Single Image Rain Noise Removal Method and Device Based on Double Fidelity Term Regular Model Based on Wavelet Transform
A technology of wavelet transform and single image, which is applied in image enhancement, image data processing, instruments, etc., can solve the problems of complicated image deraining algorithm, low real-time performance of the algorithm, residual rain marks in the image, etc., and achieve the calculation time Short, small amount of calculation, and the effect of improving performance
Active Publication Date: 2019-10-01
WENZHOU MEDICAL UNIV
View PDF10 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The rain removal method for a single image has achieved some results, but there are still some problems, mainly as follows: 1) After the rain is removed from the image, the rain marks still remain or the image becomes blurred after the rain is removed; Complex, so that the real-time performance of the algorithm is not high
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
[0035] In order to illustrate the content of the present invention more clearly, specific examples are used to illustrate the following, and the specific examples do not limit the scope of the content of the present invention.
[0036] The hardware environment used for implementation is: CPU: Intel(R) Core(TM) i7-5820K3.30GHz; graphics card: NvidiaTitan X GPU 1G video memory; memory: 4G; hard disk: Seagate 500G, and the running software environment is: MATLAB(R2015b ).
[0037] The basic process of the present invention is attached figure 1 As shown, the specific implementation is as follows:
[0038] 1. Wavelet transform: Perform wavelet transform operation on images containing rain noise. Through wavelet transform, an approximate component and three detail components can be obtained, namely horizontal, vertical and diagonal detail components. denoted as A 1 ,H 1 ,V 1 ,D 1 . The rain noise information is contained in the vertical detail component and the approximation ...
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 single image rain noise removal method and device based on a dual-fidelity item regular model of wavelet transformation, and the method comprises the steps: carrying out thewavelet transformation of an image containing rain noise, enabling the rain noise information to be usually contained in a vertical detail component and an approximate component at different resolutions, and achieving the separation of the rain noise; carrying out rain noise suppression processing on the component containing the rain noise by adopting a double-fidelity term single variation regularization model. Compared with an existing method, the method has the advantages that the numerical algorithm is achieved through an augmented Lagrangian algorithm, the advantages of double fidelity items and one-way variation are considered, meanwhile, a proposed model is combined with wavelet transform, rain noise is well removed during rain noise suppression processing, and more details of an image are reserved. Experiments show that the method is short in calculation time and small in calculation amount, the visual quality can be obviously improved when the rain removing method is applied to images affected by rain weather, and the performance of many subsequent computer visual algorithms can be improved.
Description
technical field [0001] The invention relates to the technical field of image noise processing, in particular to a method and device for removing rain noise from a single image based on a double-fidelity regular model of wavelet transform. Background technique [0002] With the rapid development of artificial intelligence technology, the application of outdoor vision systems is becoming more and more extensive. It plays an important role in target recognition and tracking, traffic monitoring, remote sensing monitoring, military monitoring and other fields. However, the robustness and practicality of an outdoor vision system will suffer greatly under harsh weather conditions. Rain is the most common dynamic weather. In the imaging system, raindrops are mainly displayed in the form of rain streaks, which greatly affects the visual effect of the image. Therefore, it is also of great value to study how to restore high-quality images from images containing rain noise. [0003] ...
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): G06T5/00
CPCG06T2207/20064G06T5/73G06T5/70
Inventor 张鑫陈伟斌
Owner WENZHOU MEDICAL UNIV