CN117893502A - Image detection method, device, equipment and storage medium - Google Patents
Image detection method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN117893502A CN117893502A CN202410061711.3A CN202410061711A CN117893502A CN 117893502 A CN117893502 A CN 117893502A CN 202410061711 A CN202410061711 A CN 202410061711A CN 117893502 A CN117893502 A CN 117893502A
- Authority
- CN
- China
- Prior art keywords
- image
- template
- printing
- characteristic points
- feature point
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 47
- 238000003860 storage Methods 0.000 title claims description 7
- 238000004364 calculation method Methods 0.000 claims abstract description 24
- 238000012216 screening Methods 0.000 claims abstract description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 11
- 238000001914 filtration Methods 0.000 claims description 10
- 239000013598 vector Substances 0.000 claims description 9
- 230000000877 morphologic effect Effects 0.000 claims description 4
- 238000004519 manufacturing process Methods 0.000 description 9
- 239000000047 product Substances 0.000 description 9
- 238000000034 method Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000004806 packaging method and process Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 239000006227 byproduct Substances 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
Abstract
The invention relates to an image detection method, which comprises the following steps: s1: after carrying out coordinate alignment processing on the printed image, extracting printing characteristic points; after coordinate alignment treatment is carried out on the template image, extracting template feature points; s2: carrying out coordinate matching on the printing characteristic points and the template characteristic points, and screening the printing characteristic points and the template characteristic points with the same coordinates as characteristic point combinations; s3: calculating the similarity between the printing characteristic points and the template characteristic points in each characteristic point combination; s4: taking the feature point combination with the similarity larger than the first threshold value as a matching feature point combination; s5: judging whether the number of the matched feature point combinations is larger than a second threshold value or not; if yes, determining that the printed image is qualified; if not, determining that the printed image is unqualified. According to the invention, only the similarity calculation is needed to be carried out on the feature point combination matched with the coordinates, so that unnecessary matching calculation is reduced, the calculation speed is effectively improved, and the detection efficiency of the printed image is further improved.
Description
Technical Field
The present invention relates to the field of printers, and in particular, to an image detection method, apparatus, device, and storage medium.
Background
On different products or packages, there are many different patterns formed by text, figures or two-dimensional codes, and related information of the products can be quickly checked through pattern recognition. In practical applications, when a product is transported on a production line, a printer is used to print on a product package being transported to form the different printed images, and when the printed image is formed on the production line, the printed image needs to be quickly identified to detect whether the printed image is qualified. However, printers may print on packaging surfaces of different sizes and proportions, and the brightness, contrast and sharpness of the printed image may vary due to the effects of light conditions and environmental noise and high-speed transmission on the production line, which may result in greater difficulty in detecting the printed image.
In the prior art, the identification detection of the printed image formed on the production line is generally realized by using SIFT (SCALE INVARIANT Feature Transform, scale-invariant feature transform matching algorithm), however, when the image is detected by using SIFT, the similarity between each feature point in the printed image and all feature points in the template image needs to be calculated, which results in huge calculation amount. As shown in fig. 1, the number of feature points is large, and the existing detection method can calculate all the pixel points not under the same coordinate, and can perform matching calculation on the pixel points under different coordinates, so that the complexity and the calculation amount of calculation are increased obviously. On the production line, the detection of the image is usually required to be performed in real time or at a high speed, and the detection of the printed image is required to be completed in a very short time, so that the unqualified printed image is determined, and therefore, the image detection method has huge calculation amount, and the detection efficiency of the printed image on the production line is low.
Disclosure of Invention
In view of the above, an object of the present invention is to provide an image detection method, apparatus, device, and storage medium that can reduce the amount of computation and improve the detection efficiency.
In a first aspect, the present invention provides an image detection method, including:
S1: after carrying out coordinate alignment processing on the printed image, extracting printing characteristic points; after coordinate alignment treatment is carried out on the template image, extracting template feature points;
S2: carrying out coordinate matching on the printing characteristic points and the template characteristic points, and screening the printing characteristic points and the template characteristic points with the same coordinates as characteristic point combinations;
S3: calculating the similarity between the printing characteristic points and the template characteristic points in each characteristic point combination;
S4: taking the feature point combination with the similarity larger than the first threshold value as a matching feature point combination;
S5: judging whether the number of the matched feature point combinations is larger than a second threshold value or not; if yes, determining that the printed image is qualified; if not, determining that the printed image is unqualified.
Further, the step S1 specifically includes:
S1_1: sequentially carrying out gray level processing and binarization processing on the printing image to obtain a processed binarized printing image;
s1_2: filtering the binarized printing image to obtain a processed first filtering image;
S1_3: performing morphological expansion on the processed first filter image to obtain a target printing area of the first filter image, and intercepting the first filter image based on the target printing area to obtain an optimized first printing image;
S1_4: performing binarization processing on the template image to obtain a processed first template image;
S1_5: performing alignment processing on the first printing image and the first template image to obtain an optimized printing image and an optimized template image;
S1_6: extracting the printing characteristic points from the optimized printing image, and extracting the template characteristic points from the optimized template image.
Further, in step S16, the print feature points are extracted from the optimized print image according to a preset extraction algorithm, and the template feature points are extracted from the optimized template image according to a preset extraction algorithm.
Further, the preset extraction algorithm comprises SIFT.
Further, in step S3, the similarity calculation formula is:
wherein cos θ is the similarity, a is the feature vector corresponding to the print feature point, and B is the feature vector corresponding to the template feature point.
Further, the first threshold is 0.8.
Further, the second threshold is 30.
In a second aspect, the present invention also provides an image detection apparatus, including:
The extraction module is used for extracting printing characteristic points after carrying out coordinate alignment processing on the printing image; after coordinate alignment treatment is carried out on the template image, extracting template feature points;
The first matching module is used for carrying out coordinate matching on the printing characteristic points and the template characteristic points, and screening the printing characteristic points and the template characteristic points with the same coordinates as characteristic point combinations;
The computing module is used for computing the similarity between the printing characteristic points and the template characteristic points in each characteristic point combination;
the second matching module is used for taking the feature point combination with similarity larger than the first threshold value as a matching feature point combination;
the determining module is used for judging whether the number of the matched feature point combinations is larger than a second threshold value; if yes, determining that the printed image is qualified; if not, determining that the printed image is unqualified.
In a third aspect, the present invention also provides a computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor is configured to implement the image detection method according to any one of the above-mentioned aspects when executing the computer program.
In a fourth aspect, the present invention further provides a computer storage medium, which includes a computer program, wherein the computer program, when executed by a processor, implements the image detection method according to any one of the above.
Compared with the prior art, the image detection method provided by the invention has the advantages that when the printed image is detected, the printed image and the standard image are subjected to coordinate alignment, after the corresponding characteristic points are extracted from the obtained optimized printed image and the optimized template image by utilizing the preset recognition algorithm, the characteristic point combination with the same coordinates is further screened out through coordinate matching, and only the similarity calculation is needed to be carried out on the characteristic point combination with the matched coordinates, so that the unnecessary matching calculation is reduced, the calculation speed is effectively improved, and the detection efficiency of the printed image is further improved.
For a clearer understanding of the present invention, specific embodiments of the invention will be set forth in the following description taken in conjunction with the accompanying drawings.
Drawings
FIG. 1 is a schematic diagram of a matching effect in image detection in the prior art;
FIG. 2 is a schematic diagram of an overall flow of an image detection method according to the present invention;
FIG. 3 is a schematic diagram of a printed image after gray scale conversion according to the present invention;
FIG. 4 is a schematic diagram of a gray scale image filtered according to the present invention;
FIG. 5 is a schematic view of a filtered image after secondary denoising according to the present invention;
FIG. 6 is a schematic view of an image obtained after the optimization process of the present invention;
FIG. 7 is a schematic illustration of a standard print image and a standard template image obtained after the alignment process of the present invention;
FIG. 8 is a schematic diagram of coordinate matching for printed feature points and template feature points in the present invention;
FIG. 9 is a probability distribution of detection accuracy for a printed image of the present invention;
Fig. 10 is a schematic diagram of a matching effect of the image detection method provided by the present invention.
Detailed Description
In order to improve the detection efficiency of the printed image, the invention firstly performs coordinate alignment processing on the printed image and the template image when detecting the printed image, respectively extracts the printed feature points and the template feature points from the optimized printed image and the optimized template image which are subjected to the coordinate alignment processing by utilizing a preset recognition algorithm, screens out feature point combinations with the same coordinates by coordinate matching, and only performs similarity calculation on the printed feature points and the template feature points in the screened feature point combinations with the same coordinates to replace the existing mode of calculating the similarity between each feature point descriptor in the first image and all feature point descriptors in the second image, thereby realizing the improvement of the detection efficiency of the printed image on the premise of greatly reducing the calculated amount.
Referring to fig. 2, the present application provides an image detection method, in which when a printer prints on a product or a package on a production line to form a printed image, a processor detects the printed image by using the image detection method provided by the present application, so that the calculated amount is reduced, and meanwhile, whether the printed image is qualified can be rapidly detected, and the detection efficiency of the printed image is improved.
Specifically, in the present embodiment, the processor performs image detection by:
s1: after carrying out coordinate alignment processing on the printed image, extracting printing characteristic points; and extracting template feature points after carrying out coordinate processing on the template image.
Specifically, the product is transmitted through a production line after production is completed, the printer starts to print on the product package after detecting the product package to obtain a printing image, and the industrial camera acquires the printing image from the product package and simultaneously acquires a pre-stored template image. Preferably, the printer can be a jet printer or an ink jet printer, the printed image can be a code-jet image comprising numbers, letters or two-dimensional codes, and the industrial camera can read the printed image through opencv.
After the print image and the template image are obtained, coordinate alignment processing is performed on the obtained print image and the template image, and corresponding feature points are extracted from the optimized print image and the optimized template image after the coordinate alignment processing. The step is therefore to extract print feature points based on the optimized print image obtained after the preprocessing and to extract template feature points based on the obtained optimized template image in the preprocessing process for the print image and the template image. Specifically, the method comprises the following steps S1_1 to S1_5:
S1_1: sequentially carrying out gray level processing and binarization processing on the printing image to obtain a processed binarized printing image;
After the printed image is read, the processor can perform gray processing on the printed image by using a conventional gray image conversion function, and convert the printed image into a gray image, as shown in fig. 3 in detail; and further performing binarization processing on the obtained gray level image by using a binarization function to obtain a binarized printing image.
S1_2: filtering the binarized printing image to obtain a processed first filtering image;
Specifically, in a binarized print image obtained by sequentially performing gradation processing and binarization processing on the print image, there may be noise caused by product packaging or printing paper, as shown in fig. 4. Therefore, after the binarized printed image is obtained, filtering denoising is further performed on the binarized printed image, so as to obtain a processed first filtered image, as shown in fig. 5. Optionally, a gaussian kernel function is used to perform a gaussian filter process on the binarized print image.
S1_3: performing morphological expansion on the processed first filter image to obtain a target printing area of the first filter image, and intercepting the first filter image based on the target printing area to obtain an optimized first printing image;
Specifically, filtering and denoising the binarized image generally can eliminate most of noise, but noise with a part of extremely large noise point area can not be eliminated after single primary filtering, so when the processed first filtered image is obtained, secondary filtering processing is further performed on the first filtered image, morphological expansion is performed by using a Gaussian kernel, all noise in a non-printing area can be removed, the outline of a target printed image in the first filtered image is found, a rectangular frame is drawn outside the outline, and therefore the target printed area of the first filtered image is identified, and the optimized first printed image is obtained by cutting out from the first filtered image based on the target printed area, as shown in fig. 6.
S1_4: performing binarization processing on the template image to obtain a processed first template image;
Specifically, the template image is an image for detecting a print image stored in advance in the processor, and it is understood that the template image records only key point information of the image, and thus the template image is much smaller than the size of the print image. After the template image is obtained, binarization processing is carried out on the template image, and a binarized template image after black-and-white processing is obtained to serve as a first template image for comparison.
S1_5: performing alignment processing on the first printing image and the first template image to obtain an optimized printing image and an optimized template image;
Specifically, since the sizes of the first print image and the first template image are different, after the first print image and the first template image are obtained, the obtained first print image and the first template image are further subjected to alignment processing, so that the optimized print image and the optimized template image obtained after the processing have the same size and are aligned in coordinates. The resulting optimized print image and optimized template image are shown in fig. 7.
S1_6: extracting the printing characteristic points from the optimized printing image, and extracting the template characteristic points from the optimized template image.
In the step S1, before the print image and the template image are obtained and the feature points of the print image and the template image are extracted, the print image and the template image are first subjected to coordinate alignment in advance, so that the obtained optimized print image and the obtained optimized template image are basically aligned in coordinates, and the print feature points and the template feature points are extracted based on the optimized print image after the coordinate alignment.
S2: and carrying out coordinate matching on the printing characteristic points and the template characteristic points, and screening the printing characteristic points and the template characteristic points with the same coordinates as characteristic point combinations.
Specifically, the size of the optimized print image and the optimized template image of the coordinate alignment process is the same, so after the optimized print image and the optimized template image of the alignment process are obtained, corresponding print feature points and template feature points are extracted based on the optimized print image and the optimized template image after the alignment process, wherein understandably, there may be a plurality of extracted print feature points, and there may be a plurality of template feature points. In the step, the extracted printing characteristic points and template characteristic points are subjected to coordinate matching, and the printing characteristic points and the template characteristic points with the same coordinates are obtained through screening to form a plurality of corresponding characteristic point combinations.
In one example, the plurality of printing feature points may be extracted from the standard printing image and the plurality of template feature points may be extracted from the standard template image according to a preset extraction algorithm, respectively, and in particular, the preset extraction algorithm includes a SIFT algorithm. For example, the print feature point d1 extracted from the optimized print image using the SIFT algorithm can be expressed as: [ A, a 1 ] the template feature point d2 extracted from the optimized template image can be expressed as: [ B, b 1 ] A is a feature vector for representing the printing feature points, and a 1 is a coordinate for representing the printing feature points; and B is used for representing the feature vector of the template feature points, and B 1 is used for representing the coordinates of the template feature points. Specifically, the print feature points and the template feature points may be represented by 128-dimensional feature vectors.
Specifically, if the coordinates a 1 of the print feature point are the same as the coordinates b 1 of the template feature point, it is determined that the print feature point d1 extracted from the optimized print image and the template feature point d2 extracted from the optimized template image are coordinate-matched, and at this time, the print feature point d1 and the template feature point d2 having the same coordinates form a feature point combination as a pair of coordinate-matched feature points. By matching the coordinates of the plurality of print feature points and the template feature points, a plurality of feature point combinations of the coordinate matching can be obtained. Further, it is understood that the print feature point d1 and the template feature point d2 included in the resulting feature point combination at this time are merely coordinate-matched, and do not mean that the print feature point d1 and the template feature point d2 are matched. Referring to fig. 8, a coordinate matching effect diagram for the print feature points and the template feature points is given. As can be seen from fig. 8, the printed feature points and the template feature points with the same coordinates are used as a feature point combination to perform subsequent similarity calculation, and no irrelevant feature points are matched, so that the subsequent calculation speed of the feature point similarity is improved.
S3: and calculating the similarity between the printing characteristic points and the template characteristic points in each characteristic point combination.
In the step, after a plurality of feature point combinations with matched coordinates are obtained, the similarity between the printing feature point d1 and the template feature point d2 in each feature point combination with matched coordinates is calculated to represent whether the corresponding printing feature point d1 and template feature point d2 are matched. Specifically, whether the corresponding print feature point d1 and template feature point d2 in the feature point combination having the same coordinates match or not may be determined by the cosine similarity between the feature vector a of the print feature point d1 and the feature vector B of the template feature point d 2. The specific calculation formula is as follows:
s4: and selecting the feature point combination with the similarity larger than the first threshold value as a matched feature point combination.
Specifically, whether the calculated similarity between the printing feature points and the template feature points is larger than a first threshold is judged, when the similarity is larger than the first threshold, the printing feature points in the feature point combination are determined to be matched with the template feature points, and the corresponding feature point combination is selected as a matched feature point combination. Preferably, the first threshold may be 0.8, and when the cosine similarity is greater than 0.8, the print feature point and the template feature point are considered to be matched, and the corresponding feature point combination is selected as the matched feature point combination. And when the calculated similarity between the first feature point and the second feature point is smaller than a first threshold value, the fact that the similarity between the first feature point and the second feature point which are matched in coordinates in the feature point combination is not matched is indicated, and the feature point combination is abandoned.
S5: judging whether the number of the matched feature point combinations is larger than a second threshold value, if so, determining that the printed image is qualified; if not, determining that the printed image is unqualified.
In the above step, the similarity calculation between the printed feature points and the template feature points in all the feature point combinations matched with the coordinates is completed, the feature point combination with the similarity larger than the first threshold value is determined as the matched feature point combination, and the number n of the matched feature point combinations is obtained. In the step, whether the number n of the matched characteristic point combinations is larger than a second threshold value is further judged, and when the number n of the matched characteristic point combinations is larger than the second threshold value, the print image is determined to be qualified and is a positive sample; if the number n of the matched feature point combinations is not greater than the second threshold, the fact that the number n of the matched feature points in the standard printed image and the standard template image is too small is indicated to be a negative sample, and the printed image detected as the negative sample can be recalled. Specifically, referring to fig. 9, a probability distribution map of detection accuracy of a printed image by the image detection method of the present application is given. As can be seen from FIG. 9, nearly 100% of the detected images do not match the standard template image, and the images are determined to be unqualified and recalled, and the number of matched characteristic points is less than 20; and nearly 100% of the images are detected to be matched with the standard template image, and the images are determined to be qualified printing images, wherein the number of matched feature points is basically more than 30. In the present embodiment, therefore, it is preferable that the second threshold value be set to 30 so as to quickly determine whether the printed image is acceptable.
Compared with the prior art, in the image detection method provided by the invention, when the processor detects the image printed by the printer, the processor corrects the coordinates of the printed image and the template image before extracting the characteristic points of the image to obtain the optimized printed image and the optimized template image after coordinate alignment, extracts the characteristic points of the optimized printed image and the optimized template image after coordinate alignment, screens the extracted characteristic points in a coordinate judgment mode to obtain the characteristic point combination of coordinate matching, and performs similarity matching calculation on the characteristic points of the characteristic point combination of coordinate matching obtained by screening, thereby reducing a large number of unnecessary calculation, effectively reducing the calculation complexity and calculation amount for image detection, and improving the detection efficiency of the image. Referring specifically to fig. 10, compared with fig. 1, the calculation complexity and calculation amount are obviously improved.
Specifically, in an embodiment, the present invention further provides an image detection apparatus, including:
The extraction module is used for extracting printing characteristic points after carrying out coordinate alignment processing on the printing image; after coordinate alignment treatment is carried out on the template image, extracting template feature points;
The first matching module is used for carrying out coordinate matching on the printing characteristic points and the template characteristic points, and screening the printing characteristic points and the template characteristic points with the same coordinates as characteristic point combinations;
The computing module is used for computing the similarity between the printing characteristic points and the template characteristic points in each characteristic point combination;
the second matching module is used for taking the feature point combination with similarity larger than the first threshold value as a matching feature point combination;
the determining module is used for judging whether the number of the matched feature point combinations is larger than a second threshold value; if yes, determining that the printed image is qualified; if not, determining that the printed image is unqualified.
In an embodiment, the present invention further provides a computer device, including a memory and a processor, where the memory stores a computer program, and the processor is configured to implement the image detection method according to any one of the above when executing the computer program.
In an embodiment, the invention also provides a computer storage medium, comprising a computer program, characterized in that the computer program, when executed by a processor, implements the image detection method according to any of the preceding claims.
The above examples illustrate only a few embodiments of the invention, which are described in detail and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention.
Claims (10)
1. An image detection method, comprising:
S1: after carrying out coordinate alignment processing on the printed image, extracting printing characteristic points; after coordinate alignment treatment is carried out on the template image, extracting template feature points;
S2: carrying out coordinate matching on the printing characteristic points and the template characteristic points, and screening the printing characteristic points and the template characteristic points with the same coordinates as characteristic point combinations;
S3: calculating the similarity between the printing characteristic points and the template characteristic points in each characteristic point combination;
S4: taking the feature point combination with the similarity larger than the first threshold value as a matching feature point combination;
S5: judging whether the number of the matched feature point combinations is larger than a second threshold value or not; if yes, determining that the printed image is qualified; if not, determining that the printed image is unqualified.
2. The image detection method according to claim 1, wherein the step S1 specifically includes:
S1_1: sequentially carrying out gray level processing and binarization processing on the printing image to obtain a processed binarized printing image;
s1_2: filtering the binarized printing image to obtain a processed first filtering image;
S1_3: performing morphological expansion on the processed first filter image to obtain a target printing area of the first filter image, and intercepting the first filter image based on the target printing area to obtain an optimized first printing image;
S1_4: performing binarization processing on the template image to obtain a processed first template image;
S1_5: performing alignment processing on the first printing image and the first template image to obtain an optimized printing image and an optimized template image;
S1_6: extracting the printing characteristic points from the optimized printing image, and extracting the template characteristic points from the optimized template image.
3. The image detection method according to claim 2, wherein in step S16, the printing feature points are extracted from the optimized printing image according to a preset extraction algorithm, and the template feature points are extracted from the optimized template image according to a preset extraction algorithm.
4. The image detection method according to claim 3, wherein the preset extraction algorithm comprises SIFT.
5. The image detection method according to claim 1, wherein in step S3, the similarity calculation formula is:
wherein cos θ is the similarity, a is the feature vector corresponding to the print feature point, and B is the feature vector corresponding to the template feature point.
6. The image detection method according to any one of claims 1 to 5, wherein the first threshold value is 0.8.
7. The image detection method according to any one of claims 1 to 5, wherein the second threshold value is 30.
8. An image detection apparatus, comprising:
The extraction module is used for extracting printing characteristic points after carrying out coordinate alignment processing on the printing image; after coordinate alignment treatment is carried out on the template image, extracting template feature points;
The first matching module is used for carrying out coordinate matching on the printing characteristic points and the template characteristic points, and screening the printing characteristic points and the template characteristic points with the same coordinates as characteristic point combinations;
The computing module is used for computing the similarity between the printing characteristic points and the template characteristic points in each characteristic point combination;
the second matching module is used for taking the feature point combination with similarity larger than the first threshold value as a matching feature point combination;
the determining module is used for judging whether the number of the matched feature point combinations is larger than a second threshold value; if yes, determining that the printed image is qualified; if not, determining that the printed image is unqualified.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor is adapted to implement the image detection method of any of claims 1-7 when executing the computer program.
10. A computer storage medium comprising a computer program, characterized in that the computer program, when executed by a processor, implements the image detection method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410061711.3A CN117893502A (en) | 2024-01-15 | 2024-01-15 | Image detection method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410061711.3A CN117893502A (en) | 2024-01-15 | 2024-01-15 | Image detection method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117893502A true CN117893502A (en) | 2024-04-16 |
Family
ID=90643927
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410061711.3A Pending CN117893502A (en) | 2024-01-15 | 2024-01-15 | Image detection method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117893502A (en) |
-
2024
- 2024-01-15 CN CN202410061711.3A patent/CN117893502A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110046529B (en) | Two-dimensional code identification method, device and equipment | |
| CN104751142B (en) | A kind of natural scene Method for text detection based on stroke feature | |
| CN110378310B (en) | Automatic generation method of handwriting sample set based on answer library | |
| CN107563377A (en) | It is a kind of to detect localization method using the certificate key area of edge and character area | |
| CN110020692B (en) | Handwriting separation and positioning method based on print template | |
| US20080310721A1 (en) | Method And Apparatus For Recognizing Characters In A Document Image | |
| CN110298353B (en) | Character recognition method and system | |
| CN110598566A (en) | Image processing method, device, terminal and computer readable storage medium | |
| CN112016563B (en) | Method for identifying authenticity of circular seal | |
| US20190354756A1 (en) | Character recognition method | |
| CN113901952A (en) | Print form and handwritten form separated character recognition method based on deep learning | |
| CN110163152A (en) | Antiforge recognizing method, method for anti-counterfeit, system, device end and storage medium | |
| CN111814673A (en) | Method, device and equipment for correcting text detection bounding box and storage medium | |
| CN115063802A (en) | PSENet-based circular seal identification method, device and medium | |
| Mei et al. | A novel framework for container code-character recognition based on deep learning and template matching | |
| CN107292255B (en) | Handwritten number recognition method based on feature matrix similarity analysis | |
| CN108090728B (en) | Express information input method and system based on intelligent terminal | |
| CN111445402B (en) | Image denoising method and device | |
| Dai et al. | Scene text detection based on enhanced multi-channels MSER and a fast text grouping process | |
| Shukla et al. | An approach for skew detection using hough transform | |
| CN116503871A (en) | Character segmentation preprocessing method, terminal device and computer readable storage medium | |
| CN116469090A (en) | Method and device for detecting code spraying pattern, electronic equipment and storage medium | |
| CN117893502A (en) | Image detection method, device, equipment and storage medium | |
| CN111814780B (en) | Bill image processing method, device, equipment and storage medium | |
| Kumar et al. | Online handwritten character recognition for Telugu language using support vector machines |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |