CN110659374A - Method for searching images by images based on neural network extraction of vehicle characteristic values and attributes - Google Patents
Method for searching images by images based on neural network extraction of vehicle characteristic values and attributes Download PDFInfo
- Publication number
- CN110659374A CN110659374A CN201910886043.7A CN201910886043A CN110659374A CN 110659374 A CN110659374 A CN 110659374A CN 201910886043 A CN201910886043 A CN 201910886043A CN 110659374 A CN110659374 A CN 110659374A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- neural network
- searching
- attributes
- image
- 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
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
- G06F16/53—Querying
- G06F16/532—Query formulation, e.g. graphical querying
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method for searching a picture by a picture based on extracting vehicle characteristic values and attributes by a neural network, which is used for acquiring a real-time bayonet snap-shot image to carry out image preprocessing operation; positioning and analyzing the vehicle from the preprocessed image; calculating and extracting global characteristic values of the positioned vehicle images; calculating and extracting attribute values of the positioned vehicle images; carrying out weight setting and characteristic value search threshold value setting on each attribute value of the vehicle, and carrying out screening and subtraction on a vehicle basic library; vehicle searching is carried out through setting weights of the global features of the vehicles and the attributes of the vehicles in a graph searching function; and sequencing the search results to obtain the search results. The invention has fast characteristic identification speed and greatly improves the accuracy of searching.
Description
Technical Field
The invention relates to a method for searching images by images, in particular to a method for searching images by images based on extraction of vehicle characteristic values and attributes by a neural network, and belongs to the field of information search.
Background
Searching pictures is a technology for searching similar pictures by inputting pictures, and provides a function of searching related graphic image data for users. The technology relates to various disciplines such as database management, computer graphics, image processing, pattern recognition, information retrieval and the like. The related technology mainly comprises three key technologies of characteristic value extraction, characteristic value representation and similarity calculation. The image searching technology is widely applied to various fields such as big data image retrieval, video detection, internet shopping search engines and the like.
The map searching function of mass vehicle data mainly adopts the conventional image contour characteristic information to perform matching analysis, and does not consider the vehicle specific relevant attributes (including the vehicle license plate, color, vehicle type, vehicle logo, hanging decoration, annual inspection mark and other relevant information) to perform vehicle early-stage screening, so that the map searching accuracy of the vehicle is not high.
Patent CN 201711393923.8 "a system and method for searching images by a multitask bayonet vehicle" focuses on training a positioning network based on an improved edge box detection technology and a cascading loss function manner, and respectively positions and detects three parts of a vehicle, an annual inspection mark and a vehicle lamp in a bayonet vehicle image, and combines global and local features, which mainly detect the vehicle, the annual mark and the vehicle lamp, thereby increasing inspection speed.
Disclosure of Invention
The invention aims to provide a method for searching a map by a map based on the extraction of vehicle characteristic values and attributes by a neural network, and improve the accuracy of vehicle searching.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows:
a method for searching a map by a map based on a neural network for extracting vehicle characteristic values and attributes is characterized by comprising the following steps:
the method comprises the following steps: acquiring a real-time bayonet snapshot image and performing image preprocessing operation;
step two: positioning and analyzing the vehicle from the preprocessed image;
step three: calculating and extracting global characteristic values of the positioned vehicle images;
step four: calculating and extracting attribute values of the positioned vehicle images;
step five: carrying out weight setting and characteristic value search threshold value setting on each attribute value of the vehicle, and carrying out screening and subtraction on a vehicle basic library;
step six: vehicle searching is carried out through setting weights of the global features of the vehicles and the attributes of the vehicles in a graph searching function;
step seven: and sequencing the search results to obtain the search results.
Further, the step one is specifically
1.1, carrying out noise reduction processing on the picture, keeping the integrity of original information as much as possible, and removing useless information in signals at the same time, wherein the used method comprises median filtering based on a spatial domain and wavelet threshold denoising based on a wavelet domain;
1.2 size processing is carried out on the picture.
Further, the second step is specifically
2.1, positioning the vehicle in the picture, and identifying the pixel point position of the vehicle in the picture;
2.2 standardize the size of the located vehicle image sample;
and 2.3, isolating the to-be-detected picture positioned to the vehicle pixel point.
Further, the third step is specifically
3.1 calculating a characteristic value of the vehicle by using a sift algorithm;
3.2 Using CNN convolutional neural network Algorithm, dimensionality reduction is performed on the calculated vehicle eigenvalues
3.3 extracting the last layer of feature vector of the CNN neural network, converting the last layer of feature vector into a one-dimensional array, and reducing the original 128-dimensional feature value of the sift feature into the one-dimensional array for convenient representation, storage and calculation.
Further, the fourth step is specifically
4.1 identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolution neural network and an SVM to obtain the license plate attribute;
4.2, identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolutional neural network and an SVM to obtain the vehicle type attribute;
4.3 identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolution neural network and an SVM to obtain the color attribute of the vehicle body;
4.4, identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolution neural network and an SVM to obtain the vehicle logo attribute;
4.5 identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolutional neural network and an SVM to obtain the attribute of the annual inspection mark of the vehicle;
4.6 identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolution neural network and an SVM to obtain vehicle hanging decoration attributes;
and 4.7, the acquired vehicle attributes are subjected to structured processing, so that comparison calculation and storage are facilitated.
Further, the fifth step is specifically that
5.1 setting weights for each attribute value of the vehicle, and setting different attribute values according to the contribution degree of each attribute value to a search result in a graph search algorithm;
5.2 setting a search threshold value for the characteristic value;
and 5.3, the vehicle basic library is screened and reduced by the two steps, so that the number of pictures to be searched in the target library of the picture searched by the picture is reduced.
Further, the sixth step is specifically that
6.1 the global characteristic value of the vehicle is subjected to similarity calculation, and calculation methods such as Euclidean distance and Mahalanobis distance are adopted for calculation
6.2 vehicle search by setting weight of vehicle attributes in the map search function
And 6.3 combining the two steps, searching the images to be detected in the screened vehicle basic image library by using the images to obtain a search result.
Further, the seventh step is specifically to perform sorting processing on the search results in the sixth step to obtain the top 10 pictures most similar to the vehicle to be searched for displaying.
Compared with the prior art, the invention has the following advantages and effects:
1. according to the invention, the analysis and extraction of the relevant attributes of the vehicle (including the license plate, the color, the type and the logo of the vehicle) are carried out by accurately positioning the vehicle in the picture, and the calculation of the matching degree of the vehicle is realized by setting the weight of the attributes and the characteristics of the vehicle, so that when a customer acquires one vehicle picture, other vehicle pictures most similar to the vehicle picture are searched from a map library;
2. through the analysis of an intelligent algorithm, the extraction of the attributes and the characteristic values of the vehicles in the picture to be searched is completed within 200 ms;
3. the vehicle searches the picture function with the picture on the premise of ten million levels of basic libraries, and the searching accuracy rate reaches 98%.
Drawings
FIG. 1 is a flowchart of a method for searching a map based on neural network extraction of vehicle feature values and attributes according to the present invention.
Detailed Description
The present invention is further illustrated by the following examples, which are illustrative of the present invention and are not to be construed as being limited thereto.
As shown in fig. 1, a method for searching a map based on a neural network to extract vehicle feature values and attributes of the present invention includes the following steps:
the method comprises the following steps: acquiring a real-time bayonet snapshot image and performing image preprocessing operation;
1.1, carrying out noise reduction processing on the picture, keeping the integrity of original information as much as possible, and removing useless information in signals at the same time, wherein the used method comprises median filtering based on a spatial domain and wavelet threshold denoising based on a wavelet domain;
1.2 size processing is carried out on the picture.
Step two: positioning and analyzing the vehicle from the preprocessed image;
2.1, positioning the vehicle in the picture, and identifying the pixel point position of the vehicle in the picture;
2.2 standardize the size of the located vehicle image sample;
and 2.3, isolating the to-be-detected picture positioned to the vehicle pixel point.
Step three: calculating and extracting global characteristic values of the positioned vehicle images;
3.1 calculating a characteristic value of the vehicle by using a sift algorithm;
3.2 Using CNN convolutional neural network Algorithm, dimensionality reduction is performed on the calculated vehicle eigenvalues
3.3 extracting the last layer of feature vector of the CNN neural network, converting the last layer of feature vector into a one-dimensional array, and reducing the original 128-dimensional feature value of the sift feature into the one-dimensional array for convenient representation, storage and calculation.
Step four: calculating and extracting attribute values of the positioned vehicle images;
4.1 identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolution neural network and an SVM to obtain the license plate attribute;
4.2, identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolutional neural network and an SVM to obtain the vehicle type attribute;
4.3 identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolution neural network and an SVM to obtain the color attribute of the vehicle body;
4.4, identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolution neural network and an SVM to obtain the vehicle logo attribute;
4.5 identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolutional neural network and an SVM to obtain the attribute of the annual inspection mark of the vehicle;
4.6 identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolution neural network and an SVM to obtain vehicle hanging decoration attributes;
and 4.7, the acquired vehicle attributes are subjected to structured processing, so that comparison calculation and storage are facilitated.
Step five: carrying out weight setting and characteristic value search threshold value setting on each attribute value of the vehicle, and carrying out screening and subtraction on a vehicle basic library;
5.1 setting weights for each attribute value of the vehicle, and setting different attribute values according to the contribution degree of each attribute value to a search result in a graph search algorithm;
5.2 setting a search threshold value for the characteristic value;
and 5.3, the vehicle basic library is screened and reduced by the two steps, so that the number of pictures to be searched in the target library of the picture searched by the picture is reduced.
Step six: vehicle searching is carried out through setting weights of the global features of the vehicles and the attributes of the vehicles in a graph searching function;
6.1 the global characteristic value of the vehicle is subjected to similarity calculation, and calculation methods such as Euclidean distance and Mahalanobis distance are adopted for calculation
6.2 vehicle search by setting weight of vehicle attributes in the map search function
And 6.3 combining the two steps, searching the images to be detected in the screened vehicle basic image library by using the images to obtain a search result.
Step seven: and sequencing the search results to obtain the search results. And sequencing the search results in the sixth step to obtain the first 10 pictures most similar to the vehicle to be searched for display.
According to the invention, the analysis and extraction of the relevant attributes of the vehicle (including the license plate, the color, the type and the logo of the vehicle) are carried out by accurately positioning the vehicle in the picture, and the calculation of the matching degree of the vehicle is realized by setting the weight of the attributes and the characteristics of the vehicle, so that when a customer acquires one vehicle picture, other vehicle pictures most similar to the vehicle picture are searched from a map library; through the analysis of an intelligent algorithm, the extraction of the attributes and the characteristic values of the vehicles in the picture to be searched is completed within 200 ms; the vehicle searches the picture function with the picture on the premise of ten million levels of basic libraries, and the searching accuracy rate reaches 98%.
Although the present invention has been described with reference to a preferred embodiment, it should be understood that various changes, substitutions and alterations can be made herein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (8)
1. A method for searching a map by a map based on a neural network for extracting vehicle characteristic values and attributes is characterized by comprising the following steps:
the method comprises the following steps: acquiring a real-time bayonet snapshot image and performing image preprocessing operation;
step two: positioning and analyzing the vehicle from the preprocessed image;
step three: calculating and extracting global characteristic values of the positioned vehicle images;
step four: calculating and extracting attribute values of the positioned vehicle images;
step five: carrying out weight setting and characteristic value search threshold value setting on each attribute value of the vehicle, and carrying out screening and subtraction on a vehicle basic library;
step six: vehicle searching is carried out through setting weights of the global features of the vehicles and the attributes of the vehicles in a graph searching function;
step seven: and sequencing the search results to obtain the search results.
2. The method for searching a map based on the neural network extraction of the vehicle characteristic values and attributes as claimed in claim 1, wherein: the step one is specifically
1.1, carrying out noise reduction processing on the picture, keeping the integrity of original information as much as possible, and removing useless information in signals at the same time, wherein the used method comprises median filtering based on a spatial domain and wavelet threshold denoising based on a wavelet domain;
1.2 size processing is carried out on the picture.
3. The method for searching a map based on the neural network extraction of the vehicle characteristic values and attributes as claimed in claim 1, wherein: the second step is specifically that
2.1, positioning the vehicle in the picture, and identifying the pixel point position of the vehicle in the picture;
2.2 standardize the size of the located vehicle image sample;
and 2.3, isolating the to-be-detected picture positioned to the vehicle pixel point.
4. The method for searching a map based on the neural network extraction of the vehicle characteristic values and attributes as claimed in claim 1, wherein: the third step is specifically that
3.1 calculating a characteristic value of the vehicle by using a sift algorithm;
3.2 Using CNN convolutional neural network Algorithm, dimensionality reduction is performed on the calculated vehicle eigenvalues
3.3 extracting the last layer of feature vector of the CNN neural network, converting the last layer of feature vector into a one-dimensional array, and reducing the original 128-dimensional feature value of the sift feature into the one-dimensional array for convenient representation, storage and calculation.
5. The method for searching a map based on the neural network extraction of the vehicle characteristic values and attributes as claimed in claim 1, wherein: the fourth step is specifically that
4.1 identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolution neural network and an SVM to obtain the license plate attribute;
4.2, identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolutional neural network and an SVM to obtain the vehicle type attribute;
4.3 identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolution neural network and an SVM to obtain the color attribute of the vehicle body;
4.4, identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolution neural network and an SVM to obtain the vehicle logo attribute;
4.5 identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolutional neural network and an SVM to obtain the attribute of the annual inspection mark of the vehicle;
4.6 identifying the positioned vehicle image by a method of combining a deep learning neural network, a CNN convolution neural network and an SVM to obtain vehicle hanging decoration attributes;
and 4.7, the acquired vehicle attributes are subjected to structured processing, so that comparison calculation and storage are facilitated.
6. The method for searching a map based on the neural network extraction of the vehicle characteristic values and attributes as claimed in claim 1, wherein: the fifth step is specifically that
5.1 setting weights for each attribute value of the vehicle, and setting different attribute values according to the contribution degree of each attribute value to a search result in a graph search algorithm;
5.2 setting a search threshold value for the characteristic value;
and 5.3, the vehicle basic library is screened and reduced by the two steps, so that the number of pictures to be searched in the target library of the picture searched by the picture is reduced.
7. The method for searching a map based on the neural network extraction of the vehicle characteristic values and attributes as claimed in claim 1, wherein: the sixth step is specifically that
6.1 the global characteristic value of the vehicle is subjected to similarity calculation, and calculation methods such as Euclidean distance and Mahalanobis distance are adopted for calculation
6.2 vehicle search by setting weight of vehicle attributes in the map search function
And 6.3 combining the two steps, searching the images to be detected in the screened vehicle basic image library by using the images to obtain a search result.
8. The method for searching a map based on the neural network extraction of the vehicle characteristic values and attributes as claimed in claim 1, wherein: and the seventh step is specifically to sort the search results in the sixth step to obtain the first 10 pictures most similar to the vehicle to be searched for displaying.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910886043.7A CN110659374A (en) | 2019-09-19 | 2019-09-19 | Method for searching images by images based on neural network extraction of vehicle characteristic values and attributes |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910886043.7A CN110659374A (en) | 2019-09-19 | 2019-09-19 | Method for searching images by images based on neural network extraction of vehicle characteristic values and attributes |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110659374A true CN110659374A (en) | 2020-01-07 |
Family
ID=69037277
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910886043.7A Pending CN110659374A (en) | 2019-09-19 | 2019-09-19 | Method for searching images by images based on neural network extraction of vehicle characteristic values and attributes |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110659374A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111814751A (en) * | 2020-08-14 | 2020-10-23 | 深延科技(北京)有限公司 | Vehicle attribute analysis method and system based on deep learning target detection and image recognition |
| CN112818736A (en) * | 2020-12-10 | 2021-05-18 | 西南交通大学 | Emergency command big data supporting platform |
| CN112990048A (en) * | 2021-03-26 | 2021-06-18 | 中科视语(北京)科技有限公司 | Vehicle pattern recognition method and device |
| CN113011540A (en) * | 2021-03-01 | 2021-06-22 | 北京骑胜科技有限公司 | Identification method, device, readable storage medium and electronic equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106611016A (en) * | 2015-10-27 | 2017-05-03 | 北京航空航天大学 | Image retrieval method based on decomposable word pack model |
| CN106776943A (en) * | 2016-12-01 | 2017-05-31 | 中科唯实科技（北京）有限公司 | A kind of vehicle retrieval method based on AutoEncoder and attribute tags |
| CN108229468A (en) * | 2017-06-28 | 2018-06-29 | 北京市商汤科技开发有限公司 | Vehicle appearance feature recognition and vehicle retrieval method, apparatus, storage medium, electronic equipment |
| CN109558823A (en) * | 2018-11-22 | 2019-04-02 | 北京市首都公路发展集团有限公司 | A kind of vehicle identification method and system to scheme to search figure |
-
2019
- 2019-09-19 CN CN201910886043.7A patent/CN110659374A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106611016A (en) * | 2015-10-27 | 2017-05-03 | 北京航空航天大学 | Image retrieval method based on decomposable word pack model |
| CN106776943A (en) * | 2016-12-01 | 2017-05-31 | 中科唯实科技（北京）有限公司 | A kind of vehicle retrieval method based on AutoEncoder and attribute tags |
| CN108229468A (en) * | 2017-06-28 | 2018-06-29 | 北京市商汤科技开发有限公司 | Vehicle appearance feature recognition and vehicle retrieval method, apparatus, storage medium, electronic equipment |
| CN109558823A (en) * | 2018-11-22 | 2019-04-02 | 北京市首都公路发展集团有限公司 | A kind of vehicle identification method and system to scheme to search figure |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111814751A (en) * | 2020-08-14 | 2020-10-23 | 深延科技(北京)有限公司 | Vehicle attribute analysis method and system based on deep learning target detection and image recognition |
| CN112818736A (en) * | 2020-12-10 | 2021-05-18 | 西南交通大学 | Emergency command big data supporting platform |
| CN113011540A (en) * | 2021-03-01 | 2021-06-22 | 北京骑胜科技有限公司 | Identification method, device, readable storage medium and electronic equipment |
| CN112990048A (en) * | 2021-03-26 | 2021-06-18 | 中科视语(北京)科技有限公司 | Vehicle pattern recognition method and device |
| CN112990048B (en) * | 2021-03-26 | 2021-11-23 | 中科视语(北京)科技有限公司 | Vehicle pattern recognition method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107679250B (en) | Multi-task layered image retrieval method based on deep self-coding convolutional neural network | |
| TWI651662B (en) | Image annotation method, electronic device and non-transitory computer readable storage medium | |
| Quack et al. | Efficient mining of frequent and distinctive feature configurations | |
| Li et al. | Automatic text detection and tracking in digital video | |
| Pan et al. | A robust system to detect and localize texts in natural scene images | |
| CN110717534B (en) | Target classification and positioning method based on network supervision | |
| CN110659374A (en) | Method for searching images by images based on neural network extraction of vehicle characteristic values and attributes | |
| CN107480620B (en) | Remote sensing image automatic target identification method based on heterogeneous feature fusion | |
| Xie et al. | Combination of dominant color descriptor and Hu moments in consistent zone for content based image retrieval | |
| Alsmadi et al. | Fish recognition based on robust features extraction from color texture measurements using back-propagation classifier | |
| Hussain et al. | Robust pre-processing technique based on saliency detection for content based image retrieval systems | |
| CN105718552A (en) | Clothing freehand sketch based clothing image retrieval method | |
| CN103955496B (en) | A kind of quick live tire trace decorative pattern searching algorithm | |
| US8027978B2 (en) | Image search method, apparatus, and program | |
| CN102176208A (en) | Robust video fingerprint method based on three-dimensional space-time characteristics | |
| TR201614118A1 (en) | A vessel identification method. | |
| JP2007531136A (en) | Method and apparatus for extracting visual object categories from a database with images | |
| Chakraborty et al. | Hand gesture recognition: A comparative study | |
| Li et al. | Performance comparison of saliency detection | |
| Badura et al. | Advanced scale-space, invariant, low detailed feature recognition from images-car brand recognition | |
| CN101118544A (en) | Method for constructing picture shape contour outline descriptor | |
| CN109902690A (en) | Image recognition technology | |
| Kavya | Feature extraction technique for robust and fast visual tracking: a typical review | |
| Shinde et al. | Street view text detection methods | |
| Amuthavalli et al. | Notice of Violation of IEEE Publication Principles: Sketch Based Image Retrieval System Using ExHoG |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 210000 No. 268, Hanzhoung Road, Nanjing, Jiangsu Applicant after: CLP Hongxin Information Technology Co., Ltd Address before: 210000 No. 268, Hanzhoung Road, Nanjing, Jiangsu Applicant before: Jiangsu Hongxin System Integration Co., Ltd. |