CN112256906A - Method, device and storage medium for marking annotation on display screen - Google Patents
Method, device and storage medium for marking annotation on display screen Download PDFInfo
- Publication number
- CN112256906A CN112256906A CN202011144962.6A CN202011144962A CN112256906A CN 112256906 A CN112256906 A CN 112256906A CN 202011144962 A CN202011144962 A CN 202011144962A CN 112256906 A CN112256906 A CN 112256906A
- Authority
- CN
- China
- Prior art keywords
- information
- module
- article
- detection frame
- display screen
- 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
- 238000000034 method Methods 0.000 title claims abstract description 20
- 239000013598 vector Substances 0.000 claims abstract description 78
- 238000001514 detection method Methods 0.000 claims abstract description 71
- 238000007781 pre-processing Methods 0.000 claims abstract description 13
- 238000012216 screening Methods 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 230000004931 aggregating effect Effects 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 230000002776 aggregation Effects 0.000 description 3
- 238000004220 aggregation Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/5866—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/54—Browsing; Visualisation therefor
-
- 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
- G06V10/255—Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
-
- 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
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Library & Information Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method for marking annotations on a display screen, which comprises the following steps: receiving a marking annotation instruction, and acquiring graphical information of annotation to be marked; preprocessing a picture, dividing the picture into a plurality of target candidate areas, extracting a feature vector of an article contour in each target candidate area, and generating a feature map; according to the feature map, eliminating candidate areas without object contour feature vectors; generating detection frame information according to the article contour feature vector, and correcting the appearance of the detection frame; matching and comparing the characteristic vector inside each detection frame with the characteristic vector in the database, and finally determining the article information; the method and the device have the advantages that the position and the information of the article are rapidly identified, and the mark and the annotation are carried out on the display screen, so that security personnel can conveniently and rapidly know the type and the information of the article.
Description
Technical Field
The invention relates to the technical field of autonomous graph marking, in particular to a method and a device for marking annotations on a display screen and a storage medium.
Background
Along with the development of national economy, more and more people choose to take public transport means to go on a journey, in order to guarantee the safety of people's trip, security installations often need to be installed in areas such as subway station, and the most commonly used security installations include the security scanner, and the security scanner is equipped with the display screen that shows scanning result, and the luggage of examining the scanner to the passenger is scanned to on sending the scanning result to the display screen, supply the security inspector to discern.
The display screen of the existing security check instrument can only display scanned pictures, and depends on security check personnel to artificially judge whether forbidden articles exist, the types of the articles are all distinguished by the security check personnel, particularly in the passenger flow peak, the security check speed is high, the retention time of images on the display screen is greatly reduced, and the security check personnel can not timely and effectively identify and distinguish the articles.
Disclosure of Invention
The present invention is directed to a method, an apparatus and a storage medium for marking annotations on a display screen, so as to solve the problems set forth in the above background art.
In order to achieve the purpose, the invention provides the following technical scheme:
a method of marking annotations on a display screen, comprising the steps of:
s1, receiving a marking annotation instruction and acquiring the graphical information of the annotation to be marked;
s2, preprocessing the picture, dividing the picture into a plurality of target candidate areas, extracting the feature vector of the outline of the article in each target candidate area, and generating a feature map;
s3, removing candidate areas without article contour feature vectors according to the feature map;
s4, generating detection frame information according to the contour feature vector of the article, and correcting the shape of the detection frame;
s5, matching and comparing the characteristic vector in each detection frame with the characteristic vector in the database, and finally determining the information of the article;
and S6, sending the detection frame information and the article information to a display terminal, wherein the display terminal displays the shape of the detection frame on a display screen and displays the article information in the detection frame.
Preferably, the step S2 includes the following steps:
s201, dividing the picture into a plurality of small areas through a simple area division algorithm, and continuously aggregating adjacent small areas through similarity and area size to form a plurality of candidate areas;
s202, extracting the feature vectors of all the areas by using a shared convolution algorithm module;
and S203, superposing and integrating the feature vectors of all dimensions into a feature map.
Preferably, in step S3, the trained binary classification unit is used to eliminate candidate regions without the item contour feature vector.
Preferably, in step S4, the detection box information is generated after the item contour feature vector is input to the area network generation module.
Preferably, the step S5 includes the steps of:
s501, comparing the similarity of the feature vector in each detection frame with the feature vector in a database, and determining the similarity;
s502, judging whether the acquaintance is larger than a threshold value of the comparison article, and if so, judging the type and the information of the article;
and S503, determining and outputting the information of the articles in the detection frame.
The invention also provides a device for marking annotations on a display screen, which comprises:
the receiving and acquiring module is used for receiving the marking annotation instruction and the picture to be marked and annotated;
the preprocessing module is used for dividing the picture into a plurality of target candidate areas, then extracting the feature vector of the outline of the article in each target candidate area and generating a feature map;
the screening module is used for reading the generated feature map and rejecting a candidate area without the object contour feature vector;
the detection frame module is used for generating a detection frame around the article image according to the article contour feature vector, correcting and adjusting the size of the detection frame and accurately positioning the detection frame;
the classified information determining module is used for matching and comparing the characteristic vector inside the detection frame with the characteristic vector in the database, and determining and outputting the information of the articles inside the detection frame;
and the data sending module reads the information of the detection frame and the information of the articles in the detection frame and sends the information to the display terminal, and the information is displayed on the display screen through the display terminal.
Preferably, the preprocessing module comprises:
the region dividing module is used for dividing the picture into a plurality of small regions and then aggregating adjacent small regions continuously through similarity and region size to form a plurality of candidate regions;
the characteristic vector extraction module is used for extracting the characteristic vectors of all the regions by using a shared convolution algorithm;
and the feature map generation module is used for superposing the feature vectors of all dimensions and integrating the feature vectors into a feature map for outputting.
Preferably, an SVM module is arranged in the screening module, the SVM module is trained by inputting positive and negative samples, the trained SVM module is used as a two-classifier to screen and reject candidate regions, a region network generation module is arranged in the detection frame module, and the network generation module forms a detection frame around the contour of the article through the feature vector of the contour of the article.
Preferably, the classification information determining module includes:
the comparison module is used for comparing the similarity between the characteristic vector in the detection frame and the characteristic vector in the database and outputting the similarity;
the judging module is used for comparing the similarity with a set threshold value, and judging the type and the information of the article if the similarity is greater than the set threshold value;
and the output module is used for reading the information of the article from the database and outputting the information after determining the information of the article in the detection frame.
The present invention additionally provides a storage medium for marking annotations on a display screen, the storage medium having stored thereon a computer program which, when executed by a processor, is capable of implementing the steps of the method for marking annotations on a display screen according to any one of the claims.
Compared with the prior art, the invention has the beneficial effects that:
(1) the position of an article in the image is rapidly determined and marked on a display screen by using a detection frame through feature extraction of the image collected by the security inspection machine, so that a security inspector can rapidly observe and identify the position of the article;
(1) the inside article characteristic of detection frame is contrasted, the kind and the information of article are discerned fast to annotate in the detection frame of mark on the display screen, let the security check personnel know the kind of article fast, thereby judge rapidly whether for forbidden article.
Drawings
FIG. 1 is a flow chart of an embodiment of the present invention 1;
FIG. 2 is a flowchart illustrating an implementation of step S2 in example 1 of the present invention;
FIG. 3 is a flowchart illustrating an implementation of step S5 in embodiment 1 of the present invention;
FIG. 4 is a schematic structural view of example 2 of the present invention;
FIG. 5 is a schematic structural diagram of a preprocessing module in embodiment 2 of the present invention;
fig. 6 is a schematic structural diagram of a classification information determination module according to embodiment 2 of the present invention;
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1:
referring to fig. 1-3, a method for marking annotations on a display screen includes the following steps:
and S1, receiving the marking annotation command and acquiring the to-be-marked annotation graphic information.
The marking instruction is sent by a security check worker, the security check worker is determined to start the method when the marking instruction is sent, and the acquired graphic information can be acquired by a security check machine during real-time security check or can be acquired from an image library storing previous images.
And S2, preprocessing the picture, dividing the picture into a plurality of target candidate areas, extracting the feature vector of the outline of the article in each target candidate area, and generating a feature map.
Specifically, the step S2 mainly includes the following steps of firstly, dividing the picture into a plurality of small regions by a simple region division algorithm, and then continuously aggregating adjacent small regions by similarity and region size to form a plurality of candidate regions, wherein when aggregating adjacent small regions, the small regions are aggregated first, so that incomplete hierarchical relationship caused by continuous aggregation of small regions by the large regions is prevented, and when aggregating regions, a greedy algorithm is adopted for aggregation of regions; secondly, extracting the feature vectors of each region by using a shared convolution algorithm module, wherein a complete convolution integral network runs in the shared convolution algorithm module, the feature vector of each region is extracted by using the convolution neural network during extraction, and then the corresponding feature of each region is extracted from the feature of the whole image; and then, superposing and integrating the feature vectors of all dimensions into a feature map, namely superposing the feature vectors of different dimensions in the same region, and dividing the feature vectors by the region.
And S3, removing candidate areas without the article contour feature vectors according to the feature map.
The rejecting tool adopts a trained two-classification unit, the two-classification unit only outputs two types of results, the characteristic vectors are input, the output is a class score, positive and negative sample input is adopted during training, and a screening threshold value is set according to the class score output during training, so that during actual work, a candidate region without the object contour characteristic vectors is judged and screened according to comparison with the set threshold value, the calculation amount of subsequent steps is reduced, and the calculation speed is improved.
And S4, generating detection frame information according to the article contour feature vector, and correcting the shape of the detection frame.
The detection frame information is generated by a regional network generation module, a regional generation network is arranged in the regional network generation module, the network input is a characteristic diagram, the output is a detection frame, the inside of the frame is an article, and the outside of the frame is a background.
And S5, matching and comparing the characteristic vector in each detection frame with the characteristic vector in the database, and finally determining the article information.
Specifically, the step S5 mainly includes the following steps: firstly, similarity comparison is carried out on a characteristic vector in each detection frame and a characteristic vector in a database, the database is established in advance, various characteristic information such as outlines, sizes and geometries of various articles are stored, information such as names and danger levels is also stored, the vectors with different dimensionalities are adopted for one-to-one corresponding comparison in the comparison process, cross comparison is avoided, comparison tools are compared through a preset network, the comparison accuracy and efficiency are improved, and the similarity is determined; secondly, judging whether the degree of identity is greater than a threshold value of a comparison article, if so, judging the type and information of the article, namely judging that the identified article is the article to be compared currently, and if not, continuing to compare the degree of identity with the next comparison article; and finally, determining and outputting the information of the articles in the detection frame, wherein the information of the articles in the detection frame is directly called out from the database during output.
And S6, sending the detection frame information and the article information to a display terminal, wherein the display terminal displays the shape of the detection frame on a display screen and displays the article information in the detection frame.
Wherein after display terminal receives detection frame information and article information, handle into the display format of display screen, directly superpose with original image on the display screen, article position in the quick definite image and use the detection frame to mark on the display screen, let security check personnel observe the position of discerning the article fast, compare the inside article characteristic of detection frame, the kind and the information of article are discerned fast, and annotate in the detection frame of mark on the display screen, let security check personnel know the kind of article fast, thereby judge rapidly whether for forbidden article.
Example 2:
referring to fig. 4-6, a device for marking annotations on a display screen includes a receiving and acquiring module, a preprocessing module, a screening module, a detection frame module, a classification information determining module and a data sending module, wherein the receiving and acquiring module receives an annotation marking instruction of a security inspector and receives a picture to be marked, the picture can be obtained from an image acquired by a security inspection machine during real-time security inspection or an image library storing past images, the receiving and acquiring module sends the received picture to the preprocessing module, the preprocessing module includes a region dividing module, a feature vector extracting module and a feature map generating module, the region dividing module divides the picture into a plurality of small regions, and then continuously aggregates adjacent small regions through similarity and region size to form a plurality of candidate regions, the region dividing module aggregates the small regions first, the method comprises the steps of preventing small areas from being continuously aggregated in a large area to cause incomplete hierarchical relation, performing area aggregation by adopting a greedy algorithm when the small areas are aggregated, extracting feature vectors of a whole graph by using a convolution integral network by using a feature vector extraction module, extracting features corresponding to each area from the features of the whole graph, integrating the feature vectors of all dimensions into a feature map by a feature map generation module in an overlapping mode, namely overlapping the feature vectors of different dimensions in the same area to classify the feature vectors by the areas, and outputting a feature vector map of a sub-area by a preprocessing module and sending the map to a screening module.
The system comprises a screening module, a classification information determining module, a SVM module and a network generation module, wherein the screening module is arranged in the screening module, the SVM module is trained by inputting positive and negative samples, the trained SVM module is used as a two-classifier to screen and remove candidate regions, the SVM module sets a screening threshold according to class scores output by training, the candidate regions without object contour feature vectors are screened according to comparison with the set threshold, the calculated amount of subsequent steps is reduced, the calculating speed is improved, the screening module sends a screened feature vector map to a detection frame module, the detection frame module internally comprises a region network generation module, the network generation module forms a detection frame around the object contour through the object contour feature vectors, and then processed data are sent to the classification information determining module.
The classification information determining module comprises a comparison module, a judging module and an output module, wherein a database is arranged in the comparison module, the database stores various characteristic information of various articles such as outlines, sizes, geometries and the like, and also stores information such as names, danger levels and the like, the comparison module compares the similarity between a characteristic vector in the detection frame and a characteristic vector in the database and outputs the similarity, the judging module compares the similarity with a set threshold, if the similarity is larger than the set threshold, the type and the information of the article are judged, after the output module determines the information of the article in the detection frame, the information of the article is read from the database and is output to the data sending module, the data sending module receives the information of the read detection frame and the information of the article in the detection frame, processes the information and sends the information to the display terminal, and the original image is superposed on the display screen for displaying, and the article information on the display screen is marked and annotated.
Example 3:
a storage medium for marking annotations on a display screen, the storage medium having stored thereon a computer program, which when executed by a processor is capable of implementing the steps of the method for marking annotations on a display screen according to any one of the claims, wherein the storage medium may be a usb disk, a removable hard disk, an optical disk, or any other storage device known in the art for storing computer program codes.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (10)
1. A method of marking annotations on a display screen, comprising the steps of;
s1, receiving a marking annotation instruction and acquiring the graphical information of the annotation to be marked;
s2, preprocessing the picture, dividing the picture into a plurality of target candidate areas, extracting the feature vector of the outline of the article in each target candidate area, and generating a feature map;
s3, removing candidate areas without article contour feature vectors according to the feature map;
s4, generating detection frame information according to the contour feature vector of the article, and correcting the shape of the detection frame;
s5, matching and comparing the characteristic vector in each detection frame with the characteristic vector in the database, and finally determining the information of the article;
and S6, sending the detection frame information and the article information to a display terminal, wherein the display terminal displays the shape of the detection frame on a display screen and displays the article information in the detection frame.
2. A method of marking annotations on a display screen according to claim 1, characterized in that: the step S2 includes the following steps:
s201, dividing the picture into a plurality of small areas through a simple area division algorithm, and continuously aggregating adjacent small areas through similarity and area size to form a plurality of candidate areas;
s202, extracting the feature vectors of all the areas by using a shared convolution algorithm module;
and S203, superposing and integrating the feature vectors of all dimensions into a feature map.
3. A method of marking annotations on a display screen according to claim 1, characterized in that: in the step S3, a trained binary classification unit is used to eliminate candidate regions without the item contour feature vector.
4. A method of marking annotations on a display screen according to claim 1, characterized in that: in step S4, the detection box information is generated after the item contour feature vector is input to the area network generation module.
5. A method of marking annotations on a display screen according to claim 1, characterized in that: the step S5 includes the steps of:
s501, comparing the similarity of the feature vector in each detection frame with the feature vector in a database, and determining the similarity;
s502, judging whether the acquaintance is larger than a threshold value of the comparison article, and if so, judging the type and the information of the article;
and S503, determining and outputting the information of the articles in the detection frame.
6. An apparatus for marking annotations on a display screen, characterized in that: comprises the following steps:
the receiving and acquiring module is used for receiving the marking annotation instruction and the picture to be marked and annotated;
the preprocessing module is used for dividing the picture into a plurality of target candidate areas, then extracting the feature vector of the outline of the article in each target candidate area and generating a feature map;
the screening module is used for reading the generated feature map and rejecting a candidate area without the object contour feature vector;
the detection frame module is used for generating a detection frame around the article image according to the article contour feature vector, correcting and adjusting the size of the detection frame and accurately positioning the detection frame;
the classified information determining module is used for matching and comparing the characteristic vector inside the detection frame with the characteristic vector in the database, and determining and outputting the information of the articles inside the detection frame; and
and the data sending module reads the information of the detection frame and the information of the articles in the detection frame and sends the information to the display terminal, and the information is displayed on the display screen through the display terminal.
7. An apparatus for marking annotations on a display screen according to claim 6, wherein: the preprocessing module comprises:
the region dividing module is used for dividing the picture into a plurality of small regions and then aggregating adjacent small regions continuously through similarity and region size to form a plurality of candidate regions;
the characteristic vector extraction module is used for extracting the characteristic vectors of all the regions by using a shared convolution algorithm; and
and the feature map generation module is used for superposing the feature vectors of all dimensions and integrating the feature vectors into a feature map for outputting.
8. An apparatus for marking annotations on a display screen according to claim 6, wherein: an SVM module is arranged in the screening module, the SVM module is trained by inputting positive and negative samples, the trained SVM module is used as a two-classifier to screen and reject candidate regions, a region network generation module is arranged in the detection frame module, and the network generation module forms a detection frame around the outline of the article through the feature vector of the outline of the article.
9. An apparatus for marking annotations on a display screen according to claim 6, wherein: the classification information determination module includes:
the comparison module is used for comparing the similarity between the characteristic vector in the detection frame and the characteristic vector in the database and outputting the similarity;
the judging module is used for comparing the similarity with a set threshold value, and judging the type and the information of the article if the similarity is greater than the set threshold value; and
and the output module is used for reading the information of the article from the database and outputting the information after determining the information of the article in the detection frame.
10. A storage medium for marking annotations on a display screen, characterized by: the storage medium has stored thereon a computer program which, when being executed by a processor, is capable of carrying out the steps of the method of marking annotation on a display screen according to any one of claims 1 to 5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011144962.6A CN112256906A (en) | 2020-10-23 | 2020-10-23 | Method, device and storage medium for marking annotation on display screen |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011144962.6A CN112256906A (en) | 2020-10-23 | 2020-10-23 | Method, device and storage medium for marking annotation on display screen |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112256906A true CN112256906A (en) | 2021-01-22 |
Family
ID=74265007
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011144962.6A Pending CN112256906A (en) | 2020-10-23 | 2020-10-23 | Method, device and storage medium for marking annotation on display screen |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112256906A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116304391A (en) * | 2023-05-12 | 2023-06-23 | 深圳市前海数据服务有限公司 | Method and system for generating recommended selection area after map area frame selection |
Citations (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106504233A (en) * | 2016-10-18 | 2017-03-15 | 国网山东省电力公司电力科学研究院 | Image electric power widget recognition methodss and system are patrolled and examined based on the unmanned plane of Faster R CNN |
CN107133570A (en) * | 2017-04-07 | 2017-09-05 | 武汉睿智视讯科技有限公司 | A kind of vehicle/pedestrian detection method and system |
CN107992841A (en) * | 2017-12-13 | 2018-05-04 | 北京小米移动软件有限公司 | The method and device of identification objects in images, electronic equipment, readable storage medium storing program for executing |
CN108009554A (en) * | 2017-12-01 | 2018-05-08 | 国信优易数据有限公司 | A kind of image processing method and device |
CN108121986A (en) * | 2017-12-29 | 2018-06-05 | 深圳云天励飞技术有限公司 | Object detection method and device, computer installation and computer readable storage medium |
CN109145901A (en) * | 2018-08-14 | 2019-01-04 | 腾讯科技(深圳)有限公司 | Item identification method, device, computer readable storage medium and computer equipment |
CN109376717A (en) * | 2018-12-14 | 2019-02-22 | 中科软科技股份有限公司 | Personal identification method, device, electronic equipment and the storage medium of face comparison |
CN109886279A (en) * | 2019-01-24 | 2019-06-14 | 平安科技(深圳)有限公司 | Image processing method, device, computer equipment and storage medium |
CN109934216A (en) * | 2017-12-19 | 2019-06-25 | 华为技术有限公司 | The method, apparatus of image procossing, computer readable storage medium |
CN110489578A (en) * | 2019-08-12 | 2019-11-22 | 腾讯科技(深圳)有限公司 | Image processing method, device and computer equipment |
CN110517262A (en) * | 2019-09-02 | 2019-11-29 | 上海联影医疗科技有限公司 | Object detection method, device, equipment and storage medium |
CN110543857A (en) * | 2019-09-05 | 2019-12-06 | 安徽启新明智科技有限公司 | Contraband identification method, device and system based on image analysis and storage medium |
CN111290040A (en) * | 2020-03-12 | 2020-06-16 | 安徽启新明智科技有限公司 | Active double-view-angle correlation method based on image recognition |
CN111428726A (en) * | 2020-06-10 | 2020-07-17 | 中山大学 | Panorama segmentation method, system, equipment and storage medium based on graph neural network |
CN111626993A (en) * | 2020-05-07 | 2020-09-04 | 武汉科技大学 | Image automatic detection counting method and system based on embedded FEFnet network |
-
2020
- 2020-10-23 CN CN202011144962.6A patent/CN112256906A/en active Pending
Patent Citations (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106504233A (en) * | 2016-10-18 | 2017-03-15 | 国网山东省电力公司电力科学研究院 | Image electric power widget recognition methodss and system are patrolled and examined based on the unmanned plane of Faster R CNN |
CN107133570A (en) * | 2017-04-07 | 2017-09-05 | 武汉睿智视讯科技有限公司 | A kind of vehicle/pedestrian detection method and system |
CN108009554A (en) * | 2017-12-01 | 2018-05-08 | 国信优易数据有限公司 | A kind of image processing method and device |
CN107992841A (en) * | 2017-12-13 | 2018-05-04 | 北京小米移动软件有限公司 | The method and device of identification objects in images, electronic equipment, readable storage medium storing program for executing |
CN109934216A (en) * | 2017-12-19 | 2019-06-25 | 华为技术有限公司 | The method, apparatus of image procossing, computer readable storage medium |
CN108121986A (en) * | 2017-12-29 | 2018-06-05 | 深圳云天励飞技术有限公司 | Object detection method and device, computer installation and computer readable storage medium |
CN109145901A (en) * | 2018-08-14 | 2019-01-04 | 腾讯科技(深圳)有限公司 | Item identification method, device, computer readable storage medium and computer equipment |
CN109376717A (en) * | 2018-12-14 | 2019-02-22 | 中科软科技股份有限公司 | Personal identification method, device, electronic equipment and the storage medium of face comparison |
CN109886279A (en) * | 2019-01-24 | 2019-06-14 | 平安科技(深圳)有限公司 | Image processing method, device, computer equipment and storage medium |
CN110489578A (en) * | 2019-08-12 | 2019-11-22 | 腾讯科技(深圳)有限公司 | Image processing method, device and computer equipment |
CN110517262A (en) * | 2019-09-02 | 2019-11-29 | 上海联影医疗科技有限公司 | Object detection method, device, equipment and storage medium |
CN110543857A (en) * | 2019-09-05 | 2019-12-06 | 安徽启新明智科技有限公司 | Contraband identification method, device and system based on image analysis and storage medium |
CN111290040A (en) * | 2020-03-12 | 2020-06-16 | 安徽启新明智科技有限公司 | Active double-view-angle correlation method based on image recognition |
CN111626993A (en) * | 2020-05-07 | 2020-09-04 | 武汉科技大学 | Image automatic detection counting method and system based on embedded FEFnet network |
CN111428726A (en) * | 2020-06-10 | 2020-07-17 | 中山大学 | Panorama segmentation method, system, equipment and storage medium based on graph neural network |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116304391A (en) * | 2023-05-12 | 2023-06-23 | 深圳市前海数据服务有限公司 | Method and system for generating recommended selection area after map area frame selection |
CN116304391B (en) * | 2023-05-12 | 2023-07-25 | 深圳市前海数据服务有限公司 | Method and system for generating recommended selection area after map area frame selection |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111080622B (en) | Neural network training method, workpiece surface defect classification and detection method and device | |
CN110503054B (en) | Text image processing method and device | |
CN111461133B (en) | Express delivery surface single item name identification method, device, equipment and storage medium | |
CN111242899B (en) | Image-based flaw detection method and computer-readable storage medium | |
CN113780087B (en) | Postal package text detection method and equipment based on deep learning | |
CN112967255A (en) | Shield segment defect type identification and positioning system and method based on deep learning | |
CN113592886B (en) | Drawing examining method and device for building drawing, electronic equipment and medium | |
CN114463767A (en) | Credit card identification method, device, computer equipment and storage medium | |
da Silva et al. | Automatic discrimination between printed and handwritten text in documents | |
CN111612747A (en) | Method and system for rapidly detecting surface cracks of product | |
CN112446259A (en) | Image processing method, device, terminal and computer readable storage medium | |
CN114549493A (en) | Magnetic core defect detection system and method based on deep learning | |
US10970531B2 (en) | Digitization of industrial inspection sheets by inferring visual relations | |
CN111738979A (en) | Automatic certificate image quality inspection method and system | |
CN112256906A (en) | Method, device and storage medium for marking annotation on display screen | |
CN114067296A (en) | Method and device for identifying surface defects of steel rail | |
Nguyen et al. | Tensor voting based text localization in natural scene images | |
CN112200789A (en) | Image identification method and device, electronic equipment and storage medium | |
WO2017058252A1 (en) | Detecting document objects | |
CN112861861B (en) | Method and device for recognizing nixie tube text and electronic equipment | |
CN112149654B (en) | Invoice text information identification method based on deep learning | |
Kumari et al. | On the use of Moravec operator for text detection in document images and video frames | |
Molefe et al. | Classification of thermite welding defects using local binary patterns and k nearest neighbors | |
CN113554024A (en) | Method and device for determining cleanliness of vehicle and computer equipment | |
CN114627457A (en) | Ticket information identification method and device |
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: 20210122 |