CN110197181B - Cable character detection method and system based on OCR - Google Patents
Cable character detection method and system based on OCR Download PDFInfo
- Publication number
- CN110197181B CN110197181B CN201910473296.1A CN201910473296A CN110197181B CN 110197181 B CN110197181 B CN 110197181B CN 201910473296 A CN201910473296 A CN 201910473296A CN 110197181 B CN110197181 B CN 110197181B
- Authority
- CN
- China
- Prior art keywords
- characters
- cable
- template
- character
- detected
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Character Discrimination (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a cable character detection method and a system based on OCR, relating to the field of image detection and recognition, wherein the method comprises the steps of obtaining characters on a normal cable sheath, and taking each character as an element; all elements are connected in sequence to form a bidirectional linked list, and a template is generated based on characters in the bidirectional linked list; training a classifier by taking the template as a training sample; and inputting the characters on the cable to be detected into the trained classifier, comparing and identifying the characters on the cable to be detected with the characters in the template by the classifier, if the characters on the cable to be detected can be identified, indicating that the characters on the cable to be detected are not defective, otherwise, indicating that the characters on the cable to be detected are defective. The invention can quickly detect whether the printed characters on the cable have defects, replaces manual detection and reduces the production cost.
Description
Technical Field
The invention relates to the field of image detection and recognition, in particular to a cable character detection method and system based on OCR.
Background
In the production process of the optical cable, for characters on the appearance of the sheath, when a lettering block is improperly installed or is pressed too shallowly, the characters are not displayed clearly, and partial characters are lost, so that the information display such as the model and length of the optical cable is incomplete, the laying construction in the later stage is influenced, and the customer experience is damaged; when the printer imprints too deeply, the cable core sleeve is stressed to cause abnormal optical fiber attenuation, and the product quality is seriously influenced.
The current character detection mode is manual detection, rely on operating personnel visual observation to discern the character quality on sheath surface, when optical cable sheath production speed is very fast, operating personnel detects speed and is difficult to keep up, and optical cable production mode is the continuity, it can arouse fatigue to last to detect, the condition of lou examining appears easily, and simultaneously, different operating personnel's subjective consciousness is different, it is difficult to unify to detect the standard, the production efficiency that leads to the optical cable receives the restriction, product quality is difficult to obtain the promotion, and the cost of labor constantly rises.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a cable character detection method and system based on OCR (optical character recognition), which can quickly detect whether the printed characters on the cable have defects or not, replace manual detection and reduce the production cost.
In order to achieve the above purposes, the technical scheme adopted by the invention is as follows:
acquiring characters on a normal cable sheath, and taking each character as an element;
all elements are connected in sequence to form a bidirectional linked list, and a template is generated based on characters in the bidirectional linked list;
training a classifier by taking the template as a training sample;
and inputting the characters on the cable to be detected into the trained classifier, comparing and identifying the characters on the cable to be detected with the characters in the template by the classifier, if the characters on the cable to be detected can be identified, indicating that the characters on the cable to be detected are not defective, otherwise, indicating that the characters on the cable to be detected are defective.
On the basis of the technical scheme, the acquiring of the characters on the normal cable sheath takes each character as an element, and the specific steps comprise:
acquiring an image of the surface of a sheath of a normal cable, and binarizing the image;
extracting the region where the character is located in the image based on the mode of obtaining the connected domain;
and acquiring characters in the region, fusing each character according to a set interval, and taking each character as an element.
On the basis of the technical proposal, the device comprises a shell,
the acquired images of the surface of the normal cable sheath are multiple frames, and elements in each frame of image are connected into a two-way linked list;
the template is generated based on the characters in the bidirectional linked list, and the specific steps are as follows: based on the similarity between each section of characters, splicing the bidirectional linked list corresponding to each frame of image, wherein the spliced bidirectional linked list comprises all characters on a normal cable sheath, and then generating a template based on the characters in the spliced bidirectional linked list.
On the basis of the above technical solution, the training of the classifier using the template as a training sample specifically includes: and carrying out normalization processing on the size of the template, inputting the template after normalization processing into a classifier as a training sample, and training the classifier by adopting an SVM algorithm.
On the basis of the technical scheme, the classifier compares and identifies the characters on the cable to be detected with the characters in the template, and the specific steps comprise:
scaling the characters on the cable to be detected and the characters in the template to a fixed size;
calculating the distance between each character on the cable to be detected and each character in the template based on a mean square error algorithm;
and establishing a distance matrix, calculating the distance d between each section of characters on the cable to be detected and each section of characters in the template by using a DTW algorithm, and selecting a minimum matching sequence of the sum of all the distances d.
On the basis of the technical scheme, the distance between each section of characters on the cable to be detected and the distance between each section of characters in the template are calculated based on a mean square error algorithm, and the calculation formula is as follows:
wherein, R (x, y) represents the distance between characters, T (x, y) represents the character position in the template, and I (x, y) represents the character position on the cable to be detected.
The invention also provides a cable character detection system based on OCR, comprising:
the acquisition module is used for acquiring characters on a normal cable sheath, and taking each character as one element;
the generating module is used for sequentially connecting all elements into a bidirectional linked list and generating a template based on characters in the bidirectional linked list;
the training module is used for training a classifier by taking the template as a training sample;
and the comparison module is used for inputting the characters on the cable to be detected into the trained classifier, the classifier compares and identifies the characters on the cable to be detected with the characters in the template, if the characters on the cable to be detected can be identified, the characters on the cable to be detected are not defective, otherwise, the characters on the cable to be detected are defective.
On the basis of the technical scheme, the acquiring of the characters on the normal cable sheath takes each character as an element, and the specific steps comprise:
acquiring an image of the surface of a sheath of a normal cable, and binarizing the image;
extracting the region where the character is located in the image based on the mode of obtaining the connected domain;
and acquiring characters in the region, fusing each character according to a set interval, and taking each character as an element.
On the basis of the technical proposal, the device comprises a shell,
the acquired images of the surface of the normal cable sheath are multiple frames, and elements in each frame of image are connected into a two-way linked list;
the template is generated based on the characters in the bidirectional linked list, and the specific steps are as follows: based on the similarity between each section of characters, splicing the bidirectional linked list corresponding to each frame of image, wherein the spliced bidirectional linked list comprises all characters on a normal cable sheath, and then generating a template based on the characters in the spliced bidirectional linked list.
On the basis of the above technical solution, the training of the classifier using the template as a training sample specifically includes: and carrying out normalization processing on the size of the template, inputting the template after normalization processing into a classifier as a training sample, and training the classifier by adopting an SVM algorithm.
Compared with the prior art, the invention has the advantages that: through obtaining the character on the normal cable sheath, connect to be the bidirectional chain table, then generate the template based on the character in the bidirectional chain table, train the classifier as the training sample with the template, and then use the classifier after the training is accomplished to treat the character on the detection line cable and discern, adopt image recognition technology, realize on the cable fast whether there is the detection of defect in the printed character, replace artifical the detection, reduction in production cost.
Drawings
FIG. 1 is a flowchart of a cable character detection method based on OCR according to an embodiment of the present invention;
FIG. 2 is an exemplary illustration of characters on a cable jacket in an embodiment of the present invention;
FIG. 3 is a diagram illustrating string matching in accordance with an embodiment of the present invention;
fig. 4 is a diagram of a DTW matching matrix in an embodiment of the invention.
Detailed Description
The embodiment of the invention provides a cable Character detection method based on OCR (Optical Character Recognition), which is based on an image Recognition technology, can quickly detect whether the printed characters on the cable have defects or not, replaces manual detection, reduces the production cost and improves the production efficiency. The embodiment of the invention also correspondingly provides a cable character detection system based on the OCR.
Referring to fig. 1, an OCR-based cable character detection method provided by an embodiment of the present invention includes:
s1: characters on a normal cable jacket are obtained, and each character is taken as one element. The characters are in English and numeric forms and are used for expressing information such as specification, production and the like of the cable. For the style of characters on the cable sheath, see fig. 2 for example, the characters exist on the sheath in paragraphs, each paragraph of characters includes at least 2 characters, where "2008" is a segment of characters, "FIBERHOME" is a segment of characters, and "CFOA-SM-AS 80-S" is a segment of characters. For the cable with determined model, the character content on the sheath is basically fixed and unchanged, the character on the sheath of the complete cable consists of two parts, namely a fixed character part and a meter part, wherein the fixed character part is used for representing information such as the model and the like, is fixed and unchanged, the fixed character part consists of a plurality of sections of characters, and the meter part changes along with the change of the length of the cable and belongs to a variable part.
S2: all elements are connected in sequence to form a bidirectional linked list, and a template is generated based on characters in the bidirectional linked list.
In the embodiment of the invention, because the cable is in a long strip shape, when the characters on the cable sheath are acquired, all the characters on the sheath cannot be acquired, and only one character can be acquired, so that multiple sections of characters of the sheath need to be acquired during operation, all the acquired characters are spliced by taking elements as units in a duplication elimination mode, and the generated template contains all the character sections on the sheath, namely the character sections on the normal cable.
S3: and training a classifier by taking the template as a training sample. The method comprises the following steps: and normalizing the size of the template, inputting the normalized template serving as a training sample into a classifier, and training the classifier by adopting an SVM (Support Vector Machine) algorithm. In the normalization process, the template size can be normalized to an image with length and width of 32, and is divided into 36 types, wherein 34 types are numbers 0-9 and letters with "I" and "O" removed, and 2 types of special symbols for storing meter number areas are left, namely "and negative examples.
S4: and inputting the characters on the cable to be detected into the trained classifier, comparing and identifying the characters on the cable to be detected with the characters in the template by the classifier, if the characters on the cable to be detected can be identified, indicating that the characters on the cable to be detected are not defective, otherwise, indicating that the characters on the cable to be detected are defective.
The template comprises all complete character segments on the sheath, the classifier compares and identifies the characters on the cable to be detected with the characters in the template, if the characters on the cable to be detected can be identified, the characters on the cable to be detected are shown to be in the template, and the characters on the cable to be detected are shown to be free of defects.
According to the OCR-based cable character detection method, characters on a normal cable sheath are obtained and connected into the two-way chain table, then the template is generated based on the characters in the two-way chain table, the template is used as a training sample to train the classifier, the trained classifier is used for identifying the characters on the cable to be detected, the image recognition technology is adopted, whether the defects exist in the printed characters on the cable or not is rapidly detected, manual detection is replaced, and production cost is reduced.
Optionally, on the basis of the embodiment corresponding to fig. 1, in a first optional embodiment of the OCR-based cable character detection method provided in the embodiment of the present invention, the method obtains characters on a normal cable sheath, and takes each character as an element, and includes the specific steps of:
s101: acquiring an image of the surface of a sheath of a normal cable, and binarizing the image;
s102: extracting the region where the character is located in the image based on the mode of obtaining the connected domain;
s103: and acquiring characters in the region, fusing each character according to a set interval, and taking each character as an element.
The acquired images of the surface of the normal cable sheath are multiple frames, and elements in each frame of image are connected into a doubly linked list.
Generating a template based on the characters in the bidirectional linked list, which comprises the following specific steps: based on the similarity between each section of characters, splicing the bidirectional linked list corresponding to each frame of image, wherein the spliced bidirectional linked list comprises all characters on a normal cable sheath, and then generating a template based on the characters in the spliced bidirectional linked list.
For the concatenation of the character segments between the two-way linked lists, the similarity between the character segments can be calculated by a DTW (dynamic time warping) algorithm, and the specific calculation process, for example, the matching features of the character segments are expressed in a BA manner, for example, the feature of "12 FO" is expressed as B12F0A1A2AFA0Taking the judgment of "TS" and "12F 0" as an example, the string matching method is shown in FIG. 3, and "B" is found by DTW algorithm12ATASAnd B12F0A1A2AFA0"the shortest distance, the calculation mode can be converted into the shortest path algorithm, a distance matrix is established, and the matching result value is shown as fig. 4.
Optionally, on the basis of the first optional embodiment of the cable character detection method based on OCR, in a second optional embodiment of the cable character detection method based on OCR provided in the embodiment of the present invention, the classifier compares and identifies the characters on the cable to be detected with the characters in the template, and the specific steps include:
scaling the characters on the cable to be detected and the characters in the template to a fixed size;
calculating the distance between each character on the cable to be detected and each character in the template based on a mean square error algorithm;
and establishing a distance matrix, calculating the distance d between each section of characters on the cable to be detected and each section of characters in the template by using a DTW algorithm, and selecting a minimum matching sequence of the sum of all the distances d.
Based on a mean square error algorithm, calculating the distance between each section of characters on the cable to be detected and the distance between each section of characters in the template, wherein the calculation formula is as follows:
wherein, R (x, y) represents the distance between characters, T (x, y) represents the character position in the template, and I (x, y) represents the character position on the cable to be detected.
Because it cannot be confirmed whether the character segment of the cable to be detected is missing or not and where the missing position is, it is necessary to arrange the character segment sequence on the cable to be detected and the character segment sequence in the template in a constrained manner, and then extract a sequence o' from all the arrangements L, so that the sequence distance f (d) is the minimum, and the calculation formula is as follows:
wherein L iso'Represents the extracted sequence, and L' represents the standard sequence.
The cable character detection system based on OCR provided by the embodiment of the invention comprises:
the acquisition module is used for acquiring characters on a normal cable sheath, and taking each character as one element;
the generating module is used for sequentially connecting all elements into a bidirectional linked list and generating a template based on characters in the bidirectional linked list;
the training module is used for training a classifier by taking the template as a training sample;
and the comparison module is used for inputting the characters on the cable to be detected into the trained classifier, the classifier compares and identifies the characters on the cable to be detected with the characters in the template, if the characters on the cable to be detected can be identified, the characters on the cable to be detected are not defective, otherwise, the characters on the cable to be detected are defective.
Acquiring characters on a normal cable sheath, and taking each character as an element, wherein the method specifically comprises the following steps:
acquiring an image of the surface of a sheath of a normal cable, and binarizing the image;
extracting the region where the character is located in the image based on the mode of obtaining the connected domain;
and acquiring characters in the region, fusing each character according to a set interval, and taking each character as an element.
The acquired images of the surface of the normal cable sheath are multiple frames, and elements in each frame of image are connected into a two-way linked list;
generating a template based on the characters in the bidirectional linked list, which comprises the following specific steps: based on the similarity between each section of characters, splicing the bidirectional linked list corresponding to each frame of image, wherein the spliced bidirectional linked list comprises all characters on a normal cable sheath, and then generating a template based on the characters in the spliced bidirectional linked list. Training a classifier by taking the template as a training sample, specifically: and carrying out normalization processing on the size of the template, inputting the template after normalization processing into a classifier as a training sample, and training the classifier by adopting an SVM algorithm.
According to the OCR-based cable character detection system, characters on a normal cable sheath are obtained and connected into the two-way chain table, then the template is generated based on the characters in the two-way chain table, the template is used as a training sample to train the classifier, the trained classifier is used for identifying the characters on the cable to be detected, the image recognition technology is adopted, whether the defects exist in the printed characters on the cable or not is rapidly detected, manual detection is replaced, and the production cost is reduced.
The present invention is not limited to the above-described embodiments, and it will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the principle of the present invention, and such modifications and improvements are also considered to be within the scope of the present invention. Those not described in detail in this specification are within the skill of the art.
Claims (8)
1. An OCR-based cable character detection method is characterized by comprising the following steps:
acquiring characters on a normal cable sheath, and taking each character as an element;
all elements are connected in sequence to form a bidirectional linked list, and a template is generated based on characters in the bidirectional linked list;
training a classifier by taking the template as a training sample;
inputting characters on the cable to be detected into a trained classifier, comparing and recognizing the characters on the cable to be detected with the characters in the template by the classifier, if the characters on the cable to be detected can be recognized, indicating that the characters on the cable to be detected are not defective, otherwise, indicating that the characters on the cable to be detected are defective;
the method for acquiring the characters on the normal cable sheath takes each character as one element, and comprises the following specific steps: acquiring an image of the surface of a sheath of a normal cable, and binarizing the image; extracting the region where the character is located in the image based on the mode of obtaining the connected domain; acquiring characters in the region, fusing each character segment according to a set interval, and taking each character segment as an element;
the acquired images of the surface of the normal cable sheath are multiple frames, and elements in each frame of image are connected into a two-way linked list;
the template is generated based on the characters in the bidirectional linked list, and the specific steps are as follows: based on the similarity between each section of characters, splicing the bidirectional linked list corresponding to each frame of image, wherein the spliced bidirectional linked list comprises all characters on a normal cable sheath, and then generating a template based on the characters in the spliced bidirectional linked list.
2. The OCR-based cable character detection method as recited in claim 1, wherein the template is used as a training sample to train a classifier, specifically: and carrying out normalization processing on the size of the template, inputting the template after normalization processing into a classifier as a training sample, and training the classifier by adopting an SVM algorithm.
3. The OCR-based cable character detection method as claimed in claim 1, wherein the classifier compares the characters on the cable to be detected with the characters in the template for recognition, and the specific steps include:
scaling the characters on the cable to be detected and the characters in the template to a fixed size;
calculating the distance between each character on the cable to be detected and each character in the template based on a mean square error algorithm;
and establishing a distance matrix, calculating the distance d between each section of characters on the cable to be detected and each section of characters in the template by using a DTW algorithm, and selecting a minimum matching sequence of the sum of all the distances d.
4. An OCR-based cable character detection method as claimed in claim 3, wherein the mean square error algorithm is used to calculate the distance between each segment of characters on the cable to be detected and the distance between each segment of characters in the template, and the calculation formula is:
wherein, R (x, y) represents the distance between characters, T (x, y) represents the character position in the template, and I (x, y) represents the character position on the cable to be detected.
5. An OCR-based cable character detection system, comprising:
the acquisition module is used for acquiring characters on a normal cable sheath, and taking each character as one element;
the generating module is used for sequentially connecting all elements into a bidirectional linked list and generating a template based on characters in the bidirectional linked list;
the training module is used for training a classifier by taking the template as a training sample;
and the comparison module is used for inputting the characters on the cable to be detected into the trained classifier, the classifier compares and identifies the characters on the cable to be detected with the characters in the template, if the characters on the cable to be detected can be identified, the characters on the cable to be detected are not defective, otherwise, the characters on the cable to be detected are defective.
6. An OCR-based cable character detection system as recited in claim 5 wherein said step of obtaining characters on a normal cable sheath with each character as an element comprises:
acquiring an image of the surface of a sheath of a normal cable, and binarizing the image;
extracting the region where the character is located in the image based on the mode of obtaining the connected domain;
and acquiring characters in the region, fusing each character according to a set interval, and taking each character as an element.
7. An OCR-based cable character detection system as recited in claim 6 wherein:
the acquired images of the surface of the normal cable sheath are multiple frames, and elements in each frame of image are connected into a two-way linked list;
the template is generated based on the characters in the bidirectional linked list, and the specific steps are as follows: based on the similarity between each section of characters, splicing the bidirectional linked list corresponding to each frame of image, wherein the spliced bidirectional linked list comprises all characters on a normal cable sheath, and then generating a template based on the characters in the spliced bidirectional linked list.
8. An OCR-based cable character detection system as recited in claim 5, wherein said template is used as a training sample to train a classifier, specifically: and carrying out normalization processing on the size of the template, inputting the template after normalization processing into a classifier as a training sample, and training the classifier by adopting an SVM algorithm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910473296.1A CN110197181B (en) | 2019-05-31 | 2019-05-31 | Cable character detection method and system based on OCR |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910473296.1A CN110197181B (en) | 2019-05-31 | 2019-05-31 | Cable character detection method and system based on OCR |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110197181A CN110197181A (en) | 2019-09-03 |
CN110197181B true CN110197181B (en) | 2021-04-30 |
Family
ID=67753714
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910473296.1A Active CN110197181B (en) | 2019-05-31 | 2019-05-31 | Cable character detection method and system based on OCR |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110197181B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114780685A (en) * | 2022-04-28 | 2022-07-22 | 贵州电网有限责任公司 | Method for automatically identifying defect information input condition and supplementing defect information through unmanned aerial vehicle |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103488986A (en) * | 2013-09-18 | 2014-01-01 | 西安理工大学 | Method for segmenting and extracting characters in self-adaptation mode |
CN104820986A (en) * | 2015-04-28 | 2015-08-05 | 电子科技大学 | Machine vision-based cable on-line detection method |
CN105260475A (en) * | 2015-10-30 | 2016-01-20 | 努比亚技术有限公司 | Data searching method, data saving method and related equipment |
CN106227668A (en) * | 2016-07-29 | 2016-12-14 | 腾讯科技(深圳)有限公司 | Data processing method and device |
CN108154144A (en) * | 2018-01-12 | 2018-06-12 | 江苏省新通智能交通科技发展有限公司 | A kind of name of vessel character locating method and system based on image |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5759830B2 (en) * | 2011-08-19 | 2015-08-05 | 株式会社日立産機システム | Inkjet recording device |
CN106650721B (en) * | 2016-12-28 | 2019-08-13 | 吴晓军 | A kind of industrial character identifying method based on convolutional neural networks |
CN108596173B (en) * | 2018-04-19 | 2022-02-11 | 长春理工大学 | Single-camera full-view line number real-time recognition device and detection method thereof |
CN109409272B (en) * | 2018-10-17 | 2021-06-18 | 北京空间技术研制试验中心 | Cable acceptance system and method based on machine vision |
CN109712162B (en) * | 2019-01-18 | 2023-03-21 | 珠海博明视觉科技有限公司 | Cable character defect detection method and device based on projection histogram difference |
-
2019
- 2019-05-31 CN CN201910473296.1A patent/CN110197181B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103488986A (en) * | 2013-09-18 | 2014-01-01 | 西安理工大学 | Method for segmenting and extracting characters in self-adaptation mode |
CN104820986A (en) * | 2015-04-28 | 2015-08-05 | 电子科技大学 | Machine vision-based cable on-line detection method |
CN105260475A (en) * | 2015-10-30 | 2016-01-20 | 努比亚技术有限公司 | Data searching method, data saving method and related equipment |
CN106227668A (en) * | 2016-07-29 | 2016-12-14 | 腾讯科技(深圳)有限公司 | Data processing method and device |
CN108154144A (en) * | 2018-01-12 | 2018-06-12 | 江苏省新通智能交通科技发展有限公司 | A kind of name of vessel character locating method and system based on image |
Also Published As
Publication number | Publication date |
---|---|
CN110197181A (en) | 2019-09-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2020155939A1 (en) | Image recognition method and device, storage medium and processor | |
CN109409272B (en) | Cable acceptance system and method based on machine vision | |
CN113344857B (en) | Defect detection network training method, defect detection method and storage medium | |
CN108492291B (en) | CNN segmentation-based solar photovoltaic silicon wafer defect detection system and method | |
CN105930836B (en) | Video character recognition method and device | |
CN109712162B (en) | Cable character defect detection method and device based on projection histogram difference | |
CN105654072A (en) | Automatic character extraction and recognition system and method for low-resolution medical bill image | |
JP2023542460A (en) | Insulator defect detection method and system based on zero sample learning | |
US8345742B2 (en) | Method of processing moving picture and apparatus thereof | |
CN110717492B (en) | Method for correcting direction of character string in drawing based on joint features | |
CN114862845B (en) | Defect detection method, device and equipment for mobile phone touch screen and storage medium | |
CN110135225B (en) | Sample labeling method and computer storage medium | |
CN108427959A (en) | Board state collection method based on image recognition and system | |
CN112381840B (en) | Method and system for marking vehicle appearance parts in loss assessment video | |
CN110197181B (en) | Cable character detection method and system based on OCR | |
CN115019294A (en) | Pointer instrument reading identification method and system | |
JP2002163637A (en) | Device and method for examining image | |
CN111340031A (en) | Equipment almanac target information extraction and identification system based on image identification and method thereof | |
CN116110066A (en) | Information extraction method, device and equipment of bill text and storage medium | |
CN115376139A (en) | Label collecting and analyzing system based on OCR high-speed image recognition | |
US9158968B2 (en) | Apparatus for extracting changed part of image, apparatus for displaying changed part of image, and computer readable medium | |
CN111382703B (en) | Finger vein recognition method based on secondary screening and score fusion | |
CN114373178A (en) | Picture character detection and identification method and system | |
CN114332898A (en) | Automatic correcting method and device for connection test questions and storage medium | |
CN114255464A (en) | Natural scene character detection and identification method based on CRAFT and SCRN-SEED framework |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |