Unlock instant, AI-driven research and patent intelligence for your innovation.
License plate recognition method based on lightweight neural network
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A neural network and license plate recognition technology, which is applied in the field of traffic monitoring and pattern recognition technology, can solve the problems of speed influence, accuracy and generalization limitation, etc.
Active Publication Date: 2020-05-15
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF9 Cites 0 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
This method is divided into two steps of character area segmentation and character classification. It is not an end-to-end method, and the speed will be affected accordingly.
And the segmentation based on the license plate character area is based on the traditional algorithm, and the accuracy and generalization ability have certain limitations.
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
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0064] The present invention designs a license plate recognition method based on a lightweight neural network, adopts the CTC loss function to train a lightweight neural network model, realizes end-to-end training without segmentation, and effectively solves the situation that the input and output are not aligned, especially The following settings are adopted: including the preparation of data sets, the construction of lightweight neural network models and the recognition of license plates,
[0065] The preparation of described data forms the license plate picture with the license plate number as the file name, specifically:
[0066] First, obtain video screenshots from the road video, form a sample image, and save it under the images folder; preferably, collect several videos of driving vehicles from the monitoring probe, and save a picture per second, for a total of Collect more than 30,000 pictures (sample images) and save them in the images folder, where the sample images ...
Embodiment 2
[0080] This embodiment is further optimized on the basis of the above embodiments, and the same parts as the aforementioned technical solutions will not be repeated here. Further, in order to better realize the present invention, the following setting methods are specially adopted: the steps 1) specifically It is: input the matrix with dimension [32,3,24,92] into the first two-dimensional convolutional layer, after normalization operation (batch normalization) and non-linear activation (relu), the dimension is [32,64,22] ,92] of the first feature matrix x1; the first two-dimensional convolution layer is provided with 64 convolution kernels with a step size of 1 and a size of 3×3.
Embodiment 3
[0082] This embodiment is further optimized on the basis of any of the above-mentioned embodiments, and the same parts as the aforementioned technical solutions will not be repeated here. Further, in order to better realize the present invention, the following setting methods are adopted in particular: the step 2 )Specifically:
[0083] 2.1) Input the first feature matrix x1 to the pooling layer to obtain the feature matrix of [32, 32, 20, 90]; in the step 2.1), the pooling area size is (1, 3, 3), and the step size is (1, 1, 1) pooling layer;
[0084] 2.2) Input the feature matrix obtained in step 2.1) into a special convolutional layer (named small_basic_block), and the output result is obtained after normalization operation (batch normalization) and nonlinear activation (relu), and the dimension is [32,128,20,90 ] of the second eigenmatrix x2.
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 license plate recognition method based on a lightweight neural network, and the method comprises the steps: preparing a data set, constructing a lightweight neural network model, and recognizing a license plate, and the construction of the lightweight neural network model comprises the following steps: 1) forming a first feature matrix x1; 2) forming a second characteristic matrix x2; 3) forming a third feature matrix x3; 4) forming a fourth characteristic matrix x4; 5) fusing channels of the first feature matrix x1, the second feature matrix x2, the third feature matrix x3 and the fourth feature matrix x4, and then obtaining a feature vector through a convolution layer of which the convolutionkernel size is 1 * 1 and the step length is 1; and 6) completing lightweight neural network model training based on the CTC loss function, performing lightweight neural network model training by adopting the CTC loss function, realizing end-to-end training without segmentation, and effectively solving the problem of input and output misalignment.
Description
technical field [0001] The invention relates to the fields of pattern recognition technology, traffic monitoring technology, etc., specifically, a license plate recognition method based on a lightweight neural network. Background technique [0002] License plate recognition is a crucial link in the modern intelligent transportation system, and it is the core function of the intelligent transportation system. It uses technical means to extract the license plate number of the vehicle to obtain vehicle information. In my country, motor vehicle license plates need to be registered with the traffic management department, and there is a one-to-one relationship between license plates and vehicles. Therefore, information acquisition, management and monitoring of motor vehicles can be performed based on motor vehicle license plates. In this case, a license plate recognition system is needed to collect license plate number information efficiently and accurately. [0003] The automati...
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
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to View More 
Login to View More