CN107330434A - Electrical symbol recognition methods in a kind of circuit diagram based on PHOG features - Google Patents
Electrical symbol recognition methods in a kind of circuit diagram based on PHOG features Download PDFInfo
- Publication number
- CN107330434A CN107330434A CN201710478828.1A CN201710478828A CN107330434A CN 107330434 A CN107330434 A CN 107330434A CN 201710478828 A CN201710478828 A CN 201710478828A CN 107330434 A CN107330434 A CN 107330434A
- Authority
- CN
- China
- Prior art keywords
- image
- circuit diagram
- electrical symbol
- phog
- electrical
- 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
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
- G06V10/26—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
- 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
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention belongs to image identification technical field, electrical symbol recognition methods in a kind of circuit diagram based on PHOG features is specifically provided;IMAQ and image preprocessing are carried out to circuit diagram image first, then horizontal linear is carried out to circuit diagram image and vertical line is extracted, image segmentation, the electrical symbol obtained in circuit diagram are subsequently carried out to circuit diagram image, the characteristic point of electrical symbol and sample electrical symbol in circuit diagram image is extracted based on PHOG algorithms again, finally utilize the PHOG features of sample electrical symbol, classification based training is carried out to SVM classifier, optimal classification surface is produced;By optimal classification surface, to extract in circuit diagram image the PHOG features of electrical symbol be identified.The present invention exactly can split electrical symbol from circuit diagram, obtain electrical symbol image, so as to realize that electrical symbol is recognized;Effectively overcome the weaker shortcoming of environmental disturbances factor ability.
Description
Technical field
The invention belongs to image identification technical field, it is related to electrical symbol recognition methods in circuit diagram, it is specific to provide a kind of
Electrical symbol recognition methods in circuit diagram based on P HOG features.
Background technology
In actual applications, engineer and architect often describe circuit, machinery zero using different graphical symbols
Part, building etc., these drawings generally require to be converted to electronic form for effectively being stored, and retrieve and transmit, Yi Jigeng
Newly, and combine and generate new drawing.With the rapid development of electronic technology, the electrical symbol in drawing recognizes problem by pole
Big concern;Electrical symbol is detected and its accuracy and rapidity of positioning can directly influence the correctness and totality of drawing judgement
The understanding of mentality of designing.But it is due to be influenceed by symbol size, the anglec of rotation especially complex background environment so that target is accorded with
Number there is many interference, solve problems and be faced with many difficulties, so as computer vision and electrical design in recent years
The focus of area research.
At present, electrical symbol identification generally uses the method based on statistical classification, and wherein key is to need to extract electrically symbol
Number feature, then Classification and Identification is carried out using the method for machine learning.Compare representative algorithm:SIFT algorithms, in space
Extreme point is found in yardstick to a sub-picture, and extracts description such as its position, yardstick, rotational invariants and obtains feature and goes forward side by side
Row Feature Points Matching, for detecting and describing the local feature in image;Shape context algorithms, based on contour of object sample
What this point was described.These algorithms are calculated on some intensive, unified space cells, and in order to improve performance, all
Overlapping local progress pixel contrast standardization is wanted, overcomes environmental disturbances factor ability weaker.
The content of the invention
It is an object of the invention to for being based on PHOG features there is provided one kind in place of above shortcomings in the prior art
Circuit diagram in electrical symbol recognition methods;Electrical symbol can be effectively recognized, environmental disturbances factor is overcome.In order to realize this
Purpose, the present invention is as follows using technical scheme:
Electrical symbol recognition methods in a kind of circuit diagram based on PHOG features, comprises the following steps;
Step 1, IMAQ and image preprocessing are carried out to circuit diagram image;
Step 2, horizontal linear and vertical line are carried out to circuit diagram image extract;
Step 3, the electrical symbol carried out to circuit diagram image in image segmentation, acquisition circuit diagram;
Step 4, based on PHOG algorithms, extract electrical symbol and sample electrical symbol (previously known class in circuit diagram image
Characteristic point not);
Step 5, the PHOG features using sample electrical symbol, to SVM (Support Vector Machine) grader
Classification based training is carried out, optimal classification surface is produced;By optimal classification surface, to extracting to obtain the PHOG of electrical symbol in circuit diagram image
Feature is identified.
Further, step 3 specifically includes following steps:
Step 3.1, step 1 is pre-processed after image subtract step 2 horizontal linear image and vertical line image, obtain
The image of electrical symbol;
Step 3.2, the image to electrical symbol are repaired using closing operation of mathematical morphology, and structural element SE selection is used
Equation below:
SE=strel (' disk', R)
Wherein, strel (' disk', R) is the structural element expression formula of closed operation, ' disk' represents that closed operation middle finger shapes
Shape, i.e. circle, R represent circular configuration element radius, take 0.5w, w to be linear width;
Step 3.3, the electrical symbol for checking defect in image after being repaired through step 3.2, judge its defect part, if defect
Part obtains any bar straight-line segment in horizontal linear or vertical line image corresponding to step 2, then assert the straight-line segment
For a part for the electrical symbol, image is added after the straight-line segment image is repaired with step 3.2, is repaired again
Image afterwards, finally removes the tie point on image;
Image is carried out and computing after step 3.4, image and step 1 that step 3.3 is obtained are pre-processed, after being split
Electrical symbol image.
Further, image preprocessing described in step 1 includes the character in image binaryzation, denoising and circuit diagram successively
Remove;Wherein, image binaryzation uses Otsu algorithm;Denoising calculates the area of connected domain using eight neighborhood searching algorithm, if even
The area in logical domain is less than 8 pixels, then regards as noise and remove;Character in circuit diagram, which is removed, uses Global thresholding from figure
Text is removed as in.
Further, horizontal linear described in step 2 and the detailed process of vertical line extraction are:
Using 10 pixels as step-length, scan mode from top to bottom, from left to right is taken to be scanned circuit diagram image,
Scanning result is represented with histogram, the value on histogram summit is taken as linear width w;Wherein, for each line width
Degree, by calculating continuous white to black, black being obtained to the distance between two turning points in vain;
Horizontal line section or vertical segment are extracted using morphology opening operation, structural element SE selection uses equation below:
SE=strel (' line', LEN, DEG)
Wherein, strel (' line', LEN, DEG) is the structural element expression formula of opening operation, ' line' represented in opening operation
Designated shape, i.e. straight line;LEN represents straight length, is chosen for 5w;DEG represents angle, is 0 or 90, when DEG is 0, obtains
Horizontal linear section, when DEG is 90, obtains vertical line section.
Further, svm classifier training is carried out using LIBSVM tool boxes in step 5, in svm classifier training, using straight
Side's figure, which intersects core (Histogram intersection Kernel), is classified.
Technical scheme advantage of the present invention is:Using the present invention electrical symbol image-recognizing method can exactly from
Electrical symbol is split in circuit diagram, electrical symbol image is obtained, so as to provide favourable support for electrical symbol identification.
The present invention carries out symbol using electrical symbol PHOG features, the shortcoming for overcoming environmental disturbances factor ability weaker is extracted with SVM
Know method for distinguishing, a kind of new method is provided for the vector quantization of electrical power engineering drawings, carried out without the pixel on the whole to drawing
Tracking fitting, makes vector quantization process simplification.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of electrical symbol recognition methods in the circuit diagram based on PHOG features in embodiment.
Fig. 2 is the schematic diagram of circuit diagram in embodiment.
Fig. 3 is the circuit diagram image schematic diagram after image preprocessing in embodiment.
Fig. 4 is the schematic diagram that embodiment cathetus is extracted, wherein, (a) represents that horizontal linear is extracted, and (b) represents vertical straight
Line drawing.
Fig. 5 is electrical symbol schematic diagram after being extracted in embodiment.
Embodiment
With reference to embodiment and accompanying drawing, the present invention is described in further detail.Those skilled in the art should manage
Solution, below specifically described content be illustrative and be not restrictive, should not be to limit the scope of the invention.
Electrical symbol recognition methods in a kind of circuit diagram based on PHOG features is provided in the present embodiment, its idiographic flow is such as
Shown in Fig. 1;It is used for extracting the electrical symbol in circuit diagram first by the dividing method of morphology operations, wherein for structural elements
The selection of element and threshold value is that the statistical analysis based on graphic assembly spatial form is obtained;Secondly, PHOG is extracted to electrical symbol
Feature;Finally, combination supporting vector machine training grader (SVM) realizes the identification to electrical symbol.Specifically include following steps:
Step 1, using camera carry out circuit diagram IMAQ, as shown in Fig. 2 and being pre-processed to image:To figure
As carrying out binary conversion treatment using Otsu algorithm (Otsu), due to the image after binary conversion treatment, have some noises and go out
It is existing;For the noise removal in image, the area of connected domain is calculated using eight neighborhood searching algorithm, if the face of some connected domains
Product is less than 8 pixels, is regarded as noise and is removed;In circuit diagram, character and figure are all present, so in order to electricity
Gas symbol is reasonably split, and it is necessary that the character in circuit diagram, which is removed, is removed for the character in circuit diagram,
Text is removed from image using Global thresholding;Picture after image preprocessing is as shown in Figure 3.
Step 2, horizontal linear and vertical line are extracted:By taking from top to bottom, scan mode from left to right is to figure
As being scanned, for calculating the width of image middle conductor;And in order to reduce counting loss, using the step-length of 10 pixels
Image is scanned;For the length of each scan line, by calculating continuous white to black, black to two white turning points
The distance between and obtain;The result of scanning is represented using histogram, takes the value on histogram summit as linear width w;
Horizontal line section or vertical segment are extracted using morphology opening operation, structural element SE selection uses equation below:
SE=strel (' line', LEN, DEG)
Wherein, strel (' line', LEN, DEG) is the structural element expression formula of opening operation, ' line' represented in opening operation
Designated shape, i.e. straight line;LEN represents straight length, is chosen for 5w;DEG represents angle, is 0 or 90, when DEG is 0, obtains
Horizontal linear section, shown in such as Fig. 4 (a);When DEG is 90, vertical line section is obtained as shown in Fig. 4 (b).
Electrical symbol is extracted in step 3, circuit diagram:Character in step 1 is removed the level obtained in image and step 2 straight
Line image, vertical line image are subtracted each other, and obtain the image of electrical symbol;Then, because some electrical symbols are by level
With vertical segment composition, so the removal of line segment may delete some parts of these symbols, the removal of these line segments may
Electrical symbol can be divided into several parts, so to be repaired to image using closing operation of mathematical morphology, wherein structural element
SE selection, using a flat circular configuration element (disk), radius R size is more than 0.5*w;Some inner spaces compared with
Big electrical symbol is bigger than other symbol areas, is come so being easy to be divided, so to after closing operation of mathematical morphology
Image carries out the reparation of endless integral symbol, and the tie point on image is removed again afterwards;Finally obtained image and step
Rapid 1 character removes image and carried out and computing, the electrical symbol image after being split;Electrical symbol picture such as Fig. 5 institutes of extraction
Show.
Step 4, extraction PHOG features:The PHOG features of electrical symbol and sample electrical symbol after segmentation are extracted, it has
Body method is:(1) coloured image is converted into gray level image;(2) using the marginal information of Canny operator extraction images, afterwards
PHOG features will be extracted in these edges;(3) will be image layered, first layer (being designated as L=0) is entire image, second
Layer (being designated as L=1) is to divide equally entire image (first layer) progress four, and third layer (being designated as L=2) is will be each in the second layer
Block subregion carries out four and divided equally, and the rest may be inferred, herein using 4 Rotating fields;Layered method HOG features, in each layer, statistics are each
Histogram of gradients feature and in series image feature under this layer of the block region on K direction, chooses K=8 herein;
(4) HOG feature of the image under each layer is subjected to series connection merging, it is 680 dimensions to constitute final PHOG intrinsic dimensionalities.
Step 5, electrical symbol identification:Electrical symbol training set is set up, the PHOG features of electrical symbol image are extracted;Finally
SVMs is trained using these PHOG features and classification information, using SVMs to obtained electrical symbol
Carry out Classification and Identification;In svm classifier training is carried out, using histogram intersection core (Histogram intersection
Kernel) classified;Histogram intersection core is also known as Pyramid match kernel, and the histogram intersection core is that one kind is based on
The kernel function of implicit corresponding relation, the problem of solving the identification and classification of unordered, variable-length set of vectors.
The foregoing is only a specific embodiment of the invention, any feature disclosed in this specification, except non-specifically
Narration, can alternative features equivalent by other or with similar purpose replaced;Disclosed all features or all sides
Method or during the step of, in addition to mutually exclusive feature and/or step, can be combined in any way.
Claims (5)
1. electrical symbol recognition methods in a kind of circuit diagram based on PHOG features, comprises the following steps;
Step 1, IMAQ and image preprocessing are carried out to circuit diagram image;
Step 2, horizontal linear and vertical line are carried out to circuit diagram image extract;
Step 3, the electrical symbol carried out to circuit diagram image in image segmentation, acquisition circuit diagram;
Step 4, based on PHOG algorithms, extract the characteristic point of electrical symbol and sample electrical symbol in circuit diagram image;
Step 5, the PHOG features using sample electrical symbol, classification based training is carried out to SVM classifier, produces optimal classification surface;
By optimal classification surface, to extract in circuit diagram image the PHOG features of electrical symbol be identified.
2. electrical symbol recognition methods in the circuit diagram based on PHOG features as described in claim 1, it is characterised in that step 3
Specifically include following steps:
Step 3.1, step 1 is pre-processed after image subtract step 2 horizontal linear image and vertical line image, obtain electrically
The image of symbol;
Step 3.2, the image to electrical symbol are repaired using closing operation of mathematical morphology, and structural element SE selection is using as follows
Formula:
SE=strel (' disk', R)
Wherein, strel (' disk', R) is the structural element expression formula of closed operation, ' disk' represent designated shape in closed operation,
I.e. circular, R represents circular configuration element radius, takes 0.5w, w to be linear width;
Step 3.3, the electrical symbol for checking defect in image after being repaired through step 3.2, judge its defect part, if defect part
Any bar straight-line segment in horizontal linear or vertical line image is obtained corresponding to step 2, then assert the straight-line segment to be somebody's turn to do
A part for electrical symbol, image is added after the straight-line segment image is repaired with step 3.2, after being repaired again
Image, finally removes the tie point on image;
Image is carried out and computing after step 3.4, image and step 1 that step 3.3 is obtained are pre-processed, electric after being split
Glyph image.
3. electrical symbol recognition methods in the circuit diagram based on PHOG features as described in claim 1, it is characterised in that step 1
Described in image preprocessing successively include image binaryzation, denoising and circuit diagram in character remove;Wherein, image binaryzation is adopted
Use Otsu algorithm;Denoising calculates the area of connected domain using eight neighborhood searching algorithm, if the area of connected domain is less than 8 pixels,
Then regard as noise and remove;Character in circuit diagram is removed removes text using Global thresholding from image.
4. electrical symbol recognition methods in the circuit diagram based on PHOG features as described in claim 1, it is characterised in that step 2
The horizontal linear and the detailed process of vertical line extraction are:
Using 10 pixels as step-length, take scan mode from top to bottom, to be from left to right scanned circuit diagram image, will sweep
Retouch result to be represented with histogram, take the value on histogram summit as linear width w;Wherein, for each scanning line width, lead to
Cross and calculate continuous white to black, black obtained to the distance between two white turning points;
Horizontal line section or vertical segment are extracted using morphology opening operation, structural element SE selection uses equation below:
SE=strel (' line', LEN, DEG)
Wherein, strel (' line', LEN, DEG) is the structural element expression formula of opening operation, ' line' represents to specify in opening operation
Shape, i.e. straight line;LEN represents straight length, is chosen for 5w;DEG represents angle, is 0 or 90, when DEG is 0, obtains level
Straightway, when DEG is 90, obtains vertical line section.
5. electrical symbol recognition methods in the circuit diagram based on PHOG features as described in claim 1, it is characterised in that step 5
Middle use LIBSVM tool boxes carry out svm classifier training, in svm classifier training, are classified using histogram intersection core.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710478828.1A CN107330434A (en) | 2017-06-22 | 2017-06-22 | Electrical symbol recognition methods in a kind of circuit diagram based on PHOG features |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710478828.1A CN107330434A (en) | 2017-06-22 | 2017-06-22 | Electrical symbol recognition methods in a kind of circuit diagram based on PHOG features |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107330434A true CN107330434A (en) | 2017-11-07 |
Family
ID=60194563
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710478828.1A Pending CN107330434A (en) | 2017-06-22 | 2017-06-22 | Electrical symbol recognition methods in a kind of circuit diagram based on PHOG features |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107330434A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111797838A (en) * | 2019-04-08 | 2020-10-20 | 上海怀若智能科技有限公司 | Blind denoising system, method and device for picture documents |
CN112528845A (en) * | 2020-12-11 | 2021-03-19 | 华中师范大学 | Physical circuit diagram identification method based on deep learning and application thereof |
CN113158999A (en) * | 2021-05-26 | 2021-07-23 | 南京云阶电力科技有限公司 | Method and device for identifying terminal jumper in electrical design drawing based on template matching |
CN113688829A (en) * | 2021-08-05 | 2021-11-23 | 南京国电南自电网自动化有限公司 | Automatic transformer substation monitoring picture identification method and system |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106650696A (en) * | 2016-12-30 | 2017-05-10 | 山东大学 | Handwritten electrical element identification method based on singular value decomposition |
-
2017
- 2017-06-22 CN CN201710478828.1A patent/CN107330434A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106650696A (en) * | 2016-12-30 | 2017-05-10 | 山东大学 | Handwritten electrical element identification method based on singular value decomposition |
Non-Patent Citations (3)
Title |
---|
S.CHOWDHURY等: "Segmentation of Text and Graphics from Document Images", 《NINTH INTERNATIONAL CONFERENCE ON DOCUMENT ANALYSIS AND RECOGNITION (ICDAR 2007)》 * |
李宏等: "一个具有自动输入功能的电路CAD系统", 《造船技术》 * |
肖豆等: "基于PHOG特征的电路图电气符号识别", 《舰船电子工程》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111797838A (en) * | 2019-04-08 | 2020-10-20 | 上海怀若智能科技有限公司 | Blind denoising system, method and device for picture documents |
CN112528845A (en) * | 2020-12-11 | 2021-03-19 | 华中师范大学 | Physical circuit diagram identification method based on deep learning and application thereof |
CN112528845B (en) * | 2020-12-11 | 2022-09-20 | 华中师范大学 | Physical circuit diagram identification method based on deep learning and application thereof |
CN113158999A (en) * | 2021-05-26 | 2021-07-23 | 南京云阶电力科技有限公司 | Method and device for identifying terminal jumper in electrical design drawing based on template matching |
CN113158999B (en) * | 2021-05-26 | 2024-04-02 | 南京云阶电力科技有限公司 | Terminal jumper wire identification method and device in electrical design drawing based on template matching |
CN113688829A (en) * | 2021-08-05 | 2021-11-23 | 南京国电南自电网自动化有限公司 | Automatic transformer substation monitoring picture identification method and system |
CN113688829B (en) * | 2021-08-05 | 2024-02-20 | 南京国电南自电网自动化有限公司 | Automatic identification method and system for monitoring picture of transformer substation |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104751142B (en) | A kind of natural scene Method for text detection based on stroke feature | |
Peng et al. | A triple-thresholds pavement crack detection method leveraging random structured forest | |
CN110838126B (en) | Cell image segmentation method, cell image segmentation device, computer equipment and storage medium | |
CN101561866B (en) | Character recognition method based on SIFT feature and gray scale difference value histogram feature | |
CN107330434A (en) | Electrical symbol recognition methods in a kind of circuit diagram based on PHOG features | |
CN108038481A (en) | A kind of combination maximum extreme value stability region and the text positioning method of stroke width change | |
CN111539330B (en) | Transformer substation digital display instrument identification method based on double-SVM multi-classifier | |
CN104680550A (en) | Method for detecting defect on surface of bearing by image feature points | |
CN110751619A (en) | Insulator defect detection method | |
CN102750531B (en) | Method for detecting handwriting mark symbols for bill document positioning grids | |
CN107016394B (en) | Cross fiber feature point matching method | |
CN103310211A (en) | Filling mark recognition method based on image processing | |
Garz et al. | A binarization-free clustering approach to segment curved text lines in historical manuscripts | |
Liang et al. | An extraction and classification algorithm for concrete cracks based on machine vision | |
CN112818952A (en) | Coal rock boundary recognition method and device and electronic equipment | |
CN113033558A (en) | Text detection method and device for natural scene and storage medium | |
CN115984186A (en) | Fine product image anomaly detection method based on multi-resolution knowledge extraction | |
Maddouri et al. | Text lines and PAWs segmentation of handwritten Arabic document by two hybrid methods | |
CN105354547A (en) | Pedestrian detection method in combination of texture and color features | |
CN109271882B (en) | Method for extracting color-distinguished handwritten Chinese characters | |
Chen et al. | A novel Fourier descriptor based image alignment algorithm for automatic optical inspection | |
Giri | Text information extraction and analysis from images using digital image processing techniques | |
Chang | Intelligent text detection and extraction from natural scene images | |
CN112418210B (en) | Intelligent classification method for tower inspection information | |
CN104036494A (en) | Fast matching computation method used for fruit picture |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171107 |
|
RJ01 | Rejection of invention patent application after publication |