An intelligent detection method of vehicle logo based on convolutional neural network
A convolutional neural network and vehicle logo detection technology, applied in neural learning methods, biological neural network models, neural architectures, etc., can solve problems such as poor robustness, low recognition accuracy, and low efficiency, and achieve good robustness Effect
Active Publication Date: 2021-09-03
WENZHOU UNIV
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] The technical problem to be solved by the embodiments of the present invention is to provide an intelligent detection method for car logos based on convolutional neural networks, which overcomes the problems of low efficiency and poor robustness of edge detection technology and traditional manual extraction of car logo features and the current existing The car logo detection method has low recognition accuracy and low positioning accuracy, so as to improve the recognition accuracy and positioning accuracy of the car logo detection model, and realize the intelligent detection of car logo with high robustness and high accuracy.
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
[0039] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings.
[0040] Such as figure 1 As shown, in the embodiment of the present invention, a kind of car logo intelligent detection method based on convolutional neural network is proposed, and the method includes the following steps:
[0041] Step S1, given the vehicle logo detection training set I train ={(a i ,b i )|i∈[1,M]} and vehicle logo detection test set I test ={(a i ,b i )|i∈[1,H]}; where, a i Represents the i-th input image containing vehicle logo information, its size is (3×K×K), 3 in (3×K×K) represents the number of color channels, and the corresponding color channel d∈{red, green, blue} , K in (3×K×K) represents the width or height of a single picture; b i Represents the corresponding car logo label in the i-th input image, and the i-th car logo label co...
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 provides an intelligent detection method for car logos based on convolutional neural networks. First, a car logo detection training set and a car logo detection test set are given, and a car logo detection model and an SGD optimizer are constructed and initialized; In each iterative calculation of the detection model, the input image in the vehicle logo detection training set is used as the input of the car logo detection model to obtain the car logo detection result, and according to the loss function, the loss value is calculated and backpropagated to the car logo detection model At the same time, the vehicle logo detection model is evaluated by using the car logo detection test set until the end of the iteration, and the final optimal network parameters are obtained to update the car logo detection model; finally, the input containing the car logo information to be tested is obtained The image input is calculated to obtain the vehicle logo detection result of the input image containing the vehicle logo information to be tested. By implementing the present invention, the recognition accuracy and positioning accuracy of the vehicle logo detection model are improved, and intelligent detection of the vehicle logo with high robustness and high accuracy is realized.
Description
technical field [0001] The invention relates to the technical field of intelligent detection of vehicle logos, in particular to a method for intelligent detection of vehicle logos based on a convolutional neural network. Background technique [0002] In the case of a large base of private cars in the world and the number is increasing year by year, intelligent transportation systems play an extremely important role. As a key information of the vehicle, the logo is not easy to be replaced and can become a distinctive feature of the vehicle. Therefore, the detection of vehicle logos is of great significance to help solve crimes and control vehicle detection. [0003] At present, most car logo detection algorithms use edge detection technology or use traditional manual design features. These methods are cumbersome and the detection robustness for car logos in different scenarios is not high enough. [0004] To solve this problem, there is an urgent need for an intelligent det...
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): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06V20/54G06V2201/08G06N3/045G06F18/241
Inventor 赵汉理卢望龙
Owner WENZHOU UNIV