Unlock instant, AI-driven research and patent intelligence for your innovation.

Vehicle and licence plate detection method based on deep learning in vehicle-mounted environment

A technology of license plate detection and vehicle detection, applied in the field of deep neural network

Inactive Publication Date: 2018-11-16
珠海亿智电子科技有限公司
View PDF5 Cites 23 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The key technical problem to be solved by the present invention is to solve the challenging vehicle and license plate detection problem in complex natural scenes

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Vehicle and licence plate detection method based on deep learning in vehicle-mounted environment
  • Vehicle and licence plate detection method based on deep learning in vehicle-mounted environment
  • Vehicle and licence plate detection method based on deep learning in vehicle-mounted environment

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 specific implementation manners of the identification method and system of the vehicle type according to the embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0040] We propose an end-to-end network to simultaneously detect vehicles and license plates, where two separate branches with different convolutional layers are designed for vehicles and license plates, respectively. We use low-level features to detect license plates, and use high-level features to locate vehicles. In order to improve the recall rate of the detection system, we add an attention mechanism to strengthen each detection branch. Besides that, we apply several task-specific strategies for real-t...

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

No PUM Login to View More

Abstract

The invention uses a SSD as a basic network for carrying out improvement and discloses an end-to-end network for simultaneously detecting a vehicle and a licence plate in a given image. The network has two independent branches and in the two branches, different convolutional layers are respectively designed for detection of the vehicle and the licence plate. Finally, a network is improved again and designed on the basis of a strategy of the SSD, and a video collected from a real scene is used for verifying validity of the network. Meanwhile, a licence plate data reinforcement method which cannot invade privacy is proposed, which has great significance for further research. In addition, an attention mechanism is used for improving a recall rate of a detection system, and experiments prove that the attention mechanism can improve performance of target detection. Moreover, a plurality of special strategies for a task are adopted for detection of the vehicle and the licence plate so as toimplement real-time and efficient detection in the video.

Description

technical field [0001] The invention relates to a target detection method in the field of computer vision, and mainly proposes an end-to-end deep neural network, which is improved based on SSD, and is used to simultaneously detect vehicles and license plates in a given image. Background technique [0002] Automatic vehicle and license plate detection are important components of intelligent transportation systems. It can be applied in many applications such as traffic monitoring and management. So far, it has attracted considerable research attention and many methods have been proposed. Due to the rapid development of machine learning (such as CNN and other deep neural networks) in recent years, this method has shown good performance through attempts and improvements in many fields. However, since there are many different targets in road images of natural scenes, they will still be greatly disturbed, and certain measures need to be taken to reduce false detections and impro...

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
no application Login to View More
IPC IPC(8): G06K9/00G06K9/34G06K9/62
CPCG06V20/584G06V30/153G06V2201/08G06V20/625G06F18/214
Inventor 殷绪成任红萍陈松路杨春
Owner 珠海亿智电子科技有限公司