A Vehicle Component Location Method for Vehicle Fine Recognition
A vehicle component and fine recognition technology, applied in the field of pattern recognition, can solve problems such as limited application range
Active Publication Date: 2017-12-29
WUHAN UNIV
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] The existing vehicle type identification method based on coarse-grained classification uses the overall feature difference of the vehicle to classify, which can only identify basic types (such as cars, buses, trucks, etc.), and its application range is limited
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0045] In order to facilitate those of ordinary skill in the art to understand and implement the present invention, the present invention will be described in further detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the implementation examples described here are only used to illustrate and explain the present invention, and are not intended to limit this invention.
[0046] The present invention provides a two-stage vehicle component positioning method for vehicle fine recognition, model training and component precise positioning;
[0047] please see figure 1 , the model training phase of the present embodiment, its specific implementation includes the following sub-steps:
[0048] Step 1.1: Preparation of training data set. The training data set includes positive sample data and negative sample data; positive sample data is the image data of marked vehicle frame and component frame, and negative sample data is image dat...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a vehicle component positioning method oriented to vehicle fine recognition, and obtains accurate vehicle component positioning results through training models and component detection. Model training is to extract HOG features from the training data set of marked vehicles and vehicle parts, and use the SVM classifier to train to obtain model parameters, and finally obtain a model that can be used to detect vehicle parts. Component detection is to extract the feature pyramid of the test image, use the obtained model to perform convolution operation on the feature pyramid, obtain the convolution response value, and combine the deformation constraints to obtain the rough positioning of the vehicle components, and then combine the rigid body characteristics to add position constraints and symmetry constraints to achieve precise positioning of vehicle components. The present invention can realize the precise positioning of the parts of the vehicle in the picture, thereby effectively solving the problems mentioned in the background technology, and is the basic work of fine vehicle recognition.
Description
technical field [0001] The invention belongs to the technical field of pattern recognition, in particular to a vehicle component positioning method oriented to fine vehicle recognition. Background technique [0002] With the continuous increase of vehicle ownership, the widespread installation of high-definition surveillance cameras on roads and the widespread use of personal smartphones, the number of various types of vehicle-related pictures has multiplied exponentially. Because of the huge amount of data and the limited ability of human identification, it is difficult to realize the information such as vehicle model and other information from the pictures manually. Therefore, how to use computer technology to automatically identify the vehicle brand and model has developed into an important research direction. It will play an important supporting role in the fields of intelligent transportation, social public safety and so on. [0003] The research on automatic identific...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/62
Inventor 胡瑞敏肖骏肖晶
Owner WUHAN UNIV