JP2734639B2 - Data reading method and device - Google Patents

Description

Description: TECHNICAL FIELD [0001] The present invention relates to a data reading method and apparatus for reading an encoded data by taking an encoded image recorded on a recording medium such as paper or a sheet.

[Background] A barcode technology has been conventionally known as an image coding technology. However, in the case of a barcode, the amount of recordable information is limited due to its structure, and there has been a problem that the recording density cannot be increased. .

Recently, the applicant has arranged a large number of meshes vertically and horizontally instead of barcodes, and encoded data by using light and dark selectively formed in each mesh, or used a mesh pattern as the data body of an encoded image. Technology that captures and decodes coded images (Japanese Patent Application No. 63-328028).
issue). In the case of this method, since each data bit is expressed by the brightness of each of the meshes arranged two-dimensionally at minute intervals on the recording medium, the recording density can be greatly improved.

The purpose of decoding data of this type of coded image is to know the brightness of each mesh in the mesh pattern, but the approach of searching for the mesh directly from the image data captured from the image sensor as an approach is generally disadvantageous. Yes, false recognition is likely to occur. In particular, when a line image scanner that scans an encoded image on a recording medium manually is used as a simple image sensor,
Due to fluctuations in scanning speed and changes in direction during scanning,
Considerable image distortion occurs in the scanning stage, and the position of each mesh on the captured image data is greatly displaced from the position of each mesh on the recording medium, making it extremely difficult to decode. By providing a special mechanism such as a rotary encoder for measuring the scanning speed in the image sensor, this problem is reduced, but the cost type cannot be avoided.

Therefore, in the above-mentioned Japanese Patent Application No. 63-328028, the center position (the center in the vertical direction) of each mesh of the mesh pattern in the main scanning direction is specified in the encoded image on the recording medium. A main scanning reference pattern and a sub-scanning reference pattern for indicating the center position in the sub-scanning direction (the center in the horizontal direction of the mesh) are added. , The position of each scanning reference pattern is recognized, the center position of each mesh on the image data is determined from the result, and the image bits (pixels) representing the brightness of the mesh there are sampled. In particular, in the embodiment, two guide lines as a main scanning reference pattern are arranged in parallel outside the mesh pattern in which meshes are arranged vertically and horizontally at equal intervals so as to sandwich the mesh pattern, and a black mesh is used as a sub-scanning reference pattern. Synchronous marks of a pattern in which white meshes (clocks) are alternately repeated are provided along the inside of each guideline, and the vertical mesh rows of the mesh pattern coincide with the lines connecting the corresponding clocks of the two synchronous marks. I am trying to do it. For such an encoded image, the line image sensor is moved substantially along the guideline direction, and the image data is fetched line by line. In the decoding operation, the pixel values are checked from both ends of each main scanning line image of the image data toward the center, and by detecting the first change from white to black, a guideline position in the main scanning line image is obtained. I have. Each main scanning line image on the image data represents an image portion on a line obliquely crossing the mesh pattern on the recording medium, not completely in the vertical direction, but with fluctuations for each line image. However, even in this case, the point at which the distance between the two guide lines on the diagonal line (guide line interval) is appropriately equally divided is the vertical center of each mesh in the mesh pattern (and the vertical mark of the synchronization mark mesh). The encoded image is configured to be located at (center). Therefore, utilizing this property, the positions between the guide line positions of each main scanning line image are equally divided according to the image format, and the image bits at each equally dividing point are sampled to obtain image data decoded in the main scanning. An array (image dot row) on both sides of the main scan decoded image data represents an image that has shifted from the center of the synchronization mark in the vertical direction. Run length (run) of white pixels representing each mesh (clock)
And the run length of the black pixel should be observed alternately. The center of each run length is considered to be the horizontal center of the clock of the synchronization mark, and the line connecting the centers of the corresponding run lengths in the image dot rows of the synchronization marks on both sides is the horizontal line of the vertical mesh row of the mesh pattern. It is thought to pass through the center of the direction. Therefore, according to this principle, the center of each clock of the synchronization mark is detected from the image data decoded in the main scanning, and each image dot (pixel value) on a line connecting the centers of the corresponding clocks on both sides is sampled. , The brightness of each mesh in the mesh pattern is identified.

As described above, the scan reference pattern that does not cost is added to the encoded image, and each position of the scan reference pattern is detected from the captured image data, so that each mesh of the mesh pattern can be directly obtained from the distorted image data. It has become possible to find the position of each mesh without the difficult problem of searching.

However, if the encoded image itself on the recording medium has a flaw such as dirt and the scanning reference pattern is partially crushed or missing or damaged, the above-described decoding method is used. In such a case, there is a possibility that a stain or the like may be mistaken for a fragment of the scanning reference pattern. was there. For example, assuming that a white clock having a synchronization pattern as a sub-operation reference pattern is crushed black and connected to the preceding and succeeding black clocks, the above-described white clock is detected when detecting each clock of such a synchronization pattern on image data. The clock is skipped (detected as one black clock of the three black, white, and black clocks), and this error propagates, causing all other decoding errors. In this case, even if an error is detected by an error correction process using a later check code (redundant code included in the mesh pattern) or the like, the error cannot be corrected because the error far exceeds the correction capability.

[Object of the Invention] Accordingly, an object of the present invention is to provide a method for correctly determining a scanning reference pattern on image data even when a scanning reference pattern of an encoded image having a net-like pattern as a data body is partially destroyed due to dirt on a recording medium or the like. An object of the present invention is to provide a data reading method and device capable of determining each position of a pattern.

[Constitution and operation of the invention] In order to achieve the above object, the present invention provides a recording method in which an image including a reticulated pattern and a row of synchronization marks for indicating a data sampling reference in the sub-scanning direction of the reticulated pattern is recorded. Reading image data representing the image from the medium, searching the row of the synchronization marks from the read image data, detecting the position of at least one synchronization mark and the interval between adjacent synchronization marks in the search; Determine the detected position and interval as a feature parameter, predict the position of the adjacent synchronization mark from the determined feature parameter, measure the adjacent synchronization mark with the predicted position as the center, the position of the measurement result and Comparing with the predicted position, and when the comparison result substantially matches, the position of the adjacent synchronization mark is determined according to the actual measurement result. Determined, when the comparison result is substantially inconsistent, determine the position of the adjacent synchronization mark from the determined feature parameter,
The position information of all the synchronization marks is generated by repeating the processing of the prediction, the actual measurement, the comparison, and the determination for all the synchronization marks, and the brightness of each mesh of the mesh pattern is identified based on the generated position information. It is characterized by.

According to this configuration, the portion of the scan reference pattern destroyed due to dirt or the like can be detected because the predicted point obtained through the above-described prediction, actual measurement, and comparison substantially does not match the actual measured point. Since the position of the broken synchronization mark is determined from the characteristic parameters (for example, the intervals and positions of the synchronization marks obtained near the broken point), even if the scanning reference parameter is partially broken, The original synchronization mark position can be obtained with high accuracy, and the distinction of the brightness of each mesh of the mesh pattern performed based on the position information thus obtained can be made sure.

If the scanner is used in an environment where the scanning direction and scanning speed fluctuate while the encoded image is being scanned by the image sensor, if the position of the synchronization mark is determined based on the actual measurement results, the next It is preferable to update the above-mentioned feature parameter which is a standard for the synchronization mark.

The meaning of “position” when comparing the position of the actual measurement result with the predicted position is to be understood in a broad sense, and may include a parameter whose value changes in relation to destruction. For example, if the center position of one of the synchronization marks greatly changes due to dirt, the interval between adjacent synchronization clocks is shifted in accordance with the change, and the expected synchronization clock interval and the actually measured synchronization interval are substantially different. Detected as mismatch. Also, the size of the sync mark can be detected by comparison since the size of the crushed sync mark greatly deviates from the size expected as a normal sync mark.

If the line type image sensor that scans the encoded image fluctuates in the moving direction (sub-scanning direction) during scanning, the scanning image may be moved along the average sub-scanning direction of the encoded image due to the scanning reference pattern. It is preferable to prepare at least two rows of synchronization marks arranged side by side, and to arrange each row at a distance in the main scanning direction. When such two synchronization mark strings are used, one synchronization mark string and the other synchronization mark string can be associated one-to-one with synchronization marks. Where one of the sync marks is broken, the relationship between its measured position and the measured position of the adjacent sound sync mark in each sync mark row (position vector ), The position can be determined with high precision (because the interval between the synchronization marks on the row can be made sufficiently small, even if there is a change in the scanning direction during that time, in most cases, the magnitude of the change is extremely small. Is slight).

As a preferable configuration example for this purpose, (A) a net-like pattern formed by arranging data in each of a plurality of meshes arranged vertically and horizontally by coding light and shade, and provided on both upper and lower sides of the net-like pattern, The image data representing the image is read from a recording medium on which an image composed of a synchronization mark sequence in which light and dark alternately repeat light and dark alternately at intervals corresponding to the horizontal length of each mesh along the horizontal direction. , (B) searching the read image data for the synchronization mark sequence, and in this search, (i) detecting the center position and the horizontal length of the first synchronization mark in the synchronization mark sequence on each side. At the same time, the inclination indicating the positional relationship between the center position of the first synchronization mark in the upper synchronization mark row and the first center position of the lower synchronization mark row is detected, and the center position and the length in the horizontal direction are detected. Determining characteristic parameters consisting of the slope, to predict a synchronization mark center position of the next synchronization mark train on each side from the characteristic parameters determined (ii), (ii
i) With respect to the adjacent synchronization mark based on the predicted center position, the center position, the length in the horizontal direction, and the inclination are measured to obtain measured parameters. (iv) For each side, the predicted center position and the measured value are measured. (V) When the comparison result is within a predetermined range on either side, the position of the adjacent synchronization mark is determined in accordance with the measured parameters, and The feature parameters are updated, and (b) when one side matches within the predetermined range but does not match within the predetermined range on the other side, the actual measurement parameter indicates the position of the synchronization mark adjacent to one side. The position of the adjacent synchronization mark on the other side is determined according to the measured center position of the synchronization mark on one side, and the position of the adjacent synchronization mark on the other side is included in the measured center position of the synchronization mark on the one side and the characteristic parameter. Determined from the slope to determine the position of the synchronization mark next to the the predicted center position when not match within the predetermined range in (c) either side,
(Vi) By repeating the above processes (ii) to (v) for all the synchronization marks, position information of all the synchronization marks is generated, and (C) each of the mesh patterns is generated based on the generated position information. A data reading method and apparatus are provided, which distinguish between light and dark of a mesh.

In the following description of the embodiment, the synchronization mark, which is a component of the synchronization mark sequence, is called a clock or a clock mark.

Embodiment An embodiment of the present invention will be described below with reference to the drawings.

First Embodiment FIG. 1 shows the overall configuration of a data reading device 10 according to a first embodiment. The purpose of the entire apparatus is to read a net-like image 20 (see FIG. 2) recorded on a recording medium such as paper with the image sensor 11 and decode data encoded in the image.

The image sensor 11 is, for example, a manual line image sensor including a detection array of CCD elements, and scans the encoded image 20 on the recording medium by being manually moved in the sub-scanning direction (horizontal direction) of the recording medium. And generate corresponding image data. In detail, in each detection element of the detection array, input light, that is, input light according to the brightness of pixels on a recording medium facing each detection element is photoelectrically converted into an analog electric signal, and the time for one main scanning line is reduced. An analog image output of each detection element is sequentially and periodically extracted via an analog multiplexer or the like so that a line image output for all the detection elements of the detection array is obtained at every predetermined period (main scanning period). . Thereafter, the image data is binarized by a binarization circuit (for example, “1” for black pixels and “0” for white pixels), and the binarized serial image data is sent to the control circuit 12 together with a timing control signal. Transmitted. The control circuit 12 controls the image sensor 11 and performs parallel conversion of image data sent in series.

On the other hand, the CPU 13 operates according to the program stored in the ROM 14, and while the image sensor 11 is scanning an image, the parallel data is written to the image RAM 15 every time the serial image data is converted into parallel by the control circuit 12. Go out. After the end of the image scanning, the CPU 13 starts decoding the image data recorded in the image RAM 15 and records the result in the RAM 16. The decoding operation is roughly divided into a process of recognizing a data sampling reference pattern from image data, and a process of identifying the brightness of each mesh of a mesh pattern as a data body based on the result. According to the present embodiment, even when the sampling reference pattern is partially destroyed due to dirt or the like, it is possible to determine each position of the sampling reference pattern on the image data with high accuracy, thereby accurately decoding the mesh pattern. Make sure you get results. The details will be described later.

FIG. 2 is an example of an encoded image on a recording medium to be read by the data reading device 10. The illustrated encoded image 20 has a mesh pattern 22 as a data body inside. The mesh pattern 22 is composed of a large number of meshes arranged vertically and horizontally,
Each mesh is selectively dark (black) or light (white) in order to represent (encode) a bit which is a unit of data. For example, a bit "1" is represented by a black mesh, The bit "0" is represented by the mesh of "." In the example in the figure, a mesh pattern
22 has 48 × 72 bits of information. Further, the encoded image 20 has a pattern or a mark for designating a data sampling reference for main scanning and sub-scanning with respect to the net pattern 22 as the data body. In the case of FIG. 2, the main scanning sampling reference mark and the sub scanning sampling reference mark are separate, and the main scanning sampling reference extends in both directions above and below the mesh pattern 22 in the sub scanning direction. The sampling reference for the sub-scan is given by two synchronization mark trains 25 having a pattern of alternating black and white meshes along the inside of each guide line 21. Since the interval between the black and white meshes of the synchronization mark row 25 is synchronized with the interval between the meshes in the sub-scanning direction of the mesh pattern 22,
Each mesh of the synchronization mark train 25 is called a clock.

Two spaced guidelines 21 are used as the main scanning reference pattern in the moving direction of the image sensor 11,
The reason why they are provided along the 20 sub-scanning directions is to cope with a case where the direction of the image sensor 11 changes during scanning of an image. That is, the image data from the image sensor 11 is stored in the image RAM 15 in a manner identifiable for each main scanning line, but it is not clear in which direction the stored main scanning line image data indicates a line image in which direction. . However, by detecting the position of the two guide lines 21 on each main scan line image data, the detection result indicates
The position on the main scanning line image data (the center position of the main scanning data sampling) corresponding to the vertical center position (or vertical existence range) of each of the 22 meshes is a positional relationship of the original image, that is, It can be determined according to a predetermined positional relationship formed between the two guidelines 21 and the mesh pattern 22 in the encoded image on the original recording medium. For example, in the case of the image format shown in FIG. 2, the two guidelines 21 are parallel to each other and also parallel to the mesh pattern 22, and one point of the upper guideline 21 and one point of the lower guideline 21 are defined. Since the connecting straight line passes through a plurality of meshes at equal intervals in the mesh pattern 22, this should be stored in the line image data representing the image of such a straight line,
Therefore, the center position in the main scanning direction of each halftone in the line image can be obtained by finding the points of both guidelines 21 from the line image data and equally dividing the points.

On the other hand, the synchronization mark row 25, which is a sub-scanning reference pattern, is used to determine the position of each mesh in the sub-scanning direction in the captured image data. That is, the respective clocks of the upper and lower synchronization mark trains 25 are detected from the image data, and attention is paid to the corresponding upper and lower clocks. In the case of the encoded image format shown in FIG. 2, the straight line connecting the centers of the corresponding clocks passes through the center in the horizontal direction (sub-scanning direction) of each mesh in a certain vertical column of the mesh pattern 22. Therefore, by detecting the position of each clock of the pair of synchronization mark trains 25 from the image data, the center position of each mesh of the mesh pattern 22 on the image data in the sub-scanning direction or the existing range in the sub-scanning direction can be determined. Can decide.

Further, in the case of the encoded image 20 shown in FIG.
There is a data start mark 23 consisting of a checkered mesh group in the area scanned prior to scanning, and the area scanned after the scanning of the mesh pattern 22 is elongated horizontally and alternately repeats black and white along the main scanning direction. A data end mark 24 is formed. Data start mark 23, end mark 24 as in this example
By using a periodic or regular pattern, these marks can be easily detected on the image data. Also, as in this example, the data start mark
The user can easily determine the scanning direction (moving direction) of the image sensor 11 with respect to the encoded image 20 by making the pattern that makes it easy to visually distinguish the 23 and the end mark 24 from each other. Further, even if the end portion of the guideline 21 as the main scanning reference is broken by dirt or the like, the start or end of data can be detected by detecting the data start mark 23 or the end mark 24. It is also possible to predict or evaluate the collapsed position.

FIG. 3 shows an image sensor using the encoded image shown in FIG.
FIG. 11 shows an image obtained in the case where manual scanning is performed with care. According to the embodiment, if the distortion is as shown, even if the sampling reference pattern is partially lost, that is, even if the guide line 21 and the synchronization mark row 25 are slightly destroyed, it is possible to sufficiently cope with the error and obtain a correct decoding result. Can be given.

Hereinafter, the decoding process of the image data according to the embodiment will be described.

FIG. 4 is a flowchart of a process of storing image data obtained by scanning the encoded image 20 and recognizing a guideline 21 which is a main scanning reference pattern from the stored image data, and performing main scanning sampling of the image data. FIG. 5 is a flowchart of a process of recognizing a synchronization mark row 25 as a sub-scanning reference pattern from image data subjected to main-scan sampling and performing sub-scan sampling of the image data. This example is based on Guideline 21
And the recognition of the synchronization mark row 25, and as described below, even if some destruction occurs in the guideline 21 or the synchronization mark row 25, the position of the destruction can be determined with high accuracy. You can figure it out.

In FIG. 4, reference numerals 4-1 to 4-5 denote an encoded image 20 of the recording medium read by the image sensor 11 and an image RA.
This is the step of writing to M15. Note that the scan end condition (full memory) shown in 4-3 is merely an example, and any other appropriate event occurrence can be used as a signal to end the scan. Further, as shown in 4-4, a byte memory is assumed as the image RAM 15. Figure 6 shows the image RAM15
The image (line image) for one line is written in one row (l consecutive addresses of the image RAM 15) on the horizontal side of the figure.

The image decoding operation starts from 4-6 in FIG. 4
At -6, a search for a guideline set (a group of pixels characterizing the guideline that is the main scanning reference) is performed from the entire image data stored in the format as shown in FIG. As shown in FIG. 2, the guide line 21 is a continuous black line, and has a feature not found in other elements of the encoded image 20. Thus, for example, as illustrated in FIG.
A guideline set may be defined by three parallel run lengths 73, 74, 75 of white, black, and white with appropriate spacing. The initial run-length interval or width 76 needed to find the guideline set may use a fixed standard limit or measure the width of the most frequent white or black dots in the line image. Alternatively, it may be determined. The search for the guide line set defined by the appropriate three run lengths 73, 74, and 75 is performed by determining the positions of white dots, black dots, and white dots at appropriate intervals on the appropriate line image in the image RAM 15 and determining the line image. In the vertical depth direction (vertical direction in the case of FIG. 6), the image is tracked, and the number of successive white, black, and white dots (run length) is checked. The comparison can be performed by repeating the processing. If the search is unsuccessful, an error occurs because decoding is impossible. (4-7) When the search is successful, the standard values of the width of the guide line and the inclination in the scanning direction on the image data are determined from the information of the guide line set. Further, in 4-8, the margin considering the variation of the scanning direction inclination from the standard value is set to the left and right (in the case of FIG. 2, it is up and down, but in accordance with FIG. 6, it is hereinafter referred to as left and right). By adding to the position, left and right search areas (search width 81 in FIG. 8) of image data to be handled in the subsequent processing are obtained. It should be noted that the limitation of the search width 81 is a processing desired when there is another image such as characters (an image which may become noise during decoding) around the coded image 20 as shown in FIG. This is not particularly necessary when the periphery of the encoded image 20 is blank.

4-9 to 4-11 are processes for limiting the search area in the depth direction of the image data. For this purpose, the data start mark 23 and the data end mark 24 are detected, and these marks 23 and 24 are detected. Position of the left and right guideline in the obtained line image. Data start mark
23 and the end mark 24 can be detected by the line image containing a periodic black and white pattern. For example, in the case of the encoded image of FIG. 2, the start and end marks 23 and 24 are 24 black and white pairs. Since it is a repetition, it is sufficient to assume that these marks are marks if about 20 black-and-white pairs having the same period are found in consideration of the margin. Data start mark
It is convenient to search 23 from the line above the image data (first line) and to search for the end-of-data mark 24 from the line below the image data (last line). When the start mark 23 or the end mark 24 cannot be detected, it is considered that the error is caused by an erroneous operation such as scanning of only a part of the encoded image 20 and is processed as an error (4).
-10, 4-12).

From 4-13 to 4-18, search width by the above processing
The image data limited in number 81 and the search depth (from the start line 82 to the end line 83 in FIG. 8) is segmented into a plurality of partial images divided in the depth direction called a search block. Is examining whether or not the guidelines have been destroyed. In the case of FIG. 8, the image data consists of eight search blocks 84.
Are divided into The size (depth) of such a search block is determined by 4-13, and the first search block starting from the line following the start line 82 is selected. The depth of the search block can be determined in consideration of, for example, information of the guideline set obtained in 4-6, processing time, accuracy, and the like.
Its value determines the required length of the white, black, and white run lengths for the guideline set searched for in 4-14. That is, when a run length in the depth direction of white, black, and white is found that has an appropriate interval and a length of the search block depth (or more), a guideline set exists, and the position (guideline set) is determined. Is stored. If the search fails, it means that there is some defect in the guideline of the search block, so set a flag according to the failure status, such as whether the left or right guideline failed, or both failed, and later The guideline position must be interpolated (4-15, 4-16). For example, in the case of FIG. 8, the dirt 85 is attached to the right guideline portion of the fourth and fifth search blocks from the top, so that the right guideline is detected as an error in these search blocks.

When the state of the guideline is inspected for one search block, the next search block is selected (4-17), inspected, and the inspection is repeated until the last search block ending with the end line 83 (4-18).

Although not performed in the flow of FIG. 4, the size of the search block may be variably localized according to the search result. For example, if a search block having a guideline error is detected in 4-15, the depth of the search block is reduced to half and the half depth 2
Guideline search for each of the three search blocks, or 4
The loop from -13 to 4-18 may be restarted.

From 4-19 to the end of FIG. 4, for each search block, the positions of the left and right guide lines in each main scanning line image are determined, and the main scanning of each line image is performed in an equal division method based on the position information. The sampling positions (the entire trajectory of which is indicated by reference numeral 31 in FIG. 3) are obtained, and the image data bits (pixels, dots) at each sampling position are extracted from the line image, and the main scanning decode array 100 (10th position) is obtained. (See the figure). The determination of the position of the guide line (the center position of the guide line width) on each main scan line image is performed by actually measuring the guide block of the search block to which the main scan line image belongs, unless a failure flag indicating an error or destruction is set. If the failure flag is set, the interpolation is performed from the position of the guideline actually measured for the normal part of the guideline. For example, referring to the list of failure flags as shown in FIG. 4, the position of the guide line in the main scan line image in question is obtained by linear interpolation from the position of the guide line in the main scan line image at the end of the last search block. Can be obtained.

According to the flow, the result of the check 4-20 is NO when both the right and left guidelines of the current search block are normal (as can be seen from the corresponding failure flag being lowered). The center position of the left and right guide lines is actually measured for each main scanning line image in the search block, and the position of an equal point according to the format of the encoded image is obtained between the two positions as the main scanning sampling point. Extract the image bit at. If the failure flag is set in at least one of the left and right guidelines in 4-20, the process proceeds to 4-21 and below, and one line above the current search block is used to interpolate the position of the guideline in which the failure flag is set. , That is, the guide line position in the last main scan line image of the previous search block (the guide line position of the start line obtained in 4-9 when the current search block is the first search block) is set as the interpolation start end ( 4-21), search for a search block with a normal guideline after the next search block (4-22, 4-
26), the position of the guide line in the uppermost main scanning line image of the normal search block is detected as an interpolation candidate (4-27), and each main scanning line image in the problematic search block between the start and end of the interpolation is detected. The positions of the above guidelines are determined by interpolation, and main scanning sampling is performed (4-28). It should be noted that, although not explicitly stated in the flow, if one of the guidelines is normal, the position is directly measured. Also, if the guidelines near the end line have been broken,
The guide position of the end line evaluated in 4-11 is set as the interpolation end (4-23, 4-24).

In this manner, the section of the guideline having an error is interpolated based on the position actually measured in the section of another normal guideline, and highly accurate main scanning sampling is performed. As a result, an array 100 for main scan decoding as shown in FIG. 10 is completed. Here, the rows on both sides of the main scan decoded image data 100 are one-dimensional arrays (dot rows) of image dots along the main scanning center locus 31 (FIG. 3) of the synchronization mark row 25 which is the sub-scanning reference pattern. It has become. By examining the dot rows of the left and right synchronization mark rows 25, the center position in the sub-scanning direction in each clock of the synchronization mark row 25 is determined, and based on the result, the sub-scanning data sampling of the mesh pattern 22 is performed. The flow in FIG. 5 identifies the brightness of the mesh.

In FIG. 5, at 5-1 the main scanning width, that is, the interval of the left and right guide lines 21 (obtained in FIG. 4), the clock of the standard value (comparison reference value) is checked at 5-3 of the start clock. The length in the depth direction (the sub-scanning direction, the vertical direction in FIG. 10) is determined. However, in this embodiment, it is assumed that the image sensor 11 is manually scanned with respect to the encoded image 20. Therefore, a considerable amount of fluctuation is expected in the clock length, so that a considerably large margin is added to the standard value. There is a need. Instead of calculating the standard value from the main scanning width, for example, by examining the dot row of the synchronization mark row 25 on the main scanning decode array 100, an average white and black run length is obtained, and the obtained value is calculated as the standard value. You may make it. 5-2 Main scan decoded image data 100
, The first (black) clock of the left and right synchronization mark trains 25 on both sides is detected, and the center point, length, and main scanning direction of the image data 100 between the first clocks on the left and right (first
The inclination with respect to the horizontal direction in FIG. And 5
At -3, it is determined whether or not the clock length of the measurement result is obtained at 5-1 and is within a standard value range. At this stage, if the value is not within the range of the standard value, a reading error occurs. When the value is within the standard value, the value of a dot on a straight line connecting the center positions of the left and right start clocks is extracted from the main scan decoded image data 100 for sub-scan sampling, and the mesh of the first column in the mesh pattern 22 is extracted. Data indicating lightness and darkness are obtained, and the characteristic parameters (center position, length, inclination, etc.) actually measured in 5-2 are set as standard values for the next clock.

In 5-5, the length of the previous clock is added to the center position of the previous clock, and the center position of the next clock is predicted. In the case of this embodiment, this prediction is performed on the encoded image on the recording medium.
This is due to the fact that the synchronization mark sequence of 20 (FIG. 2) has a repetitive pattern of black and white clocks of equal length, and therefore the clock length is equal to the clock interval. Next, in step 5-6, a search is performed along the dot row of the synchronization mark on the image data decoded in the main scan from this prediction point until a dot value different from the dot value of the prediction point appears up and down, and the next clock is measured. I do. For example, if the prediction point is "1" indicating a black pixel, the next clock is a "1" that continues vertically from that point. Then, the length, center, left / right inclination, and the like of the next clock are obtained as actual measurement results (5-7). In this way, when the prediction and the actual measurement are performed, the center point of the prediction and the central point of the actual measurement should fall within a certain range if there is no error in the clock. As shown in FIG. 10 and FIG. 10, the clock length and the slope should greatly change from the standard values (the length of the previous clock and the slope of the previous left and right clocks). In FIG. 9, since the dirt 63 has crushed the second black clock of the right synchronization mark row 25, the dots of this synchronization mark row 25 on the main scan decoded image data 100 of FIG. In FIG. 10, a black dot row indicating a fragment of the dirt 63 surrounded by a dotted line is formed in a row (a row of pixels on the locus 31 in FIG. 9). Therefore, the length of the next clock is longer by the amount of dirt than the length of the previous clock (white clock, the center of the actual measurement is indicated by P1) in the dot row of the right synchronization mark row 25 and observed. Is done.
In addition, the predicted center P2 of the next clock is also the measured center P3.
Will greatly deviate from On the other hand, since there is no dirt on the corresponding portion of the left synchronization mark row 25, the difference between the center P2 of the next clock predicted from the previous clock (the center is indicated by P1) and the actually measured value P3 is small, if any. It is. Therefore, the occurrence of a clock error due to contamination or the like can be detected by comparing the characteristic parameter predicted for the next clock with the actually measured characteristic parameter and examining the difference between the two. In the flow of FIG. 5, the difference between the predicted center point of the next clock and the actually measured center point of the next clock is obtained for each of the left and right synchronization marks (5-8), and it is checked whether the difference is within the allowable range. (5-9, 5-10, 5-13), it is determined whether the next clock is appropriate. 5-11
Is a process performed when the clock is not appropriate for both the left and right. In such a situation, only the state before the synchronization mark row 25 can be reliably viewed. determine. In step 5-16, a dot row on a straight line connecting the center points is sampled from the main scanning decode array 100, and is used as data indicating the brightness of each associated mesh. 5-12, 5-14, and the right and left next clocks, one of which is appropriate and the other is incorrect. In this case, the center point of the incorrect clock uses the predicted point from the previous clock. The center point of the incorrect clock is determined from the measured center point of the appropriate clock based on the slope, so that the accuracy may be improved as much as possible. For example,
In FIG. 9 and FIG. 10, the inclination of the right and left proper white clocks seen upward can be evaluated by the vertical difference 2 (dot) between the center point P1 of the left clock and the center point P1 of the right clock. Since the right side of the black clock located below each of these two white clocks is incorrect, the left side is appropriate and the center point is actually measured at P2 shown in the figure, the horizontal line passing through this point and the right synchronization mark The position two dots above the intersection with the straight line which is the dot row of row 25 (in this case, it coincides with the position P2 predicted from the previous clock center P1 on the right side) is set as the center point of the right incorrect clock. It is determined.
Then, at 5-15, only the appropriate clock length is updated as the standard value of the clock interval, and at 5-16, the dot row between the left and right clock centers is sampled. If both clocks are appropriate, the center of the actually measured clock is determined and the dot row between them is extracted. In this case, the standard parameters are all updated with the actually measured clock characteristic parameters in 5-15.

Reference numeral 5-17 denotes a decoding end determination, in which a predetermined number of data according to the format of the encoded image 20 is compared with the number of times of execution of the decoding process 5-16. When the number of times of execution of the decoding processing 5-16 is equal to the number of times determined by the format, and the clock is not at the end line in the dot row of the synchronization mark row 25 thereafter (5-1)
8) That is, when the number of clocks detected by actual measurement or interpolation from the dot row of the synchronization mark row 25 on the main scan decoded image data 100 is equal to the number determined by the format, appropriate processing is performed. And the sampling process is completed. If erroneous clock interpolation or the like is performed on the way, the end line of the array 100 is reached before the number of data in the format is obtained (NO in 5-19), or after the number of data in the format is obtained. Further, since a clock is found (NO in 5-18), it can be detected as an error.

As described above, according to the first embodiment, the reading of the encoded image 20 under considerably severe environmental conditions, for example, during the scanning that occurs to some extent because the scanning of the encoded image 20 by the image sensor 11 is performed manually. Sampling criteria to be included in the encoded image 20 (Guideline 21, Synchronization mark string) in consideration of the change in the scanning speed and direction of the image, and the fact that characters that are noise factors are recorded around the encoded image 20 In addition to devising the pattern of 25), in recognizing the sampling reference pattern for the read image data, in the case where a part of the sampling reference pattern is broken, Method enables actual measurement of the position of guideline 21 in other parts and based on the position information By interpolating the guideline position of the problem part, the position of each clock is estimated by the characteristic parameter of the nearby clock, the actual measurement from the prediction point, the actual measurement result and the prediction result , And the position of the clock error comparison result is determined again. Therefore, misleading characters, scan direction fluctuations,
Despite the relatively severe use environment such as dirt,
Each position of the sampling reference pattern can be determined from the image data with as high accuracy as possible. As a result, the brightness of each mesh of the mesh pattern 22 can be identified with high accuracy.

Second Embodiment FIG. 11 shows the overall configuration of a data reading device 110 according to a second embodiment of the present invention. In this embodiment, the scanning of the image sensor 111 on the recording medium 118 is performed mechanically. As shown in the figure, a stepping motor 117 is driven by a control circuit 112, and the image sensor 111 is moved along a path such as a predetermined rail by the stepping motor 117. In the case of this configuration, since the fluctuation of the moving speed and the direction of the image sensor 111 is much smaller than the manual operation, there is an advantage that the distortion of the read image data is significantly reduced (the motor 117 is specially controlled by a servo mechanism or the like). This is true even if control is not performed.)
Therefore, by utilizing this property, it is possible to further simplify or omit some of the components and processes described in the first embodiment. For example, even if the encoded image 120 without the data end mark is used as shown in FIG. 12, it is correctly determined how many lines below the data start mark 23 the data ends from the accurate scanning speed of the image sensor 111. Can be predicted. Even if the search block 84 is made large, the position of the destroyed guide line can be interpolated with high accuracy. Furthermore, a synchronization mark string which is a sub-scanning reference pattern
Even in the recognition processing of 25, it is possible to use a global parameter such as an average from the entire image instead of a local parameter such as a feature parameter of the previous clock as a standard value for comparison. This enables prediction and evaluation of the position of the next clock with higher accuracy.

[Modifications] The description of the embodiment is finished above, but various modifications and changes can be made within the scope of the present invention.

For example, the data start mark 23 described in the first embodiment
The data end mark 24 predicts the position of the guide line at the corresponding portion of the guide line 21 when the corresponding portion of the mark is broken (4- in FIG. 4).
9, 4-11), but the function of the marks 23, 24 in the normal case, ie, the function of indicating the start and end of data, can be achieved without these marks. For example, in FIG. 2, the data start mark 23 and the end mark
If there is no 24, the main scanning line image passing through these parts (white part) will appear as a series of white pixels between the two black fragments of the guideline 21.
Start and end of data can be detected.

In the above embodiment, two guide lines 21 are arranged outside the mesh pattern 22 as main scanning reference patterns in the encoded images 20 and 120. However, any other suitable main scanning reference patterns can be used. For example, as shown in FIG. 13, if three (or more) parallel guide lines PR are used as the main scanning reference pattern for the mesh pattern MP, a stronger effect against partial destruction of the guide line PR is obtained. The position of the guide line can be restored. For example, for a portion of one of the three guide lines PR where there is a destruction, the position of the destruction guide line portion can be accurately determined from the corresponding position of the remaining two guide lines.
Further, the main scanning reference pattern may be arranged inside the mesh pattern 22 as exemplified by the guide line PR in the middle of FIG. In the case where the scanning direction of the image sensor is stable, the main scanning reference pattern may be configured by one guide line PR along the main scanning direction of the mesh pattern MP as illustrated in FIG. Good. In this case, the scanning direction (moving direction) of the image sensor can be determined, for example, by examining a main scanning line image passing the guide line PR. Also, by examining such a main scan line image, the position of the end point of the guideline PR can be determined,
From the result and the positional relationship between the end point of the guideline PR defined as the image format and each mesh of the mesh pattern MP, determine the center position of each mesh in the main scanning direction on the read image data, and perform main scanning sampling. Can be performed. Further, the main scanning reference pattern is advantageously a continuous line of black or a color different from other colors, but may also be formed of discontinuous marks.

In the coded images 20 and 120 of the embodiment, as a sub-scanning reference pattern, two rows of synchronization mark strings in which white and black meshes (clocks) are alternately repeated on both sides of the mesh pattern 22.
Although 25 is used, any appropriate pattern that can be synchronized (positionally associated) with a row of the mesh pattern 22 in the sub-scanning direction in the sub-scanning direction can be used as the sub-scanning reference. For example, when the scanning direction of the image sensor is stable, as shown in FIG. 15, one row of synchronization marks SR extending along the sub-scanning direction of the mesh pattern 22 is used.
Is enough. Further, three or more rows of synchronization marks may be arranged at intervals in the main scanning direction so that more accurate restoration can be performed against partial clock destruction. Further, the length of the clock in the synchronization mark can be arbitrarily set.For example, as shown in FIG. 16, a synchronization mark SR in which clocks forming elongated black fragments are arranged,
The sub-scanning reference pattern may be formed by arranging lines connecting the corresponding clocks above and below the mesh pattern MP so as to pass through the mesh boundary of the mesh pattern MP. As shown in FIG. 17, in addition to the first clock mark row SR1, a second clock such that one clock mark is provided for each predetermined number of clock marks in the first clock mark row SR. A row of marks SR2 may be provided, and a sub-scanning reference for the mesh pattern 17 may be given by both of them. In the case of this configuration, the number of clocks can be easily managed in the image decoding operation, and the restoration of the partial loss of the clock mark can be more reliably performed. For example, at a position where a clock that matches between the first clock mark string SR1 and the second clock mark string SR2 is found on the read image data, the clock mark counter for the first clock mark string SR1 is counted. Initialized, and thereafter, the first clock mark sequence
Each time a clock mark on the SR is detected or determined by restoration according to the embodiment, the clock mark counter is incremented. Then, again, the clock of the first clock mark train SR1 and the second clock mark train S
When a match with the clock of R2 is detected, the clock mark counter is checked. In the case of the format in Fig. 17,
If the value is equal to 8n (n is a natural number), it is considered that the first clock mark sequence has been properly recognized, and n is 2
In this case, it can be evaluated that there is a defect in the second clock mark sequence SR2, and if it is not equal to 8n, it can be determined that there is an error in the recognition of the first clock mark and an error can be detected. As a result, the first clock mark string SR is partially broken, and the clock is calculated through the process of predicting, measuring, comparing, and re-evaluating the next clock position as described in the embodiment (see FIG. 5). Can be narrowed down to a localized area called the clock interval of the second clock mark sequence SR, and further, each clock position of the portion can be reduced to the second clock mark sequence SR. Clock mark row SR2
Can be obtained by dividing the two clocks equally. Therefore, it is possible to recognize the brightness of each mesh for the entire mesh pattern MP while recognizing that such a local error has occurred (in the case of the embodiment, the fifth embodiment).
As shown in 5-17 to 5-19 in the figure, the whole becomes an error and cannot be decoded.) Practically, since the inspection data is included in the data of the mesh pattern, there is a sufficient opportunity for the correction of the light / dark identification data where the local error as described above has occurred in the error correction processing using the detection codes. Will be given.

Further, in the embodiment, it is assumed that there is an image that is not desirable for decoding of characters or the like near or around the encoded image, but it is not necessary to do so, in which case the recognition of the sampling reference pattern becomes considerably easier. , The conditions of the reference pattern (conditions such as arrangement, geometrical shape, etc.) desired to be distinguished from images such as characters are reduced.

In the embodiment, as the sampling reference pattern,
The main scanning (guideline 21) and sub-scanning (synchronization mark array 25) are realized by separate components on the image, but use a sampling reference pattern that combines them. Can also.

For example, returning to FIG. 16, the illustrated synchronization mark array SR can be used as a reference mark for both main scanning and sub scanning. In the case of FIG. 16, the synchronization mark string SR is a net pattern MP which is the data body.
Outside. Therefore, when there is no unnecessary image other than the mesh pattern MP and the synchronization mark string SR, it is not difficult to detect the synchronization mark string SR from the read image data. For example, when focusing on each main scanning line image of the read image data, black pixels appearing near both ends of the main scanning line image indicate a possibility of a fragment of the synchronization mark sequence SR. In the case of manual operation, each main scanning line image is generally read not in a completely vertical direction but in a slightly inclined direction while fluctuating in FIG. Therefore, if a main scanning line image having a group of black pixels near both ends continuously appears for several lines (for example, two to three lines) in each of the main scanning line images, the main scanning line image of the synchronization mark string SR is displayed. Indicates a high likelihood of a clock mark. Among such black pixel groups, those that are distributed on the image data with a certain degree of periodicity are considered to be clock marks without error.
Therefore, focusing on such a group of black pixels for several lines and actually measuring the center thereof, it is the center of the clock mark. For the broken clock mark, for example, the distance between the previous clock mark and the center of the previous clock mark (the distance between the center of the previous clock mark and the center of the previous clock mark) in the same manner as the corresponding processing in FIG. ), The measured value is measured at the center, the predicted value is compared with the measured result, and the difference can be detected because the difference is large. In this case, the predicted point is determined as the center of the current clock mark, or another column is determined. The position extracted from the center of the current clock mark based on the inclination between the previous clock marks can be determined as the center of the current clock mark. In this way, when the center position of each clock mark is determined, the corresponding clock marks are selected from the upper and lower synchronization mark trains SR, the centers are connected, and the straight line is equally divided according to the image format. The brightness at the center of each of the meshes can be known by reading the image bits in (1). In the last explanation, each clock of the synchronization mark sequence SR in FIG. 16 matches the center of each of the meshes of the mesh pattern MP. Thus, it is assumed that each clock is shifted by half a mesh in the horizontal direction).

As shown in FIG. 16, when the length direction of each clock is arranged so as to coincide with the boundary between the meshes of the mesh pattern MP, for example, the i-th and (i +
1) The centers U _i , U _{i + 1} , B _i , and B _{i + 1} of the _first clock mark are selected, and the respective center positions of the target vertical mesh row can be calculated from the position information. This will be described with reference to FIGS. 18 and 19. In FIG. 18, the centers U _i , U i of the i-th and (i + 1) -th clock marks in the upper synchronization mark sequence SR are shown.
_{Connect i + 1} (the position on the read image data) and calculate its center C1. Similarly, the center of the i-th and (i + 1) -th clock marks in the lower synchronization mark sequence SR
B _i and B _{i + 1} are connected to determine the midpoint C2. FIG. 19 shows a portion including the middle points C1 and C2 in the read image data. Each row in the horizontal direction in FIG. 19 is a main scanning line image. The first square in FIG. 19 represents a memory cell of one bit (one pixel). This midpoint C1
And C2 are connected and equally divided according to the format of the image, thereby determining the center position, that is, the storage location (indicated by a cross in the figure) of each mesh in the target mesh column in the image data memory. In the case of the above-described middle point direction, since the data sampling point is obtained from the pair of two adjacent clock centers, the positions U _i , U _{i + 1} , which are the synchronization mark string SR measurement points (or interpolation points), The errors included in B _i and B _{i + 1} can be absorbed and reduced by averaging.

Note that the brightness of each mesh can be identified by data sampling only the points indicated by the crosses in FIG. 19, but if desired, the pixel values in the vicinity (for example, up, down, left and right) of the crosses can be identified. , And the brightness of the mesh may be determined from the result. Alternatively, when the pixel value of the near stop is inconsistent with the pixel value of the cross mark point obtained from the recognition result of the synchronization mark string SR, this is stored, and the information is stored in a later error correction process using an inspection code. And the error correction capability can be enhanced thereby. For example, if an error is detected when the light / dark identification data of a block is checked according to the inspection word, and if the block includes the previously found contradiction, it is estimated as the error occurrence position. So-called erasure correction (error correction when an error position is determined) becomes possible.

The relationship between the sampling reference pattern and the net-like pattern, which is the data body, can be defined on the image data based on the defined positional relationship and the information on each position of the sampling reference pattern obtained on the image data, if it can be defined. The center position of each mesh of the mesh pattern can be obtained. This will be described with reference to FIGS. 20 to 22. In this case, it is assumed that there is some variation in the scanning speed and direction of the image sensor with respect to the encoded image, and for convenience of description, two rows of synchronous marks having clock marks are used as sampling reference patterns. It is assumed that a net-like pattern is arranged between the two rows of synchronization marks.

FIG. 20 shows the positional relationship on the image data read from the image sensor, where U _i and U _{i + 1} are the center positions of the i-th and (i + 1) -th clocks in the upper synchronization mark sequence, B _i and B _{i + 1} are _i in the lower synchronization mark row
This is the center position of the (i + 1) th and (i + 1) th clocks. For example, U _i and B _i , and U _{i + 1} and B _{i + 1} may be clocks having the shortest distance in the sub-scanning direction on the read image data. In FIG. 20, the coordinates of the point U _i are the origin (0,
0), and the coordinates of the point U _{i + 1} are shown as a point (x ₁ , 0) on the X axis. This is merely a normalized representation of the position on the image memory for convenience of explanation. The direction of the Y axis is the direction of the main scanning line image.

FIG. 21 shows the positional relationship on the recording medium corresponding to FIG. T _i and T _{i + 1} are the centers of the i-th and (i + 1) -th clocks in the upper synchronization mark sequence, and A _i and A _{i + 1} are the i-th and (i + 1) -th synchronization clock sequences in the lower synchronization mark sequence. Is the center of the clock mark. The center of each clock {T _i }, {A _i }, the reference pattern, and the center position {P} of each mesh of the synchronization mark sequence on the recording medium are known to the data reading device and have as an encoded image model. ing. For example, all the data at these positions may be stored, or if there is synchronization between the positions, all the center positions can be calculated from one or several position data and the synchronization data. In FIG. 21, the line of the upper synchronization mark string connecting T _i and T _{i + 1} and the line of the lower synchronization mark string connecting A _i and A _{i + 1} are drawn in parallel, but it is not always necessary. Absent. Also in FIG. 21, the center of the i-th clock in the upper synchronization mark row is defined as the origin (0, 0), and the center of the (i + 1) -th clock is illustrated as a point (D, 0) on the x-axis.

Now, it is assumed that the respective clock centers of the synchronization mark sequence as the reference pattern have been obtained from the read image data in the manner described above, and the four clock centers U _i , U _{i + 1} , B as shown in FIG. _It is assumed that attention is focused on a square image block surrounded by _i and B _{i + 1} . The problem is to know the center position of each mesh within this square image block. The data reader then converts the encoded image model to the corresponding clock center on the model (and therefore on the recording medium).
Select T _i , T _{i + 1} , A _i , and A _{i + 1} . Since the coded image model includes the position information {P} of the center of each mesh in all the coded images, a square surrounded by T _i , T _{i + 1} , A _i , and A _{i + 1} (in this case, A set of center positions of the mesh inside the parallelogram can be created. For example, in the case of FIG. 21, a set of coordinates (x, y) satisfying 0 ≦ x + Ly ≦ D (L is a slope) may be selected from the set {P} of the center positions of all the meshes.

The internal points of the rectangle surrounded by T _i , T _{i + 1} , A _i , and A _{i + 1} on the model are represented by U _i ,
It corresponds to the interior point surrounded by U _{i + 1} , B _i , and B _{i + 1} and does not go outside. Therefore, when the center position of each mesh inside the rectangle is found in the coordinate system xy of the coded image model in the manner described above, each position is then referred to in the coordinate system XY of the read image data. What is necessary is to convert the coordinates to the position and extract the image bit at the converted position.

An example of coordinate exchange is shown in FIG. FIG. 2A shows an image block in question on the model, and the center of a certain mesh is indicated by coordinates (x, y). FIG. 11F shows the corresponding image block (target image block) on the read image data, and the coordinates corresponding to the coordinates (x, y) are indicated by (X, Y). (A) and (b) are horizontal scalings for adjusting the length of the upper side of the rectangle, and (b) to (c) are gradients of the lower side (U _i to B _i ) of the target image block. (C) to (d) are scalings for adjusting the length of the left side to the length of the left side of the target image block, and the four vertices U of the target image block have been described so far. _{Of the i} , U _{i + 1} , B _i , and B _{i + 1} , three points U _i , U _{i + 1} , and B _i are the same image blocks. Therefore, as shown in (e), if the corresponding point of this image block (d) is shifted to a point B _{i + 1} which is a diagonal of the point U _i, a target image block is obtained. Image blocks are local areas,
Even in the case of manual scanning, since the speed fluctuation and the direction change in such a local area are small, the conversion from (d) to (e) is performed with uniform distortion (proportional to the area ratio). distortion)
There is no problem to think.

As a result, the desired coordinates (X, Y) are here, (The multipliers for △ D and △ H in equations (1) and (2) are represented by the solid rectangle and the coordinates (X
d, Yd) indicates the area ratio with the dotted rectangle having the diagonal points, ΔD is the difference between the length of the upper side and the lower side of the target image block, and ΔH is the right and left of the target image block. The difference between the lengths of the sides in the Y-axis direction, and other X ₁ , X ₂ ,
X ₃ , Y ₂ and Y ₃ are shown in FIG. 20). In this way, the coordinates (X, Y) of the center of the corresponding mesh in the target image block are obtained from the coordinates (x, y) of the center of the mesh in the model. Accordingly, by sampling the image bit at the position (X, Y), information indicating the brightness of the mesh can be obtained. Note that points T _i , T _{i + 1} ,
Since the frame of the image formed by A _i and A _{i + 1} is actually a very elongated local area (see FIG. 2), even if the manual scanning is performed, the deviation ΔH is ignored in the normal case. (△ H = 0), and therefore can be simplified to Y = Yd instead of equation (2).

[Effects of the Invention] As is apparent from the above description, according to the data reading method and the scanning according to the first and second aspects, in the row of the synchronization marks indicating the data sampling reference in the sub-scanning direction of the mesh pattern. If there is a partially destroyed part of
This is detected from the mismatch between the predicted position and the actual measurement position, and the position of the destroyed sync mark is determined from the determined feature parameters, so the sync mark row is partially destroyed by dirt or scratches. In this case, the position of the synchronization mark can be accurately determined without skipping the synchronization mark, and there is an advantage that the distinction of the brightness of each mesh of the mesh pattern performed based on the position information can be surely performed.

The data reading method and apparatus according to the third and fourth aspects are particularly suitable for a use environment in which a line-type image sensor for scanning an encoded image is fluctuated in a moving direction (sub-scanning direction) during scanning. ing. In this case, if the synchronization mark on one side is not destroyed, the position of the synchronization mark is used as the measured position and the characteristic parameter of the synchronization mark on one side even if the corresponding synchronization mark on the other side is destroyed. It can be determined with high accuracy from the included tilt information.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram of a data reading apparatus according to a first embodiment of the present invention, FIG. 2 is a diagram illustrating an encoded image on a recording medium to be read, and FIG. FIG. 4 is a diagram showing an example of an image obtained by scanning an encoded image with a manual image sensor. FIG. 4 stores image data obtained by scanning an encoded image, and recognizes a guideline which is a main scanning reference pattern from the stored image data. FIG. 5 is a flowchart for performing main scanning decoding, FIG. 5 is a flowchart for performing sub scanning decoding by recognizing a synchronization mark string which is a sub scanning reference pattern from the main scanning decoded image data, and FIG. FIG. 7 is a diagram showing a memory map of a RAM, FIG. 7 is an explanatory diagram of a guideline set characterizing fragments of the guideline, and FIG. FIG. 9 is a diagram showing a plurality of search blocks obtained by dividing an encoded image, FIG. 9 is a diagram showing an example of an image of a guide line with a stain and a synchronization mark string, and FIG. 10 is a diagram corresponding to the image shown in FIG. FIG. 11 is a diagram showing scan-decoded image data, FIG. 11 is an overall configuration diagram of a data reading device according to a second embodiment, FIG. 12 is a diagram showing an example of an encoded image without a data end mark, and FIG. FIG. 14 is a schematic diagram of an example of an encoded image in the case of three guidelines, FIG. 14 is a schematic diagram of an example of an encoded image in the case of one guide line, and FIG. 15 is an encoded image of one synchronous mark sequence. Schematic diagram of the example, FIG. 16 is a schematic diagram of an example of an encoded image with a synchronization mark sequence in which elongated black marks are arranged as clocks, and FIG. 17 is an encoded image with two types of synchronization mark sequences having different periods. Schematic illustration of the example, Figure 18 shows the four FIG. 19 is a diagram used to describe the process of identifying the lightness and darkness of each mesh from the center of the lock using the middle point method. FIG. 20 is a diagram showing the relationship between the four positions of the sampling reference pattern on the read image data and the positions of the meshes inside, and FIG. 21 corresponds to FIG. 20 but shows sampling on the recording medium. FIG. 22 is a diagram showing a relationship between four positions of a reference pattern and positions of meshes inside the image. FIG. 22 is a view for explaining a process of converting the center coordinates of the mesh on the recording medium into the center position of the mesh on the image data. FIG. 20 coded image, 21 guideline (main scanning reference pattern), 22 mesh pattern, 25 synchronization mark array (sub-scanning reference pattern), 11, 111 image sensor, 13 CPU 14 ... ROM, 15 ... Image RAM.

Claims (4)

(57) [Claims] 1. A data reading method for reading data encoded on an image recorded on a recording medium, comprising the steps of: (A) forming a mesh in which data is encoded by light and dark formed on a plurality of meshes arranged vertically and horizontally; Patterns and
Reading image data representing the image from a recording medium on which an image comprising a sequence of synchronization marks for designating a data sampling reference in the sub-scanning direction of the mesh pattern is recorded; (I) detecting a position of at least one synchronization mark and an interval between adjacent synchronization marks, and detecting the detected position; (Ii) predicting the position of the adjacent synchronization mark from the determined characteristic parameter; (iii) actually measuring the adjacent synchronization mark around the predicted position; (Iv) comparing the position of the actual measurement result with the predicted position; (v) actual measurement when the comparison result substantially coincides Determining the position of the adjacent synchronization mark according to the result, and determining the position of the adjacent synchronization mark from the determined characteristic parameter when the comparison result substantially does not match; and (vi) from (ii) to (ii). generating position information of all the synchronization marks by repeating step v) for all the synchronization marks; and (C) identifying the brightness of each mesh of the mesh pattern based on the generated position information. A data reading method, comprising: 2. A data reading apparatus for reading data encoded on an image recorded on a recording medium, comprising: (A) a mesh which encodes data by light and dark formed on a plurality of meshes arranged vertically and horizontally; Patterns and
(B) image sensor means for reading image data representing an image from a recording medium on which an image comprising a row of synchronization marks for indicating a data sampling reference in the sub-scanning direction of the mesh pattern is recorded; Searching means for searching for a row of the synchronization mark from the image data obtained, the search means comprising: (i) detecting a position of at least one synchronization mark and an interval between adjacent synchronization marks; Means for determining the detected position and interval as a feature parameter; (ii) means for estimating the position of an adjacent synchronization mark from the determined characteristic parameter; and (iii) the adjacent synchronization mark with the estimated position as the center. (Iv) means for comparing the position of the actual measurement result with the predicted position, (v) when the comparison results substantially coincide with each other Means for determining the position of the adjacent synchronization mark according to the actual measurement result, and determining the position of the adjacent synchronization mark from the determined characteristic parameter when the comparison result is substantially inconsistent; and (vi) from (ii) above Means for generating position information of all synchronization marks by repeating the means of (v) for all synchronization marks; and (C) identification for identifying the brightness of each mesh of the mesh pattern based on the generated position information. Means for reading data. 3. A data reading method for reading data encoded on an image recorded on a recording medium, comprising: (A) a mesh which encodes data by light and dark formed on each of a plurality of meshes arranged vertically and horizontally. Patterns and
An image is provided which is provided on both upper and lower sides of the mesh pattern, and which is composed of a synchronization mark string which alternates between light and dark alternately along the horizontal direction of the mesh pattern at intervals corresponding to the horizontal length of each mesh. Reading the image data representing the image from the read recording medium; (B) searching for the synchronization mark string from the read image data, comprising the following steps: The center position of the first sync mark of the side sync mark row and the length in the horizontal direction are detected, and the center position of the first sync mark of the upper sync mark row and the first center position of the lower sync mark row are detected. Detecting a tilt representing the positional relationship with the center and determining a characteristic parameter comprising the center position, the length in the horizontal direction, and the tilt; (ii) a synchronization mark sequence on each side from the determined characteristic parameter. Estimating the center position of the adjacent synchronization mark; (iii) measuring the center position, the length in the horizontal direction, and the inclination of the adjacent synchronization mark based on the predicted center position to obtain an actually measured parameter; (Iv) comparing the predicted center position with the actually measured center position for each side; (v) the comparison result is (a) when the comparison result is within a predetermined range on either side, the actually measured parameter The position of the adjacent synchronization mark is determined according to the above, and the feature parameter is updated with the actually measured parameter. (B) If the one side matches within the predetermined range but the other side does not match within the predetermined range, The position of the adjacent synchronization mark on the other side is determined according to the measured center position of the synchronization mark on the one side indicated by the measured parameter, and the position of the adjacent synchronization mark on the other side is determined. The, determined from the slope contained in the measured center position and the feature parameter of the sync mark on one side of said,
(C) determining the position of the adjacent synchronization mark based on the predicted center position when no match is found within the predetermined range on either side; and (vi) performing all of the steps (ii) to (v) above. (C) identifying the brightness of each mesh of the mesh pattern based on the generated location information. Data reading method. 4. A data reading device for reading data encoded on an image recorded on a recording medium, comprising: (A) a net-like data in which data is encoded by light and dark formed on a plurality of meshes arranged vertically and horizontally; Patterns and
An image is provided which is provided on both upper and lower sides of the mesh pattern, and which is composed of a synchronization mark string which alternates between light and dark alternately along the horizontal direction of the mesh pattern at intervals corresponding to the horizontal length of each mesh. Means for reading image data representing the image from the read recording medium; (B) search means for searching for the synchronization mark string from the read image data, the search means comprising: ) The center position and the horizontal length of the first synchronization mark of the synchronization mark sequence on each side are detected, and the center position of the first synchronization mark of the upper synchronization mark sequence and the first position of the lower synchronization mark sequence are detected. Means for detecting a tilt representing a positional relationship with the center position, and determining a characteristic parameter comprising the center position, a lateral length, and a tilt; (ii) a synchronization mark sequence on each side from the determined characteristic parameter Means for predicting the center position of the adjacent synchronization mark; (iii) means for measuring the center position, the length in the horizontal direction, and the inclination of the adjacent synchronization mark based on the predicted center position to obtain an actually measured parameter; (Iv) means for comparing the predicted center position with the actually measured center position for each side; (v) the comparison result is (a) when the comparison result matches within a predetermined range on either side, the actually measured parameter The position of the adjacent synchronization mark is determined according to the above, and the feature parameter is updated with the actually measured parameter. (B) If the one side matches within the predetermined range but the other side does not match within the predetermined range, The position of the adjacent synchronization mark on the other side is determined in accordance with the measured center position of the one side synchronization mark indicated by the measured parameter, and the position of the adjacent synchronization mark on the other side is determined by the above-mentioned one. Actually measuring the center position of the synchronization mark on the side of the determination from the slope included in the feature parameter,
(C) means for determining the position of the adjacent synchronization mark on the basis of the predicted center position when no match is found within the predetermined range on any side; and (vi) all means from (ii) to (v) above (C) identification means for identifying the brightness of each mesh of the mesh pattern on the basis of the generated location information. Characteristic data reading device.