CN106056020A - Bar code detection and recognition method and system based on machine vision system - Google Patents

Abstract

The invention discloses a barcode detection and recognition method and system based on a machine vision system. The method is applied to a machine vision system host, and the machine vision system host is connected with a light source module and an image acquisition module. The method includes the following steps : S1, image data acquisition step, S2, image data preprocessing step, S3, barcode area positioning step, S3A, barcode area adjustment step, S4, barcode recognition step. The invention can automatically scan products with barcodes and identify the barcodes without manual operation, and has the advantages of high degree of automation, fast identification speed, and high work efficiency.

Description

A barcode detection and recognition method and system based on machine vision system

technical field

The invention relates to the technical field of barcode detection by machine vision systems, in particular to a method and system for barcode detection and recognition based on machine vision systems.

Background technique

With the continuous deepening of the concept of "Industry 4.0" and the wide application of machine vision in industrial automation, some manufacturing enterprises have begun to enter the stage of automated production. In the process of production control management, barcodes and their identification and reading technologies have become the main products Means of identification and tracking. Traditional scanning guns need to be manually aligned with the barcode area at close range, and their disadvantages such as low efficiency, slow speed, low degree of automation, and high labor costs have been unable to meet the needs of large-scale automated industrial production. At present, domestic and foreign scholars have conducted research on the regional positioning of barcode labels and proposed corresponding solutions, mainly including methods based on DCT, methods based on differences, and methods based on mathematical morphology. The barcode area of the DCT algorithm is automatically positioned. This algorithm can locate the barcode at any rotation angle, but when the texture ratio of other areas in the image is greater than the barcode area, the algorithm will fail. The barcode is positioned based on the difference method. This method achieves comparison Simple, good for vertical and horizontal barcode positioning, but poor robustness for complex backgrounds; using mathematical morphology expansion operation to define the barcode area, it is easy to connect the barcode area with other areas, and it is easy to appear False detections and missed detections. Therefore, it is necessary to provide a new barcode detection method to overcome the shortcomings of existing barcode detection and recognition methods.

Contents of the invention

In order to overcome the deficiencies in the prior art, the present invention provides a barcode detection and recognition method and system based on a machine vision system. Using the present invention, products with barcodes can be automatically scanned and the barcodes can be recognized without manual operation. , which has the advantages of high degree of automation, fast recognition speed, and high work efficiency.

The present invention is achieved through the following technical solutions:

A method for detecting and recognizing barcodes based on a machine vision system, the method is applied in a machine vision system host, the machine vision system host is connected with a light source module and an image acquisition module, and the method comprises the following steps:

S1. The image data acquisition step, through the cooperation of the image acquisition module and the light source module, image data acquisition is performed on labels with multiple barcodes, and the acquired image data is transmitted to the host of the device vision system;

S2. Image data preprocessing step, the machine vision system host performs preprocessing on the image data, so as to separate the acquired image data from the label and the background, thereby filtering background interference;

S3, the barcode area positioning step, the machine vision system host performs barcode feature extraction on the image data processed by the image data preprocessing step, and performs area positioning on the barcode to determine the location of all the barcodes on the label. The area is the barcode area with all the barcodes on the label;

Wherein, the barcode area includes the upper and lower areas of the barcode and the left and right areas of the barcode, and the positioning of the barcode area includes the positioning of the upper and lower areas of the barcode and the positioning of the left and right areas of the barcode. , the number of barcodes existing in the same row of barcodes, so as to realize the regional positioning of all barcodes on the label;

S4. A barcode recognition step, performing barcode recognition on each barcode in the barcode area.

Further, after the barcode area locating step S3, the following steps are also included before the barcode recognition step S4:

S3A, the barcode area adjustment step, the left and right boundaries of the barcode area are respectively adjusted by expanding K pixels, and the upper and lower boundaries of the barcode area are respectively adjusted by reducing P pixels, and finally the adjusted The barcode area is used as the final barcode area, wherein the K and P are positive integers, and the specific values are selected according to requirements.

Preferably, in the barcode region locating step S3, in the process of locating the upper and lower regions of the barcode, a normalized cross-correlation matching algorithm and a Hough transform line detection algorithm are used to determine the upper and lower regions of the barcode; the barcode region locating In step S3, in the process of locating the left and right regions of the barcode, a linear gray value scan is performed on the center line l of each barcode on the grayscale image of the preprocessed image to obtain the center line l of each barcode. Gray value distribution diagram, the pixel data on the gray value distribution diagram is cumulatively calculated, the central line 1 gray value array is LineA, a new array is generated after the cumulative operation is LineB, according to the array LineA and data LineB A center line gray value accumulation distribution diagram of the barcode is established, and the left and right area positioning of the barcode is determined through the straight lines in the horizontal part of the centerline gray value accumulation distribution diagram.

Preferably, in the image data preprocessing step S2, the machine vision system host performs preprocessing on the image data, including filtering, denoising, enhancement and/or binarization processing; preferably, the The image acquisition module is an industrial camera.

Based on the inventive concept of the above method, the present invention also provides a barcode detection and recognition system based on a machine vision system. The system includes a machine vision system host, and the machine vision system host is connected with a light source module and an image acquisition module. The image acquisition module is used to cooperate with the light source module to collect image data on labels with multiple barcodes, and transmit the collected image data to the machine vision system host, and the machine vision system host includes:

An image data preprocessing module, configured to preprocess the image data, so as to separate the acquired image data from the label and the background, thereby filtering background interference;

The barcode area positioning module is used to perform barcode feature extraction on the image data preprocessed by the image data preprocessing module, and perform area positioning on the barcode to determine the area where all the barcodes on the label are located. The area is a barcode area with all barcodes on said label;

Wherein, the barcode area includes the upper and lower areas of the barcode and the left and right areas of the barcode, and the positioning of the barcode area includes the positioning of the upper and lower areas of the barcode and the positioning of the left and right areas of the barcode. , the number of barcodes existing in the same row of barcodes, so as to realize the regional positioning of all barcodes on the label;

The barcode recognition module is used to perform barcode recognition on each barcode in the barcode area.

Further, the machine vision system host also includes:

The barcode area adjustment module is used to adjust the left and right boundaries of the barcode area by expanding K pixels respectively, and adjust the upper and lower boundaries of the barcode area by shrinking P pixels, and finally adjust the adjusted The barcode area is used as the final barcode area, wherein the K and P are positive integers, and the specific values are selected according to requirements.

Preferably, the barcode area locating module, in the process of locating the upper and lower areas of the barcode, uses a normalized cross-correlation matching algorithm and a Hough transform line detection algorithm to determine the upper and lower areas of the barcode; the barcode area locating module , in the process of positioning the left and right regions of the barcode, perform linear grayscale value scanning on the centerline l of each barcode on the grayscale image of the preprocessed image, so as to obtain the grayscale value of the centerline l of each barcode Distribution diagram, the pixel data on the gray value distribution diagram is accumulated, set the central line 1 gray value array as LineA, after the accumulation operation, a new array is generated as LineB, and the barcode is established according to the array LineA and data LineB The central line gray value accumulation distribution diagram, the positioning of the left and right regions of the barcode is determined by the straight lines in the horizontal part of the central line gray value accumulation distribution diagram.

Preferably, the image data preprocessing module performs preprocessing on the image data, including filtering, denoising, enhancement and/or binarization; the image acquisition module is an industrial camera.

The barcode detection and recognition method and system based on the machine vision system provided by the present invention can effectively replace the traditional barcode reader, even for oblique barcodes, single-label multiple barcodes, single-line multiple barcodes, partially missing barcodes, stained etc. can get a satisfactory recognition effect, and can realize automatic recognition of barcodes in any direction and multiple barcodes, and has the advantages of fast recognition speed and high recognition efficiency, so that it can effectively meet the needs of large-scale automated industrial production of enterprises.

Description of drawings

Accompanying drawing 1 is the schematic flow chart of the method described in the embodiment of the present invention;

Accompanying drawing 2 is the structural module block diagram of the system described in the embodiment of the present invention;

Accompanying drawing 3 is the straight line that utilizes Hough transform algorithm to carry out straight line detection to the bar code on the label and obtains, and carries out the schematic diagram of coordinate marking of the two ends of each straight line;

Accompanying drawing 4 is the distribution diagram of the gray value with the centerline of a barcode;

Accompanying drawing 5 is the gray value distribution figure that has the center line of two barcodes in the same row;

Accompanying drawing 6 is the accumulative distribution diagram of the gray value corresponding to the barcode center line in the accompanying drawing 4;

Accompanying drawing 7 is the accumulative distribution diagram of the gray value corresponding to the center line of the barcode in the accompanying drawing 5.

detailed description

In order to facilitate the understanding of those skilled in the art, the present invention will be further described below in conjunction with the accompanying drawings.

As shown in accompanying drawing 1, a kind of barcode detection and recognition method based on machine vision system, described method is applied in machine vision system host computer, described machine vision system host computer is connected with light source module and image acquisition module, in the present embodiment, The image acquisition module is preferably an industrial camera, and the method comprises the following steps:

S1. The image data acquisition step, through the cooperation of the image acquisition module and the light source module, image data acquisition is performed on labels with multiple barcodes, and the acquired image data is transmitted to the host of the device vision system;

S2, the image data preprocessing step, the machine vision system host performs preprocessing on the image data, so as to separate the acquired image data from the label and the background, thereby filtering the interference of the background; a method based on intra-class variance can be used The adaptive threshold segmentation method performs preprocessing on the image data, and the preprocessing includes filtering, denoising, enhancing and/or binarization processing;

S3, the barcode area positioning step, the machine vision system host performs barcode feature extraction on the image data processed by the image data preprocessing step, and performs area positioning on the barcode to determine the location of all the barcodes on the label. The area is the barcode area with all the barcodes on the label; because there are inclined barcodes, after the barcode feature extraction, if there is an image of the barcode inclined, the image of the inclined barcode is rotated and corrected before Regional positioning of the barcode;

Wherein, the barcode area includes the upper and lower areas of the barcode and the left and right areas of the barcode, and the positioning of the barcode area includes the positioning of the upper and lower areas of the barcode and the positioning of the left and right areas of the barcode. , the number of barcodes existing in the same row of barcodes, so as to realize the regional positioning of all barcodes on the label;

S4. A barcode recognition step, performing barcode recognition on each barcode in the barcode area.

As a preferred embodiment, after the barcode area positioning step S3, the following steps are also included before the barcode recognition step S4:

S3A, the barcode area adjustment step, the left and right boundaries of the barcode area are respectively adjusted by expanding K pixels to prevent left and right missing codes, and the upper and lower boundaries of the barcode area are respectively reduced by P pixels Adjust the processing to ensure the stability of barcode recognition. Finally, the adjusted barcode area is used as the final barcode area. By using the adjusted barcode as the final barcode area and performing barcode identification on the final barcode area, you can Effectively prevent the occurrence of missing codes and effectively improve the stability of barcode recognition. Wherein, the K and P are positive integers, and the specific values are selected according to the needs. In this embodiment, the K=P=15,

As a preferred embodiment, in the barcode area positioning step S3, the barcode feature extraction is first performed, and the image is tilt-corrected, and then the barcode area positioning is carried out. The bar code area positioning includes bar code upper and lower area positioning and bar code left and right area positioning; In the barcode upper and lower region positioning process, the normalized cross-correlation matching algorithm (NCC) and the Hough transform line detection algorithm are used to determine the upper and lower regions of the barcode; in the barcode region positioning step S3, the left and right regions of the barcode are In the positioning process, the gray value of the center line l of each barcode is scanned in a straight line on the gray image of the preprocessed image to obtain the gray value distribution map of the center line l of each bar code. The pixel data on the figure is accumulated, and the central line 1 gray value array is set as LineA, and a new array is generated after the accumulated operation is LineB, and the center line gray value accumulation distribution diagram of the barcode is established according to the array LineA and the data LineB The location of the left and right regions of the barcode is determined by the straight line in the horizontal part in the gray value accumulation distribution diagram of the center line.

The barcode area positioning step S3 in the embodiment of the present invention will be further described in detail below:

A barcode is a graphic identifier that arranges multiple black bars and blanks with different widths according to certain coding rules to express a set of information. If the barcode is regarded as a set of regularly arranged straight lines, then it is only necessary to detect all the straight lines on the label. In the embodiment of the present invention, the linear detection method is preferably used to locate the area of the barcode. Using the method of linear detection to locate the area of the barcode can effectively deal with the situation where the quality and appearance of the barcode are significantly degraded, even if the angle, size and shadow of the barcode change. In the case of a barcode defect, the position of the barcode can still be effectively located, so that the location of the barcode area can be realized; moreover, by using the straight line detection method, the position of the barcode can be determined by reading only part of the black bar of a barcode, without the need for precision The barcode position can be roughly located by reading all the black bars.

In the embodiment of the present invention, according to the characteristics of the bar-shaped arrangement of the barcode, the Hough transform algorithm is used to detect the barcode on the label as a straight line. The main principle is to use the polar coordinate formula of the straight line: ρ=хcos(θ)+уsin(θ ); ρ, θ are a pair of parameters (please give a supplementary explanation of the specific parameter definitions of ρ, θ, thank you), take a fixed point O in the plane, called the pole, lead a ray Ox, called the polar axis, and then select a The positive direction of length units and angles (usually counterclockwise). For any point M in the plane, use ρ to represent the length of line segment OM, θ to represent the angle from Ox to OM, ρ is called the polar diameter of point M, θ is called the polar angle of point M, and the ordered pair (ρ, θ) is called The polar coordinates of point M, the coordinate system established in this way is called polar coordinate system. By applying this formula to the pixels in the image, the two-dimensional ρ, θ parameter space can be obtained, and a peak will be formed in the parameter space, so that the straight line can be detected by using the peak. The two endpoints of any detected straight line are recorded as: p _i =( _xi ,y _i ),q _i =(m _i ,n _i )(i=1,2...k), as shown in Figure 3 Show. By calculating the position coordinates of the two endpoints of each straight line, the length and slope of all detectable black bars of the barcode are obtained. The length and inclination angle of any straight line are:

${\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\sqrt{{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; arctan</span>}\frac{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; .</span>$

If x _i -m _i ＝0, then the straight line is in the vertical direction, no counterclockwise rotation is required; if x _i -m _i ≠0, based on the fact that the rotation angles of the barcodes on the same label are consistent, if the If the straight line belongs to the barcode area, the inclination angle of the straight line should be very close. Due to the influence of the binarization processing in the aforementioned preprocessing, there will be a certain error between the inclination angle of the straight line calculated by formula (2) and the real angle. Therefore, the median value of the inclination angles of all detected straight lines is selected as a reference value, and as long as θ _i is within the range of plus or minus 1 degree of the median value, the inclination angle is retained. Calculate the average value of all retained tilt angles as the tilt correction degree of the label, based on which the original label image is corrected counterclockwise, and these barcodes are reserved as the rough positioning area of the barcode.

When performing area positioning on all barcodes in a multi-barcode label, due to the influence of actual noise, if only the line detection method is used to determine the upper and lower areas of the barcode of the straight line, the error will be relatively large. Through the analysis, the blanks in the non-barcode area are relatively obvious. According to the position coordinates of most of the barcode line endpoints obtained from the above detection, straight lines with equivalent lengths can be reserved, and the average value of all reserved straight lines can be recorded as At the same time get the midpoint position of each straight line Obtain the centerline l of each barcode, and rotate it at an oblique angle to the up and down of the centerline l As a coarse positioning area for barcodes.

In order to accurately locate the upper and lower areas of the barcode, a normalized cross-correlation (NCC) algorithm is used in the rough positioning area to perform precise positioning. The NCC algorithm measures the degree of matching by comparing the correlation of two columns of data, and the position with the largest correlation coefficient is the best matching position. Assuming that f( _xi ), g( _xi ) are any two lines of the barcode, their corresponding mean and variance are recorded as and Randomly select two lines in the barcode area to translate up and down respectively, then the normalized cross-correlation coefficient is defined as:

$\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\frac{\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; f</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; g</span>}}<span\; class="notranslate">\; )</span>}{{\mathrm{<span\; class="notranslate">\; \&delta;</span>}}_{\mathrm{<span\; class="notranslate">\; f</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; \&delta;</span>}}_{\mathrm{<span\; class="notranslate">\; g</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

Set the threshold T, then when R>T, otherwise it is the upper and lower areas of the barcode. Because the NCC algorithm has high accuracy and adaptability, but the amount of calculation is relatively large, the Hough transform line detection algorithm is used to extract the region of interest from the original image, and the upper and lower boundaries of each barcode area are roughly positioned. The barcode area is effectively eliminated, reducing the amount of calculation. Through the above-mentioned NCC algorithm and the Hough transform line detection algorithm, the edge line of the area where the barcode is located can be accurately determined.

Barcode left and right area positioning, including the detection of the number of barcodes in the same line and the barcode left and right area positioning of each barcode, so as to realize the area positioning of all barcodes in multi-barcode labels. Carry out linear gray value scanning on the center line l on the gray scale image, and the gray value distribution map of the center line l of each barcode can be obtained. Because in the process of image binarization, the left and right borders of the barcode are easily damaged due to quality, noise and other reasons. If you scan directly on the binary image, this defect will directly lead to inaccurate positioning of the left and right areas of the barcode, so use The grayscale image is used to locate the left and right areas of the barcode area.

According to the characteristics of the barcode, there will be a blank area at the left and right borders of the barcode, as shown in Figure 4 (the gray value distribution diagram with a barcode center line in Figure 4) where the gray value is 0; If multiple barcodes appear in a row, the blank area between the barcodes will be larger, as shown in Figure 5 (the gray value distribution diagram of the center line with two barcodes in the same row in Figure 5). In order to further determine the left and right edges of the barcode area, the pixel data on the distribution map is accumulated, and the gray value array of the center line is LineA. After the accumulation operation, a new array is generated as LineB, namely:

LineA＝{x ₁ ,x ₂ ,x ₃ ... x _n },

Among them, LineB[0]=LineA[0].

Fig. 6 is the accumulative distribution diagram of the barcode centerline gray value in Fig. 4, and Fig. 7 is the accumulative distribution diagram of the barcode centerline gray value in Fig. 6; wherein, the horizontal part (that is, the slope is 0) in Fig. 6 and Fig. 7 That is, the boundary between the barcode areas, according to which the positioning of the left and right areas of the barcode can be realized.

The remaining steps S1, S2, S3A, S4 and other steps can be realized by using existing technologies, and will not be described in detail here.

Using the barcode detection and recognition method based on the machine vision system provided in this embodiment, in the process of barcode recognition for labels, even if it is an oblique barcode, multiple barcodes on a single label, or multiple barcodes on a single line, partially missing barcodes, stained Barcodes can achieve satisfactory recognition results, and at the same time, automatic recognition of barcodes in any direction and multiple barcodes can be realized, and the recognition speed is fast and the recognition efficiency is high, so that it can effectively meet the needs of large-scale automated industrial production of enterprises.

Based on the inventive concept of the method described in the above embodiments, the embodiment of the present invention also provides a barcode detection and recognition system based on a machine vision system, as shown in Figure 2, the system includes a machine vision system host, the The machine vision system host is connected with a light source module and an image acquisition module. The image acquisition module is used to cooperate with the light source module to collect image data on labels with multiple barcodes, and transmit the collected image data to the machine vision In the system host, the machine vision system host includes:

The image data preprocessing module is used to preprocess the image data, so as to separate the acquired image data from the label and the background, so as to filter the interference of the background; the preprocessing preferably includes filtering, denoising, enhancing and/or binarization;

The barcode area positioning module is used to perform barcode feature extraction on the image data preprocessed by the image data preprocessing module, and perform area positioning on the barcode to determine the area where all the barcodes on the label are located. The area is a barcode area with all barcodes on said label;

Wherein, the barcode area includes the upper and lower areas of the barcode and the left and right areas of the barcode, and the positioning of the barcode area includes the positioning of the upper and lower areas of the barcode and the positioning of the left and right areas of the barcode. , the number of barcodes existing in the same row of barcodes, so as to realize the regional positioning of all barcodes on the label;

The barcode recognition module is used to perform barcode recognition on each barcode in the barcode area.

As a preferred embodiment, the machine vision system host also includes:

The barcode area adjustment module is used to adjust the left and right boundaries of the barcode area by expanding K pixels respectively, and adjust the upper and lower boundaries of the barcode area by shrinking P pixels, and finally adjust the adjusted The barcode field acts as the final barcode field. By adding the barcode area adjustment module, the occurrence of missing codes can be effectively prevented and the stability of barcode recognition can be effectively improved. Wherein, the K and P are positive integers, and the specific values are selected according to the needs. In this embodiment, the K=P=15.

In this embodiment, the barcode area positioning module includes a barcode feature extraction module, a tilt correction module, a barcode left and right area positioning module, and a barcode upper and lower area positioning module. The connection relationship of each functional module is shown in Figure 2, and will not be repeated here detail.

In this embodiment, the image acquisition module is preferably an industrial camera with a GIGE interface. Correspondingly, the machine vision system host has a GIGE interface, and the machine vision system host is connected to the industrial camera through the GIGE interface. In addition, the machine vision system host is also provided with a light source control module, and the machine vision system host is connected to the light source module through the light source control module.

In the barcode detection and recognition system based on the machine vision system in this embodiment, the working principle/working process of each functional module is as described in the corresponding steps of the barcode detection and recognition method based on the machine vision system described in the above embodiment. This will not be repeated here.

It should be noted that, the above functional modules can be implemented by integrating multiple functional modules among them into one circuit functional module, or can be realized by making one functional module into multiple functional modules respectively.

Using the barcode detection and recognition method and system based on the machine vision system provided by the embodiment of the present invention, products with barcodes can be automatically scanned and barcodes can be recognized without manual operation, which has a high degree of automation and fast recognition speed. High work efficiency and other advantages; in the process of barcode recognition, even oblique barcodes, multiple barcodes on a single label, multiple barcodes on a single line, partly missing barcodes, and stained barcodes can obtain satisfactory recognition results, and at the same time can achieve arbitrary Automatic recognition of direction barcodes and multi-barcodes, and the advantages of fast recognition speed and high recognition efficiency, which can effectively meet the needs of large-scale automated industrial production of enterprises.

The content mentioned in the above examples is the preferred implementation mode of the present invention, and is not a limitation of the present invention. Without departing from the concept of the present invention, any obvious replacements are within the protection scope of the present invention.

Claims (10)

1. A barcode detection and recognition method based on a machine vision system, said method is applied in a machine vision system host, said machine vision system host is connected with a light source module and an image acquisition module, it is characterized in that said method comprises the following steps : S1. The image data acquisition step, through the cooperation of the image acquisition module and the light source module, image data acquisition is performed on labels with multiple barcodes, and the acquired image data is transmitted to the host of the device vision system; S2. Image data preprocessing step, the machine vision system host performs preprocessing on the image data, so as to separate the acquired image data from the label and the background, thereby filtering background interference; S3, the barcode area positioning step, the machine vision system host performs barcode feature extraction on the image data processed by the image data preprocessing step, and performs area positioning on the barcode to determine the location of all the barcodes on the label. The area is the barcode area with all the barcodes on the label; Wherein, the barcode area includes the upper and lower areas of the barcode and the left and right areas of the barcode, and the positioning of the barcode area includes the positioning of the upper and lower areas of the barcode and the positioning of the left and right areas of the barcode. , the number of barcodes existing in the same row of barcodes, so as to realize the regional positioning of all barcodes on the label; S4. A barcode recognition step, performing barcode recognition on each barcode in the barcode area. 2. The method according to claim 1, characterized in that: after the barcode area positioning step S3, before the barcode recognition step S4, the following steps are also included: S3A, the barcode area adjustment step, the left and right boundaries of the barcode area are respectively adjusted by expanding K pixels, and the upper and lower boundaries of the barcode area are respectively adjusted by reducing P pixels, and finally the adjusted The barcode area is used as the final barcode area, wherein the K and P are positive integers, and the specific values are selected according to requirements. 3. The method according to claim 2, characterized in that: in the barcode area positioning step S3, in the barcode upper and lower area positioning process, the normalized cross-correlation matching algorithm and the Hough transform line detection algorithm are used to Determine the upper and lower regions of the barcode. 4. The method according to claim 3, characterized in that: in the barcode area locating step S3, in the process of locating the left and right areas of the barcode, each barcode on the grayscale image of the preprocessed image Scan the gray value of the center line in a straight line to obtain the distribution map of the gray value of the center line of each barcode, and perform an accumulation operation on the pixel data on the distribution map of the gray value. Set the gray value array of the center line as LineA, and perform the accumulation operation Generate a new array after the LineB, according to the array LineA and data LineB, set up the center line gray value cumulative distribution map of the barcode, determine the straight line of the horizontal part in the horizontal part of the center line gray value cumulative distribution map. Barcode left and right area positioning. 5. The method according to any one of claims 1 to 4, characterized in that: in the image data preprocessing step S2, the preprocessing performed by the machine vision system host on the image data includes filtering, Denoising, enhancement and/or binarization. 6. A barcode detection and recognition system based on a machine vision system, the system includes a machine vision system host, the machine vision system host is connected with a light source module and an image acquisition module, and the image acquisition module is used to cooperate with the light source module, Carry out image data collection to labels with multiple barcodes, and transmit the collected image data to the machine vision system host, it is characterized in that the machine vision system host includes: An image data preprocessing module, configured to preprocess the image data, so as to separate the acquired image data from the label and the background, thereby filtering background interference; The barcode area positioning module is used to perform barcode feature extraction on the image data preprocessed by the image data preprocessing module, and perform area positioning on the barcode to determine the area where all the barcodes on the label are located. The area is a barcode area with all barcodes on said label; Wherein, the barcode area includes the upper and lower areas of the barcode and the left and right areas of the barcode, and the positioning of the barcode area includes the positioning of the upper and lower areas of the barcode and the positioning of the left and right areas of the barcode. , the number of barcodes existing in the same row of barcodes, so as to realize the regional positioning of all barcodes on the label; The barcode recognition module is used to perform barcode recognition on each barcode in the barcode area. 7. The system according to claim 6, characterized in that: the machine vision system host also includes: The barcode area adjustment module is used to adjust the left and right boundaries of the barcode area by expanding K pixels respectively, and adjust the upper and lower boundaries of the barcode area by shrinking P pixels, and finally adjust the adjusted The barcode area is used as the final barcode area, wherein the K and P are positive integers, and the specific values are selected according to requirements. 8. The system according to claim 7, characterized in that: the barcode area positioning module, in the process of positioning the upper and lower areas of the barcode, adopts a normalized cross-correlation matching algorithm and a Hough transform line detection algorithm to determine The upper and lower regions of the barcode. 9. The system according to claim 8, characterized in that: the barcode area positioning module, in the process of positioning the left and right areas of the barcode, the center line of each barcode on the grayscale image of the preprocessed image Carry out straight-line gray value scanning to obtain the gray value distribution map of the center line of each barcode, and perform cumulative operations on the pixel data on the gray value distribution map, set the center line gray value array as LineA, and generate after the cumulative operation A new array is LineB, and according to the array LineA and the data LineB, the center line gray value accumulation distribution map of the barcode is established, and the left and right sides of the bar code are determined by the straight lines in the horizontal part of the center line gray value accumulation distribution map. Regional targeting. 10. The system according to any one of claims 6-9, characterized in that: the image data preprocessing module performs preprocessing on the image data, including filtering, denoising, enhancement and/or binary processing.