CN106127817A - A kind of image binaryzation method based on passage - Google Patents
A kind of image binaryzation method based on passage Download PDFInfo
- Publication number
- CN106127817A CN106127817A CN201610504100.7A CN201610504100A CN106127817A CN 106127817 A CN106127817 A CN 106127817A CN 201610504100 A CN201610504100 A CN 201610504100A CN 106127817 A CN106127817 A CN 106127817A
- Authority
- CN
- China
- Prior art keywords
- pixel
- value
- row
- image
- passage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Facsimile Image Signal Circuits (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of image binaryzation method based on passage, including step: for needing to carry out the image of binary conversion treatment, image being considered as a picture element matrix I, the width of image is designated as w, is highly designated as h；Line by line to picture element matrix I process, scan the pixel of a line every time, collect three channel value red, blue, green and the information of gray value of statistical pixel point I [i] [j] when processing the i-th row, and calculate the meansigma methods of these four amounts in the i-th row pixel；I-th row each passage pixel value is added up and carried out binary conversion treatment according to meansigma methods and the threshold value of respective channel by each passage of the i-th row pixel；If i < h 1, i=i+1；Then repeat step S2 S3, next line pixel is carried out binary conversion treatment based on passage.Compared with prior art, the present invention can carry out quickly and accurately binary conversion treatment to the image of uneven illumination.
Description
Technical field
The present invention relates to digital image processing techniques field, particularly relate to a kind of image binaryzation method based on passage.
Background technology
From appearance and the proposition of artificial intelligence technology of the computer forties in 20th century, computer is just hoped to substitute people
The mental activity work of class, wherein handle the pictures is the most constantly by practice and research, it is desirable to calculate function
Enough automatically process image, finally the result processed is presented on people at the moment.It is suggested such as from the OCR technique twenties in 20th century
The present, OCR technique has reached higher level, also occurs in that a collection of OCR product simultaneously, the most most commonly used for image car plate
Identify, capture of i.e. number-plate number being taken pictures, then carry out number-plate number extraction by image processing techniques.Visual picture processes
Technology occupies certain status in computer realm.In the image processing arts, image binaryzation processes is Image semantic classification
In an important step, the quality of binary conversion treatment directly decides difficulty and the accuracy of subsequent step.
Although at present OCR technique research is the most effective, but is in developmental stage, to skills such as the pretreatment of image
Art also needs to continue research and practice.A kind of good image binaryzation processing method can be greatly improved the character feature in image
The efficiency extracted and identify and accuracy.From IBM Corporation after 20th century, the sixties proposed the scheme of Chinese Character Recognition, Chinese character is known
Do not start studied.In Chinese Character Recognition, need to extract each validity feature of Chinese character, and carry out with the Chinese character in feature database
Join.In OCR technique, owing to the quantity of information of coloured image is relatively big, and in feature extraction operation, it is desirable that character itself
Architectural feature, therefore carries out binary conversion treatment to image and advantageously reduces the complexity of subsequent operation.Binary image processes
The pixel value of the coloured image of input is analyzed, divides foreground and background according to the threshold value that special algorithm obtains.Two-value
Change processes and the character area i.e. foreground part of image is set to black, and background is set to white, prominent character area, significantly facilitates
Feature extraction and the carrying out of character recognition.
Binary processing method is wherein broadly divided into two classes: Global treatment and Local treatment.Global treatment is i.e. to whole figure
As taking a threshold value to pass judgment on, after calculating threshold value, this threshold value completes constant for holding until image procossing.Wherein have
Kittler algorithm and based on histogrammic global threshold algorithm etc..The most dynamic threshold value of Local treatment is passed judgment on, and is divided into greatly by image
If causing impartial stem portion, carrying out calculating the threshold value of current portions each time, then carrying out binary conversion treatment, until whole
Image procossing completes.Wherein there are Wall algorithm and Wellner adaptive-filtering thresholding algorithm.In image procossing, affect image
Process qualitative factor to have a lot.The situation that different Binarization methods is suitable for also is not quite similar.Although above-mentioned algorithm can solve greatly
Subproblem, but it is as the raising of the complexity of image, process the most how many results drawn allows people be unsatisfied with, particularly scheme
In the case of picture uneven illumination is even, the result images gone out after process is fuzzy or major part is not highlighted out.At word
In the image procossing identified, the whether outstanding quality directly influencing process of picture quality and discrimination.And major part is clapped
It cannot be guaranteed that good picture quality when of taking the photograph picture, modal is exactly the even situation of uneven illumination.
Summary of the invention
For overcoming the deficiencies in the prior art, solve to cause the image procossing of binaryzation poor effect to ask because uneven illumination is even
Topic, the present invention proposes a kind of image binaryzation method based on passage.
The technical scheme is that such: a kind of image binaryzation method based on passage, comprise the following steps:
S1: for needing to carry out the image of binary conversion treatment, image is considered as a picture element matrix I, the width note of image
For w, highly it is designated as h；
S2: line by line to picture element matrix I process, scans the pixel of a line every time, collects statistics picture when processing the i-th row
Three channel value red, blue, green of vegetarian refreshments I [i] [j] and the information of gray value, be designated as R respectivelyij、Gij、BijAnd GREYij, wherein
GREYijIt is gray value, GREYij=(Rij+Gij+Bij)/3, and calculate the meansigma methods of these four amounts in the i-th row pixel, it may be assumed that
redAvgi=(∑0≤j<wRij)/w,
greenAvgi=(∑0≤j<wGij)/w,
blueAvgi=(∑0≤j<wBij)/w,
greyAvgi=(∑0≤j<wGREYij)/w；
S3: by each passage of the i-th row pixel according to the meansigma methods of respective channel and threshold value to the i-th row each passage pixel
Value carries out adding up and carrying out binary conversion treatment:
S31: the redness of pixel, green, blue each channel value summation and gray value summation are designated as respectively
redSumi, greenSumi, blueSumi, greySumi；Satisfactory redness, green, blueness, the pixel of gray value
Number is designated as redCount respectivelyi, greenCounti, blueCounti, greyCounti；It is all initialized as 0；
S32: to red channel RiProcess, use redAvgi+ α, as cut off value, travels through the i-th row pixel, if Rij
<redAvgi+ α, then it is assumed that I [i] [j] is red pixel point, and red color channel value is added to summation redSumi, by redCounti
Add 1；Use identical method to Gi、BiAnd GREYiProcess, obtain greenSumi、greenCounti、blueSumi、
blueCounti、greySumi, greyCounti, wherein α is 1/10th of graphics standard variance；
S33: if redCountiIt is 0, the i-th row pixel is set to white entirely；If redCountiIt is not 0, makes redAvgi
Equal to the meansigma methods of all red pixel points, i.e. redAvgi=redSumi/redCounti；Traversal entire row of pixels point, if Rij<
redAvgi+ β, then arranging I [i] [j] is black；Wherein β is 1/10th of graphics standard variance；
S34: by red channel RiReplace with G respectivelyij、Bij、GREYijChannel information, repeats step S32-S33, revises two
Value result, unlike red channel, is only modified to black by original white pixel value；
S4: if i is < h-1, i=i+1；Then repeat step S2-S3, next line pixel is carried out binaryzation based on passage
Process.
Further, step S2 is for be scanned described picture element matrix the most line by line.
Further, step S32 is for from left to right to travel through the i-th row pixel.
The beneficial effects of the present invention is, compared with prior art, the image of uneven illumination can be carried out soon by the present invention
Speed and exactly binary conversion treatment.
Accompanying drawing explanation
Fig. 1 is present invention image binaryzation method based on passage flow chart.
Fig. 2 is the step S3 detailed step flow chart in Fig. 1.
Fig. 3 is the receipt picture before non-binaryzation.
Fig. 4 is the binaryzation result after the method for the present invention processes of the receipt picture in Fig. 3.
Detailed description of the invention
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Describe, it is clear that described embodiment is only a part of embodiment of the present invention rather than whole embodiments wholely.Based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under not making creative work premise
Embodiment, broadly falls into the scope of protection of the invention.
Refer to Fig. 1 and Fig. 2, a kind of image binaryzation method based on passage of the present invention, comprise the following steps:
For needing to carry out the image of binary conversion treatment, image being considered as a picture element matrix I, the width of image is designated as w,
Highly it is designated as h.Each element I [i] [j] in each pixel i.e. matrix, wherein i ∈ [0, h), j ∈ [0, w), all comprise r, g, b
The information of three passages.
The most line by line to picture element matrix I process, scan the pixel of a line every time.(i ∈ when processing the i-th row
[0, h)), collect three channel value red, blue, green and the information of gray value of statistics I [i] [j], be designated as R respectivelyij、Gij、Bij
And GREYij, wherein GREYijIt is gray value, uses GREY hereij=(Rij+Gij+Bij)/3, calculate in the i-th row pixel this four
The meansigma methods of individual amount, it may be assumed that
redAvgi=(∑0≤j<wRij)/w,
greenAvgi=(∑0≤j<wGij)/w,
blueAvgi=(∑0≤j<wBij)/w,
greyAvgi=(∑0≤j<wGREYij)/w。
By each passage of the i-th row pixel according to the meansigma methods of respective channel and threshold value to the i-th row each passage pixel value
Process and add up:
The redness of pixel, green, blue each channel value summation and gray value summation are designated as redSum respectivelyi,
greenSumi, blueSumi, greySumi；Satisfactory redness, green, blueness, the pixel number of gray value are remembered respectively
For redCounti, greenCounti, blueCounti, greyCounti；It is all initialized as 0.
To red channel RiProcess, use redAvgi+ α, as cut off value, from left to right travels through the i-th row pixel,
If Rij<redAvgi+ α, then it is assumed that I [i] [j] is red pixel point, and red color channel value is added to summation redSumi, will
redCountiAdd 1；Use identical method to Gi、BiAnd GREYiProcess, obtain greenSumi、greenCounti、
blueSumi、blueCounti、greySumi、greyCounti, wherein α is the value of the boundary best results that experiment draws, typically
It is taken as 1/10th of graphics standard variance.
The binaryzation result of result of calculation correction I [i] [j] according to above-mentioned steps:
If redCountiIt is 0, the i-th row pixel is set to white (255) entirely；If redCountiIt is not 0, makes redAvgiDeng
In the meansigma methods of all red pixel points, i.e. redAvgi=redSumi/redCounti；Traversal entire row of pixels, if Rij<
redAvgi+ β, then arranging I [i] [j] is black (0)；Wherein β is that experiment draws the value that regulating effect is optimal, and usually, β is figure
As standard variance 1/10th.
Then according to Gij、Bij、GREYijChannel information continues to revise binaryzation result, unlike r passage, only by former
First white pixel point is modified to black.
If i is < h-1, i=i+1；Then continue next line pixel is carried out binary conversion treatment based on passage.
As a example by the receipt image of supermarket, flow process of the present invention will be further elaborated below.
Fig. 3 is the supermarket receipt that undressed printing (illumination) is uneven, it can be seen that receipt image bright-dark degree is not
All, along with position down, its brightness is also gradually reduced.It is-20 that experiment acquisition processes the experiment value α value of red channel, processes
The experiment value α value of green channel is-25, and the experiment value α value processing blue channel is-23, processes the experiment value α of gray value
Value is-20.Regulation arranges the experiment value β value condition of monochrome pixels value and is respectively, and regulation red channel experiment value β value is
18, regulation green channel experiment value β value is 9, and regulation blue channel experiment value β value is 8, regulates gray scale experiment value β value
It is 8.
Step 2, is considered as a picture element matrix I by Fig. 3, and the width of image is designated as w=350, is highly designated as h=500.Often
Each element I [i] [j] in individual pixel i.e. matrix, wherein i ∈ [0,500), j ∈ [0,350), all comprise tri-passages of r, g, b
Information.
Step 3, from 0 to 500 line by line to picture element matrix I process, scans the pixel of a line every time.Process the i-th row
Time (i ∈ [0,500)), collect three passages red, blue, green and the information of gray value of statistics I [i] [j], be designated as R respectivelyij、
Gij、BijAnd GREYij(note: GREYijIt is gray value, uses GREY hereij=(Rij+Gij+Bij)/3), calculate the i-th row pixel
In these four amount meansigma methodss, it may be assumed that
redAvgi=(∑0≤j<350Rij)/350,
greenAvgi=(∑0≤j<350Gij)/350,
blueAvgi=(∑0≤j<350Bij)/350,
greyAvgi=(∑0≤j<350GREYij)/350。
I-th row each passage pixel value is entered by each passage of the i-th row pixel according to meansigma methods and the threshold value of respective channel
Row processes and adds up:
1) redness of pixel, green, blue each channel value summation and gray value summation are designated as redSum respectivelyi,
greenSumi, blueSumi, greySumi；Satisfactory redness, green, blueness, the pixel number of gray value are remembered respectively
For redCounti, greenCounti, blueCounti, greyCounti；It is all initialized as 0.
2) to red channel RiProcess, use redAvgi20 as cut off value, from left to right travels through the i-th row pixel,
If Rij<redAvgi20, then it is assumed that I [i] [j] is red pixel point, and red color channel value is added to summation redSumi, will
redCountiAdd 1；Use identical method to Gi、BiAnd GREYiProcess, obtain greenSumi、greenCounti、
blueSumi、blueCounti、greySumi, greyCounti。
Such as during i=200, during statistics 200 row, experiment show that the amount of several meansigma methods is redAvg200=138,
greenAvg200=168, blueAvg200=148, greyAvg200=151.
Following all kinds of pixels such as adding up red in the 200th row, green, blueness and gray scale of collecting, the most respectively meet
Condition: R200, j<138–20、G200, j<168–25、B200, j<148–23、GREY200, j< point of 151 20, and calculate all kinds of pixel
Summation sum of some correspondence and number count.Experiment draws, as i=200, and redSum200=6343, redCount200=77,
greenSum200=7757, greenCount200=72, blueSum200=6763, blueCount200=74, greySum200=
7286, greyCount200=77.
Step 4, according to the binaryzation result of result of calculation correction I [i] [j] of step 3:
1) if redCountiIt is 0, the i-th row pixel is set to white (255) entirely；If redCountiIt is not 0, makes redAvgi
Equal to the meansigma methods of all red pixel points, i.e. redAvgi=redSumi/redCounti；Traversal entire row of pixels, if Rij<
redAvgi+ β, then arranging I [i] [j] is black (0)；
2) according to Gij、Bij、GREYijChannel information continues to revise binaryzation result, unlike r passage, and only will be original
White pixel value is modified to black.
Such as during i=200, step 3 the pixels statistics data drawn can carry out two-value to the pixel of the 200th row
Changing, i.e. arranging pixel value is black (0) or white (255).
Experiment draws red pixel meansigma methods redAvg200=82, green pixel meansigma methods greenAvg200=107, blue
Pixel average blueAvg200=91, gray-scale pixels meansigma methods greyAvg200=94.
By red channel information, by R200,j< pixel of 82+18 is considered prospect, and arranging its pixel is black, otherwise sets
It is set to white；
According to the information such as green, blueness, gray value, the 200th row pixel is carried out three times and revises.G will be met respectively200, j<
107+9、B200, j<91+8、G200, j< the white pixel point of 94+8 is set to black, gradually revises the error of binaryzation for the first time.
Step 5, if i < 499, i=i+1；Return step 2, next line pixel is carried out at binaryzation based on passage
Reason.Such as during i=200, then need to return step 2 and carry out the binary conversion treatment of next line；If i=499, then whole image two is described
Value is disposed.The image 4 image after binary conversion treatment is as shown in Figure 4.
The above is the preferred embodiment of the present invention, it is noted that for those skilled in the art
For, under the premise without departing from the principles of the invention, it is also possible to make some improvements and modifications, these improvements and modifications are also considered as
Protection scope of the present invention.
Claims (3)
1. an image binaryzation method based on passage, it is characterised in that comprise the following steps:
S1: for needing to carry out the image of binary conversion treatment, image being considered as a picture element matrix I, the width of image is designated as w,
Highly it is designated as h；
S2: line by line to picture element matrix I process, scans the pixel of a line every time, collects statistical pixel point when processing the i-th row
Three channel value red, blue, green of I [i] [j] and the information of gray value, be designated as R respectivelyij、Gij、BijAnd GREYij, wherein
GREYijIt is gray value, GREYij=(Rij+Gij+Bij)/3, and calculate the meansigma methods of these four amounts in the i-th row pixel, it may be assumed that
redAvgi=(∑0≤j<wRij)/w,
greenAvgi=(∑0≤j<wGij)/w,
blueAvgi=(∑0≤j<wBij)/w,
greyAvgi=(∑0≤j<wGREYij)/w；
S3: the i-th row each passage pixel value is entered by each passage of the i-th row pixel according to meansigma methods and the threshold value of respective channel
Row statistics also carries out binary conversion treatment:
S31: the redness of pixel, green, blue each channel value summation and gray value summation are designated as redSum respectivelyi,
greenSumi, blueSumi, greySumi；Satisfactory redness, green, blueness, the pixel number of gray value are remembered respectively
For redCounti, greenCounti, blueCounti, greyCounti；It is all initialized as 0；
S32: to red channel RiProcess, use redAvgi+ α, as cut off value, travels through the i-th row pixel, if Rij<
redAvgi+ α, then it is assumed that I [i] [j] is red pixel point, and red color channel value is added to summation redSumi, by redCounti
Add 1；Use identical method to Gi、BiAnd GREYiProcess, obtain greenSumi、greenCounti、blueSumi、
blueCounti、greySumi, greyCounti, wherein α is 1/10th of graphics standard variance；
S33: if redCountiIt is 0, the i-th row pixel is set to white entirely；If redCountiIt is not 0, makes redAvgiIt is equal to
The meansigma methods of all red pixel points, i.e. redAvgi=redSumi/redCounti；Traversal entire row of pixels point, if Rij<
redAvgi+ β, then arranging I [i] [j] is black；Wherein β is 1/10th of graphics standard variance；
S34: by red channel RiReplace with G respectivelyij、Bij、GREYijChannel information, repeats step S32-S33, revises binaryzation
As a result, unlike red channel, only original white pixel value is modified to black；
S4: if i is < h-1, i=i+1；Then repeat step S2-S3, next line pixel is carried out binary conversion treatment based on passage.
2. image binaryzation method based on passage as claimed in claim 1, it is characterised in that step S2 be from top to bottom by
Described picture element matrix is scanned by row.
3. image binaryzation method based on passage as claimed in claim 1, it is characterised in that step S32 is from left to right
I-th row pixel is traveled through.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610504100.7A CN106127817B (en) | 2016-06-28 | 2016-06-28 | A kind of image binaryzation method based on channel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610504100.7A CN106127817B (en) | 2016-06-28 | 2016-06-28 | A kind of image binaryzation method based on channel |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106127817A true CN106127817A (en) | 2016-11-16 |
| CN106127817B CN106127817B (en) | 2018-11-06 |
Family
ID=57467597
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610504100.7A Expired - Fee Related CN106127817B (en) | 2016-06-28 | 2016-06-28 | A kind of image binaryzation method based on channel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106127817B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106778780A (en) * | 2016-12-14 | 2017-05-31 | 江苏维普光电科技有限公司 | A kind of edge-detected image binarization method based on GPU |
| CN110569684A (en) * | 2019-09-19 | 2019-12-13 | 南京全鉴信息技术有限公司 | Image binarization method for small-capacity cache |
| CN112257830A (en) * | 2020-10-23 | 2021-01-22 | 上海烟草集团有限责任公司 | Smoke box information identification method and system |
| CN112288754A (en) * | 2020-11-09 | 2021-01-29 | 珠海市润鼎智能科技有限公司 | Real-time binarization threshold value selection method for high-speed image |
| CN112733834A (en) * | 2021-03-30 | 2021-04-30 | 恒银金融科技股份有限公司 | Character area positioning method based on non-matrix window mode |
| CN114155241A (en) * | 2022-01-28 | 2022-03-08 | 浙江华睿科技股份有限公司 | Foreign matter detection method and device and electronic equipment |
| CN117649661A (en) * | 2024-01-30 | 2024-03-05 | 青岛超瑞纳米新材料科技有限公司 | Carbon nanotube preparation state image processing method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101515325A (en) * | 2009-04-08 | 2009-08-26 | 北京邮电大学 | Character extracting method in digital video based on character segmentation and color cluster |
| KR20110106261A (en) * | 2011-09-08 | 2011-09-28 | (주)이컴앤드시스템 | An extracting/decoding device for 2d barcode image of arbitrary contrast |
-
2016
- 2016-06-28 CN CN201610504100.7A patent/CN106127817B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101515325A (en) * | 2009-04-08 | 2009-08-26 | 北京邮电大学 | Character extracting method in digital video based on character segmentation and color cluster |
| KR20110106261A (en) * | 2011-09-08 | 2011-09-28 | (주)이컴앤드시스템 | An extracting/decoding device for 2d barcode image of arbitrary contrast |
Non-Patent Citations (2)
| Title |
|---|
| HUANG JIAN等: "Complex Scene Text Binarization based on Graph Cut", 《2016 IEEE INFORMATION TECHNOLOGY,NETWORKING,ELECTRONIC AND AUTOMATION CONTROL CONFERENCE》 * |
| 戴美玲等: "单幅彩色条纹投影的不连续物体表面三维形貌测量", 《光学学报》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106778780A (en) * | 2016-12-14 | 2017-05-31 | 江苏维普光电科技有限公司 | A kind of edge-detected image binarization method based on GPU |
| CN106778780B (en) * | 2016-12-14 | 2020-05-01 | 江苏维普光电科技有限公司 | GPU-based edge detection image binarization method |
| CN110569684A (en) * | 2019-09-19 | 2019-12-13 | 南京全鉴信息技术有限公司 | Image binarization method for small-capacity cache |
| CN110569684B (en) * | 2019-09-19 | 2023-05-12 | 南京全鉴信息技术有限公司 | Image binarization method for small-capacity cache |
| CN112257830A (en) * | 2020-10-23 | 2021-01-22 | 上海烟草集团有限责任公司 | Smoke box information identification method and system |
| CN112288754A (en) * | 2020-11-09 | 2021-01-29 | 珠海市润鼎智能科技有限公司 | Real-time binarization threshold value selection method for high-speed image |
| CN112288754B (en) * | 2020-11-09 | 2023-12-05 | 珠海市润鼎智能科技有限公司 | Real-time binarization threshold selection method for high-speed image |
| CN112733834A (en) * | 2021-03-30 | 2021-04-30 | 恒银金融科技股份有限公司 | Character area positioning method based on non-matrix window mode |
| CN112733834B (en) * | 2021-03-30 | 2021-07-06 | 恒银金融科技股份有限公司 | Character area positioning method based on non-matrix window mode |
| CN114155241A (en) * | 2022-01-28 | 2022-03-08 | 浙江华睿科技股份有限公司 | Foreign matter detection method and device and electronic equipment |
| CN117649661A (en) * | 2024-01-30 | 2024-03-05 | 青岛超瑞纳米新材料科技有限公司 | Carbon nanotube preparation state image processing method |
| CN117649661B (en) * | 2024-01-30 | 2024-04-12 | 青岛超瑞纳米新材料科技有限公司 | Carbon nanotube preparation state image processing method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106127817B (en) | 2018-11-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106127817A (en) | A kind of image binaryzation method based on passage | |
| CN105046196B (en) | Front truck information of vehicles structuring output method based on concatenated convolutional neutral net | |
| CN101515325B (en) | Character extracting method in digital video based on character segmentation and color cluster | |
| CN106934386B (en) | A kind of natural scene character detecting method and system based on from heuristic strategies | |
| CN104408449B (en) | Intelligent mobile terminal scene literal processing method | |
| CN1312625C (en) | Character extracting method from complecate background color image based on run-length adjacent map | |
| CN110706235B (en) | Far infrared pedestrian detection method based on two-stage cascade segmentation | |
| CN105825216A (en) | Method of locating text in complex background image | |
| CN110969129A (en) | End-to-end tax bill text detection and identification method | |
| CN112750106A (en) | Nuclear staining cell counting method based on incomplete marker deep learning, computer equipment and storage medium | |
| CN104598907B (en) | Lteral data extracting method in a kind of image based on stroke width figure | |
| CN109446984A (en) | Traffic sign recognition method in natural scene | |
| CN113705579A (en) | Automatic image annotation method driven by visual saliency | |
| CN109800762A (en) | A kind of fuzzy license plate recognizer based on the Dynamic Matching factor | |
| CN106570885A (en) | Background modeling method based on brightness and texture fusion threshold value | |
| CN111340032A (en) | Character recognition method based on application scene in financial field | |
| CN110781898A (en) | Unsupervised learning method for Chinese character OCR post-processing | |
| CN112819837A (en) | Semantic segmentation method based on multi-source heterogeneous remote sensing image | |
| Wang et al. | MRF based text binarization in complex images using stroke feature | |
| CN108664968B (en) | Unsupervised text positioning method based on text selection model | |
| CN106066887A (en) | A kind of sequence of advertisements image quick-searching and the method for analysis | |
| CN111783673B (en) | Video segmentation improvement method based on OSVOS | |
| CN113657407A (en) | High-recall-rate accurate positioning method for large-amplitude picture characters | |
| Zhao et al. | An effective binarization method for disturbed camera-captured document images | |
| Lézoray et al. | Histogram and watershed based segmentation of color images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20181106 Termination date: 20190628 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |