Vehicle logo positioning method based on multi-scale target detection
A technology for vehicle logo positioning and target detection, which is applied in neural learning methods, image data processing, image enhancement, etc.
Inactive Publication Date: 2021-09-10
ZHEJIANG SHUREN UNIV
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
For traditional methods, if you want to locate small targets in large-scale images, you need to build a model that takes up a lot of memory or video memory, and the accuracy is difficult to guarantee
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
[0037] The technical solutions provided by the present invention will be further described below with reference to the accompanying drawings.
[0038] The technical solution of the present invention belongs to the object detection technology, and its main task is to find out all objects in the image and determine their categories and positions, which is one of the core problems of computer vision. There are currently four major categories of tasks in computer vision, namely: classification, localization, detection, and segmentation. Vehicle logo detection is one of the applications of target detection in the field of transportation. It is mostly used to locate the position of the vehicle logo in the road monitoring screen, which is of great significance for proofreading vehicle registration information. Due to the location of road monitoring, the captured images often have the characteristics of large-scale and high-resolution. Because the vehicle is far away from the camera, ...
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 logo positioning method based on multi-scale target detection. The vehicle logo positioning method comprises the following steps of S1, acquiring a vehicle image and performing preprocessing; S2, inputting the preprocessed vehicle image into an image processing unit; and S3, after the image processing unit carries out image processing, outputting the position coordinates and the size of the vehicle logo, wherein in the step S2, the image processing unit at least comprises a vehicle logo positioning unit, a position area cutting unit and an output unit, and the vehicle logo positioning unit is a pre-trained deep neural network model. Compared with the prior art, based on the relative position relation of the vehicle license plate and the vehicle logo, a relatively large target, namely the vehicle license plate, is positioned from a large-scale image, then the image is cut according to the obtained position of the vehicle license plate, the position area of the vehicle logo is obtained, and the area is the small-scale image containing the vehicle license plate and the vehicle logo, and the obtained small-scale image into the model again is input to obtain the accurate position and size of the vehicle logo.
Description
technical field [0001] The invention belongs to the field of target detection and vehicle logo positioning, relates to a deep neural network-based target detection and Bounding Box Regression-based precise vehicle logo positioning technology, and in particular relates to a vehicle logo positioning method based on multi-scale target detection. Background technique [0002] With the advancement of technology, cars have become the most common means of transportation. However, with the increasing number of vehicles, traffic management is becoming increasingly complex. At present, there are a large number of cameras on urban roads and highways. The traffic management system can quickly capture the accident vehicles, illegal vehicles, blacklisted vehicles, etc. by obtaining monitoring image information, which is not only for traffic planning, traffic management, and road maintenance departments It provides important basic and operational data, and at the same time provides import...
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 Applications(China)
IPC IPC(8): G06K9/38G06K9/42G06K9/46G06K9/62G06N3/04G06N3/08G06T7/11G06T7/73
CPCG06N3/084G06T7/11G06T7/73G06T2207/10004G06T2207/20076G06T2207/20081G06T2207/20084G06T2207/20132G06N3/047G06N3/048G06N3/045G06F18/241G06F18/2415
Inventor 吴建锋吴尚明郑鹏勇蒋燕君尉理哲徐振宇阮越叶芳芳江俊许森王金铭吕何新王章权
Owner ZHEJIANG SHUREN UNIV