CN113269150A - Vehicle multi-attribute identification system and method based on deep learning - Google Patents
Vehicle multi-attribute identification system and method based on deep learning Download PDFInfo
- Publication number
- CN113269150A CN113269150A CN202110703877.7A CN202110703877A CN113269150A CN 113269150 A CN113269150 A CN 113269150A CN 202110703877 A CN202110703877 A CN 202110703877A CN 113269150 A CN113269150 A CN 113269150A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- attribute
- image
- scene image
- deep learning
- 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
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/08—Detecting or categorising vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a vehicle multi-attribute identification system and method based on deep learning, relating to the technical field of computer vision; the method comprises the steps of obtaining a vehicle image in a scene image, inputting a multi-scale convolutional neural network for vehicle attribute identification, extracting a feature vector of the vehicle image, obtaining an attribute probability vector of the vehicle attribute according to the corresponding feature vector by using a multi-label classification method, and outputting the corresponding vehicle attribute with the attribute probability vector larger than a probability threshold value as an identification result.
Description
Technical Field
The invention discloses a system and a method, relates to the technical field of computer vision, and particularly relates to a system and a method for vehicle multi-attribute recognition based on deep learning.
Background
In order to identify different vehicle attributes under the service requirements of an intelligent traffic monitoring system and the like, it is generally necessary to identify accessories such as vehicle colors, driver behavior attributes, annual inspection marks and the like. Therefore, the automatic understanding of the vehicle attributes by the computer can be improved, and the intelligent traffic security problem which is difficult to solve, such as vehicle heavy identification, traffic control, vehicle target retrieval and the like, can be solved.
The traditional vehicle attribute identification method mainly comprises the steps of extracting features such as HOG (hot object) and the like by a manually pre-designed feature extractor, wherein the manually designed feature extractor is relatively complex, has insufficient feature expression capability and low model robustness, and a monitoring scene is complex and changeable (vehicle type, illumination, visual angle change and the like), so that the traditional method is difficult to obtain a good identification effect. With the continuous development of deep learning technology, researchers use the deep learning technology to identify vehicle attributes, but at present, no perfect method is available to effectively identify various attributes of a vehicle.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a vehicle multi-attribute identification system and method based on deep learning, which realize end-to-end reasoning operation, can accurately and efficiently identify the attribute characteristics of the image of the vehicle to be identified, greatly improve the attribute analysis of the vehicle and the identification and retrieval of the suspected vehicle under the scene of unlimited security and protection and greatly reduce the labor cost.
The specific scheme provided by the invention is as follows:
the method for vehicle multi-attribute recognition based on deep learning comprises the steps of obtaining a vehicle image in a scene image, inputting a multi-scale convolutional neural network for vehicle attribute recognition, extracting a feature vector of the vehicle image, obtaining an attribute probability vector of a vehicle attribute according to the corresponding feature vector by using a multi-label classification method, and outputting the corresponding vehicle attribute of which the attribute probability vector is larger than a probability threshold value as a recognition result.
Further, the method for vehicle multi-attribute identification based on deep learning acquires the vehicle image in the scene image:
and positioning the vehicle target in the scene image, and cutting the scene image according to the positioning information to obtain all the vehicle images.
Further, the method for identifying the vehicle multiple attributes based on the deep learning comprises the following specific steps:
step 1: acquiring a scene image, inputting the scene image into a pre-trained vehicle target detection model, outputting the positioning position coordinate information of a vehicle target,
step 2: and cutting the scene image according to the coordinate information of the positioning position of the vehicle target to obtain the vehicle image.
Further, the vehicle image is preprocessed in the deep learning-based vehicle multi-attribute identification method:
and (4) scaling the vehicle images to a uniform size and carrying out normalization processing.
The system for vehicle multi-attribute identification based on deep learning comprises a vehicle target positioning module and a vehicle attribute identification module,
the vehicle target positioning module acquires a vehicle image in a scene image, inputs the vehicle image into a multi-scale convolutional neural network of the vehicle attribute identification module, the vehicle attribute identification module extracts a feature vector of the vehicle image, acquires an attribute probability vector of a vehicle attribute according to the corresponding feature vector by using a multi-label classification method, and outputs the corresponding vehicle attribute of which the attribute probability vector is greater than a probability threshold value as an identification result.
Further, the vehicle target positioning module in the system for vehicle multi-attribute identification based on deep learning acquires the vehicle image in the scene image:
and positioning the vehicle target in the scene image, and cutting the scene image according to the positioning information to obtain all the vehicle images.
Further, the specific steps of the vehicle target positioning module in the system for vehicle multi-attribute identification based on deep learning to acquire the vehicle image in the scene image are as follows:
step 1: acquiring a scene image, inputting the scene image into a pre-trained vehicle target detection model, outputting the positioning position coordinate information of a vehicle target,
step 2: and cutting the scene image according to the coordinate information of the positioning position of the vehicle target to obtain the vehicle image.
Further, the vehicle attribute identification module in the deep learning-based vehicle multi-attribute identification system preprocesses the vehicle image:
and (4) scaling the vehicle images to a uniform size and carrying out normalization processing.
An apparatus for deep learning based vehicle multi-attribute identification, comprising at least one memory and at least one processor;
the at least one memory to store a machine readable program;
the at least one processor is used for calling the machine readable program to execute the deep learning-based vehicle multi-attribute identification method.
The invention has the advantages that:
compared with the prior art, the invention realizes the identification of the vehicle attribute through the multi-scale convolutional neural network and the improved multi-label loss, wherein:
the method can be built by adopting an end-to-end neural network, compared with the existing method, the method can automatically extract the characteristics without human intervention, and improves the speed and the precision of vehicle attribute identification;
by using the multi-scale convolutional neural network to extract the vehicle attribute characteristics, the attribute characteristics of the small target can be accurately positioned, and the attribute identification effect is improved;
the problem of unbalanced attribute data is solved by using the multi-label loss, and various attributes of the vehicle can be effectively and accurately identified.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a schematic diagram of the system architecture of the present invention.
FIG. 2 is a schematic diagram of a picture of object identification according to the method of the present invention.
Detailed Description
The present invention is further described below in conjunction with the following figures and specific examples so that those skilled in the art may better understand the present invention and practice it, but the examples are not intended to limit the present invention.
The invention provides a vehicle multi-attribute identification method based on deep learning, which comprises the steps of obtaining a vehicle image in a scene image, inputting a multi-scale convolutional neural network for vehicle attribute identification, extracting a feature vector of the vehicle image, obtaining an attribute probability vector of the vehicle attribute according to the corresponding feature vector by using a multi-label classification method, and outputting the corresponding vehicle attribute of which the attribute probability vector is greater than a probability threshold value as an identification result.
Compared with the prior art, the method realizes simultaneous recognition of multiple attributes of the vehicle through the multi-scale convolutional neural network and the improved multi-label loss, and solves the problems of low vehicle attribute recognition accuracy and low recognition efficiency caused by multiple vehicles, fuzzy image quality and the like in a scene.
In specific application, in some embodiments of the method, the process of the method is described by taking the security scene as an example for vehicle identification. The specific process is as follows:
step 1: acquiring a to-be-detected monitoring image of a security camera, inputting the to-be-detected image into a pre-trained vehicle target detection model, and outputting position coordinate information of all vehicle targets;
step 2: cutting the security image according to the coordinate information of all vehicle target detection obtained in the step 1 to obtain all vehicle images;
and step 3: the output of all vehicle images is taken as input, and the vehicle images are preprocessed to obtain a result which accords with the standard input of a classification model;
and 4, step 4: and (3) transmitting the output of the step (3) to a convolutional neural network for vehicle attribute identification, extracting a feature vector of a vehicle image, acquiring an attribute probability vector of the vehicle attribute according to the corresponding feature vector by using a multi-label classification method, and outputting the corresponding vehicle attribute with the attribute probability vector larger than a probability threshold value as an identification result, namely outputting vehicle attribute identification information.
On the basis of the above embodiment, when performing end-to-end vehicle multi-attribute identification by the method of the present invention, the following steps may be specifically performed:
step 101: the method comprises the steps of using a security camera to obtain an image to be detected, wherein the image to be detected comprises vehicle types (such as trucks, cars and buses), and scenes including images which are not limited to driving or bayonets.
Step 102: and (3) using a pre-trained target detection network, not limited to target detection models such as YoloV3, SSD, fast-rcnn and the like as vehicle detection models, inputting the image to be detected obtained in the step (101) into the target detection network, and performing forward propagation calculation once to obtain coordinate information corresponding to all vehicles.
Step 103: for the vehicle object detection coordinate information acquired in step 102, a corresponding vehicle image is cut out from the original image and used as an input of the vehicle attribute identification module.
Step 201: preprocessing the vehicle image result of the step 103, performing pixel value filling operation on the short edge of the image on the premise of keeping the image not deformed, so that the length-width ratio of the vehicle image is 1 and no deformation is generated, then scaling the vehicle image to a uniform size, such as 299 by 299, and finally performing normalization operation, so that the vehicle image meets the model input standard; the formula is as follows:
wherein, X is the pixel value matrix of the original image, X' is the pixel value matrix of the normalized image, mu is the mean value of the imagenet data set, and sigma is the standard deviation value of the imagenet data set.
Step 202: common convolutional neural networks such as resnet50 and inceptionV3 are used as feature extraction backbone networks; and extracting the features of the image by adopting multi-scale convolution and outputting a feature vector.
Step 203: inputting the result of the step 201 into the step 202 to obtain a feature vector corresponding to the vehicle image, inputting the feature vector into the multi-label classifier, outputting an attribute probability vector with the dimension same as that of the attribute list, and determining the corresponding attribute with the attribute probability vector larger than the probability threshold as a vehicle multi-attribute identification output result according to the probability threshold, such as 0.5.
Step 204: and mapping the result obtained in the step 203 corresponding to the attribute list to obtain the identified attribute output.
The attribute list may refer to table 1.
TABLE 1
The method of the invention is based on vehicle target positioning and vehicle attribute identification, and can effectively solve the problem of inaccurate attribute identification caused by a plurality of vehicles in the image and fuzzy image quality. The method can be built based on an end-to-end neural network, can accurately and efficiently identify the attribute characteristics of the vehicle image to be identified, greatly improves the attribute analysis of the vehicle and the identification and retrieval of the suspected vehicle under the unlimited security scene, and greatly reduces the labor cost.
Meanwhile, the invention provides a system for vehicle multi-attribute recognition based on deep learning, which comprises a vehicle target positioning module and a vehicle attribute recognition module,
the vehicle target positioning module acquires a vehicle image in a scene image, inputs the vehicle image into a multi-scale convolutional neural network of the vehicle attribute identification module, the vehicle attribute identification module extracts a feature vector of the vehicle image, acquires an attribute probability vector of a vehicle attribute according to the corresponding feature vector by using a multi-label classification method, and outputs the corresponding vehicle attribute of which the attribute probability vector is greater than a probability threshold value as an identification result. The information interaction, execution process and other contents between the modules in the system are based on the same concept as the method embodiment of the present invention, and specific contents can be referred to the description in the method embodiment of the present invention, and are not described herein again. Similarly, the system can be built by adopting an end-to-end neural network, compared with the prior art, the system can automatically extract features without human intervention, and the speed and the precision of vehicle attribute identification are improved; by using the multi-scale convolutional neural network to extract the vehicle attribute characteristics, the attribute characteristics of the small target can be accurately positioned, and the attribute identification effect is improved; the problem of unbalanced attribute data is solved by using the multi-label loss, and various attributes of the vehicle can be effectively and accurately identified.
And an apparatus for vehicle multi-attribute identification based on deep learning, comprising at least one memory and at least one processor;
the at least one memory to store a machine readable program;
the at least one processor is used for calling the machine readable program to execute the deep learning-based vehicle multi-attribute identification method.
The contents of information interaction, readable program process execution and the like of the processor in the device are based on the same concept as the method embodiment of the present invention, and specific contents can be referred to the description in the method embodiment of the present invention, and are not described herein again. Similarly, compared with the prior art, the device can automatically extract features in the end-to-end neural network construction, does not need human intervention, and improves the speed and the precision of vehicle attribute identification; by using the multi-scale convolutional neural network to extract the vehicle attribute characteristics, the attribute characteristics of the small target can be accurately positioned, and the attribute identification effect is improved; the problem of unbalanced attribute data is solved by using the multi-label loss, and various attributes of the vehicle can be effectively and accurately identified.
It should be noted that not all steps and modules in the processes and system structures in the preferred embodiments are necessary, and some steps or modules may be omitted according to actual needs. The execution order of the steps is not fixed and can be adjusted as required. The system structure described in the above embodiments may be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by a plurality of physical entities, or some components in a plurality of independent devices may be implemented together.
The above-mentioned embodiments are merely preferred embodiments for fully illustrating the present invention, and the scope of the present invention is not limited thereto. The equivalent substitution or change made by the technical personnel in the technical field on the basis of the invention is all within the protection scope of the invention. The protection scope of the invention is subject to the claims.
Claims (9)
1. The method for vehicle multi-attribute recognition based on deep learning is characterized by obtaining a vehicle image in a scene image, inputting a multi-scale convolutional neural network for vehicle attribute recognition, extracting a feature vector of the vehicle image, obtaining an attribute probability vector of the vehicle attribute according to the corresponding feature vector by using a multi-label classification method, and outputting the corresponding vehicle attribute of which the attribute probability vector is greater than a probability threshold value as a recognition result.
2. The method for vehicle multi-attribute recognition based on deep learning of claim 1, wherein the vehicle image in the scene image is acquired by:
and positioning the vehicle target in the scene image, and cutting the scene image according to the positioning information to obtain all the vehicle images.
3. The method for vehicle multi-attribute recognition based on deep learning of claim 2, which is characterized by comprising the following specific steps:
step 1: acquiring a scene image, inputting the scene image into a pre-trained vehicle target detection model, outputting the positioning position coordinate information of a vehicle target,
step 2: and cutting the scene image according to the coordinate information of the positioning position of the vehicle target to obtain the vehicle image.
4. A method for deep learning based vehicle multi-attribute recognition according to any of claims 1-3, wherein the vehicle image is pre-processed by:
and (4) scaling the vehicle images to a uniform size and carrying out normalization processing.
5. The system for vehicle multi-attribute recognition based on deep learning is characterized by comprising a vehicle target positioning module and a vehicle attribute recognition module,
the vehicle target positioning module acquires a vehicle image in a scene image, inputs the vehicle image into a multi-scale convolutional neural network of the vehicle attribute identification module, the vehicle attribute identification module extracts a feature vector of the vehicle image, acquires an attribute probability vector of a vehicle attribute according to the corresponding feature vector by using a multi-label classification method, and outputs the corresponding vehicle attribute of which the attribute probability vector is greater than a probability threshold value as an identification result.
6. The system of claim 5, wherein the vehicle target location module obtains an image of the vehicle in the scene image by:
and positioning the vehicle target in the scene image, and cutting the scene image according to the positioning information to obtain all the vehicle images.
7. The system of claim 6, wherein the vehicle target location module obtains the vehicle image from the scene image by the steps of:
step 1: acquiring a scene image, inputting the scene image into a pre-trained vehicle target detection model, outputting the positioning position coordinate information of a vehicle target,
step 2: and cutting the scene image according to the coordinate information of the positioning position of the vehicle target to obtain the vehicle image.
8. The system for deep learning-based vehicle multi-attribute recognition according to any one of claims 5-7, wherein the vehicle attribute recognition module preprocesses the vehicle image by:
and (4) scaling the vehicle images to a uniform size and carrying out normalization processing.
9. The device for vehicle multi-attribute recognition based on deep learning is characterized by comprising at least one memory and at least one processor;
the at least one memory to store a machine readable program;
the at least one processor is configured to invoke the machine readable program to perform the method for deep learning based vehicle multi-attribute recognition according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110703877.7A CN113269150A (en) | 2021-06-24 | 2021-06-24 | Vehicle multi-attribute identification system and method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110703877.7A CN113269150A (en) | 2021-06-24 | 2021-06-24 | Vehicle multi-attribute identification system and method based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113269150A true CN113269150A (en) | 2021-08-17 |
Family
ID=77235913
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110703877.7A Pending CN113269150A (en) | 2021-06-24 | 2021-06-24 | Vehicle multi-attribute identification system and method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113269150A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118053260A (en) * | 2024-04-02 | 2024-05-17 | 惠州市新益鸿科技有限公司 | Production workshop construction operation safety early warning system and method based on Internet of things |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107430693A (en) * | 2015-03-13 | 2017-12-01 | 北京市商汤科技开发有限公司 | For vehicle classification and the equipment and system of checking |
| CN109816001A (en) * | 2019-01-10 | 2019-05-28 | 高新兴科技集团股份有限公司 | A kind of more attribute recognition approaches of vehicle based on deep learning, device and equipment |
| CN111291715A (en) * | 2020-02-28 | 2020-06-16 | 安徽大学 | Vehicle type identification method based on multi-scale convolutional neural network, electronic device and storage medium |
| 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 |
-
2021
- 2021-06-24 CN CN202110703877.7A patent/CN113269150A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107430693A (en) * | 2015-03-13 | 2017-12-01 | 北京市商汤科技开发有限公司 | For vehicle classification and the equipment and system of checking |
| CN109816001A (en) * | 2019-01-10 | 2019-05-28 | 高新兴科技集团股份有限公司 | A kind of more attribute recognition approaches of vehicle based on deep learning, device and equipment |
| CN111291715A (en) * | 2020-02-28 | 2020-06-16 | 安徽大学 | Vehicle type identification method based on multi-scale convolutional neural network, electronic device and storage medium |
| 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 |
Non-Patent Citations (1)
| Title |
|---|
| 王亚敏: "基于多尺度卷积神经网络的车型识别", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118053260A (en) * | 2024-04-02 | 2024-05-17 | 惠州市新益鸿科技有限公司 | Production workshop construction operation safety early warning system and method based on Internet of things |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3564854B1 (en) | Facial expression recognition method, apparatus, electronic device, and storage medium | |
| CN110705405B (en) | Target labeling method and device | |
| CN110197146B (en) | Face image analysis method based on deep learning, electronic device and storage medium | |
| CN112949578B (en) | Vehicle lamp state identification method, device, equipment and storage medium | |
| CN112613387A (en) | Traffic sign detection method based on YOLOv3 | |
| CN110599453A (en) | Panel defect detection method and device based on image fusion and equipment terminal | |
| CN112328822B (en) | Picture pre-marking method and device and terminal equipment | |
| CN111191582A (en) | Three-dimensional target detection method, detection device, terminal device and computer-readable storage medium | |
| CN113255444A (en) | Training method of image recognition model, image recognition method and device | |
| CN113902740A (en) | Construction method of image blurring degree evaluation model | |
| CN112966687B (en) | Image segmentation model training method and device and communication equipment | |
| CN114119695A (en) | Image annotation method and device and electronic equipment | |
| CN113269150A (en) | Vehicle multi-attribute identification system and method based on deep learning | |
| CN112784675B (en) | Target detection method and device, storage medium and terminal | |
| CN113888496A (en) | Sperm morphology recognition method based on deep convolutional neural network and storage medium | |
| CN113592807A (en) | Training method, image quality determination method and device, and electronic equipment | |
| US11887292B1 (en) | Two-step anti-fraud vehicle insurance image collecting and quality testing method, system and device | |
| CN115631197B (en) | Image processing method, device, medium, equipment and system | |
| CN111738264A (en) | Intelligent acquisition method for data of display panel of machine room equipment | |
| CN114882469B (en) | Traffic sign detection method and system based on DL-SSD model | |
| CN110610177A (en) | Training method of character recognition model, character recognition method and device | |
| CN112861861B (en) | Method and device for recognizing nixie tube text and electronic equipment | |
| JP2024516642A (en) | Behavior detection method, electronic device and computer-readable storage medium | |
| Das et al. | Object Detection on Scene Images: A Novel Approach | |
| CN112819953A (en) | Three-dimensional reconstruction method, network model training method and device and electronic equipment |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210817 |