MRF license plate denoising algorithm based on gradient information and block region combination prior
A gradient prior and gradient information technology, applied in computing, computer components, character and pattern recognition, etc., can solve the problems of single noise components and difficulty in adapting to complex noise components, and achieve the effect of improving the signal-to-noise ratio
Active Publication Date: 2017-02-22
NAT UNIV OF DEFENSE TECH
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Existing denoising algorithms often can only solve the situation where the noise component is relatively single, and it is difficult to adapt to the situation where the noise component is complex in low-light environments
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
[0052] The present invention will be further described below in conjunction with drawings and embodiments.
[0053] The invention provides an MRF license plate denoising algorithm based on gradient information and block region combination prior, which can effectively denoise and restore license plate images in noise video sequences. Such as figure 1 Shown, be flow chart of the present invention, comprise the steps:
[0054] Step 1: Using the richness of license plate digital gradient information and the characteristics of blocky digital areas, construct a license plate image combination prior based on gradients and blocky areas;
[0055]1. License plate gradient information
[0056] Let I v is the denoised image of the license plate in the form of vector, and the numbers of rows and columns of the image pixels are m and n respectively. define I' c , I' r are the matrix of partial derivatives of pixel value changes in the column direction and the row direction, respective...
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 belongs to the field of image information processing, and in particular relates to an MRF license plate denoising algorithm based on gradient information and block region combination priors. The specific implementation includes the following steps: S1, using the digital gradient information of the license plate, constructing the prior information of the license plate image, including the gradient prior and the block area prior; S2, performing noise estimation on the video image sequence, and determining the probability density function of the noise distribution; S3 , Construct the binary Markov random field model of the license plate image, combine the Markov random field model and the prior information of the license plate image, establish the optimization problem model and solve it. Compared with the traditional license plate denoising method, this method can make full use of the characteristics of the license plate image, effectively improve the signal-to-noise ratio of the license plate image, obtain more accurate license plate digital edges, and provide higher quality license plate source images for license plate information extraction and analysis.
Description
technical field [0001] The invention belongs to the field of image information processing, and in particular relates to an MRF license plate denoising algorithm based on gradient information and block region combination priors. Background technique [0002] License plate image information is of great significance in the fields of security monitoring, intelligent transportation and social security, and plays a very important role in vehicle speeding supervision, car anti-theft, road and parking lot charges, and traffic accident monitoring. However, the license plate images collected under low illumination conditions often have the characteristics of low signal-to-noise ratio, high noise, and complex noise components, which have a great impact on the extraction and analysis of license plate information, resulting in a decline in the accuracy of license plate recognition and tracking. Therefore, it has important practical significance and practical application value to effectiv...
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): G06K9/40G06T5/00
CPCG06V10/457G06V10/267G06V20/625
Inventor 刘煜尹晓晴王炜徐玮熊志辉
Owner NAT UNIV OF DEFENSE TECH