Real-time defogging method for surveillance video based on improved dehazenet
A monitoring video, real-time technology, applied in image enhancement, image analysis, instruments, etc., can solve the problems of slow speed and low accuracy
Inactive Publication Date: 2021-06-04
大象智能科技(南京)有限公司
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0007] In order to solve the problems of low precision and slow speed in the traditional single image algorithm, combined with the convolutional neural network and parallel computing structure, the overall scheme design of the new video defogging algorithm is carried out
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
[0089] Below in conjunction with accompanying drawing and specific embodiment the case is further described:
[0090] (1) Overall Design of Dehazing Method
[0091] Based on the atmospheric scattering model, video dehazing is to estimate the transmittance and atmospheric light constant to solve the original image when the fog image is known. The present invention divides the whole real-time defogging process into three parts (the overall block diagram is as follows: figure 2 shown):
[0092] The first part is to capture the video through the video acquisition device, and then cut the video frame by frame into a single picture, which is provided to the neural network for processing.
[0093] The second part is to use the trained neural network, that is, the weights of each layer are known, and the input image is divided into blocks to obtain the transmittance and atmospheric light constant of each part, and finally the transmittance distribution map and atmospheric light con...
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
A real-time defogging method for surveillance video based on improved DehazeNet. The steps include: 1) collecting video through a video acquisition device, then cutting the video frame by frame into a single picture, and providing it to the neural network for processing; 2) using the trained The improved DehazeNet neural network, that is, the weights of each layer are known, and the input picture is processed in blocks to obtain the transmittance t(x) and atmospheric light constant of each part, and finally form a transmittance distribution map and atmospheric light constant distribution Fig. 3) Obtain the output of the neural network, solve the fog-free image according to the dehazing algorithm based on the atmospheric scattering model, and then re-splice the fog-free image into a video. Compared with traditional single image defogging, the present invention realizes real-time defogging of video, ensures the effect of defogging at the same time, and solves the problems of oversaturation and blurred skyline existing in traditional defogging methods.
Description
technical field [0001] The invention belongs to the application technology of computer technology in video processing, in particular to a real-time defogging method for monitoring video based on improved DehazeNet. Background technique [0002] Fog is a common atmospheric phenomenon. The ubiquity of dust, smoke or other particles in the air reduces the clarity of the atmosphere. Since light is scattered by particles in the atmosphere, the visual contrast of objects is reduced when imaging. Because of this, fog often causes a lot of problems for photographic imaging. The existence of fog essentially changes the atmospheric transmittance, and the contrast and color of the outdoor scene image will be changed, making many features contained in the image covered or blurred, resulting in the inability of video surveillance products to collect clear on-site images. It has a serious impact on the security of important places in the city. [0003] The transmittance of the fog in ...
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): G06T5/00G06T3/40G06N3/04
CPCG06T3/4038G06T2207/30232G06T2207/10016G06N3/045G06T5/73
Inventor 陈天悦
Owner 大象智能科技(南京)有限公司