License plate detection and recognition method and system in complex scene
A technology for license plate detection and complex scenes, applied in the direction of neural learning methods, character and pattern recognition, instruments, etc., can solve the problems of wide range and poor scalability, poor performance, poor recognition accuracy, etc., to improve the accuracy of license plate recognition, improve accuracy and Time-sensitive, practical effect
Inactive Publication Date: 2021-11-05
北京睿芯高通量科技有限公司
View PDF6 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] 2. Coarse positioning of license plate area
[0010] 1. This method is extensive and has poor scalability. It only supports the detection and recognition of frontal vehicle images, and cannot be applied to multi-angle vehicle detection.
[0011] 2. This method has relatively strict requirements on image quality, and external objective factors such as light intensity and shooting angle have a greater impact on the corresponding method, and the performance in complex scenes is poor
[0012] 3. Character recognition is heavily dependent on character segmentation accuracy, and character segmentation accuracy depends on image quality, resulting in poor recognition accuracy
[0019] 1. In the existing license plate detection and recognition, the target detection method such as the YOLO framework is generally used for the license plate location. The disadvantage is that the YOLO framework can only generate a rectangular frame. Although it is generally corrected, the detection effect for the license plate with a large degree of inclination Not good, it directly affects the accuracy of license plate character segmentation and license plate character recognition
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
[0083] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the figures in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention. Moreover, it needs to be further emphasized here that the preferred technical solutions provided by the following specific examples can be used in cooperation or in combination with each other.
[0084] like figure 1 As shown, the present invention provides a flow chart of a license plate detection and recognition method in a complex scene based on deep learning technology, and the method includes the following steps:
[0085] Step S1: collect the vehicle video stream, and ext...
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 relates to a license plate detection and recognition method and system in a complex scene, and the method comprises the following steps: collecting a vehicle video stream, and extracting image data from the video stream; constructing a vehicle local positioning module, and taking the picture data as training data of the vehicle local positioning module to obtain a license plate local positioning model; performing picture correction according to the obtained license plate corner point information and external rectangle information; constructing a license plate recognition module, and taking the license plate picture as training data of a license plate recognition model to obtain the license plate recognition model; and inputting the corrected picture as a license plate recognition module to obtain a picture recognition result. The invention has relatively high practical applicability.
Description
technical field [0001] The invention relates to the technical field of license plate detection, in particular to a license plate detection and recognition method and system based on deep learning technology in complex scenarios. Background technique [0002] Traditional license plate detection methods generally use methods such as OpenCV detection. In complex scenes, the accuracy is low and the robustness is poor, and it cannot achieve the effect of practical application. However, the existing depth-based license plate detection technology generally passes through three steps, including License plate location, license plate character segmentation and character recognition, this method has poor timeliness and low detection performance for complex scenes. [0003] Prior Art 1 discloses a license plate detection method based on OpenCV, which is mainly divided into the following steps: [0004] 1. Vehicle detection. By reading pictures or videos, based on the background differ...
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 & AuthorityApplications(China)
IPC IPC(8): G06K9/00G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06N3/045G06F18/253
Inventor林海华琚午阳
Owner北京睿芯高通量科技有限公司