Mixed noise suppression algorithm suitable for unmanned aerial vehicle water surface target detection
A technology of mixing noise and water surface targets, applied in the field of image processing, which can solve problems with high requirements and high computational complexity
Pending Publication Date: 2021-04-23
HENAN POLYTECHNIC UNIV
View PDF2 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The noise reduction effect of the image noise reduction algorithm based on deep learning depends on the learned image data set, the structure of the deep network and the setting of the target loss function. airborne equipment, obviously such a requirement is too 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
[0038] The present invention will be described in detail below in conjunction with the accompanying drawings and implementation examples, but the scope of the present invention is not limited.
[0039] A hybrid noise suppression algorithm suitable for UAV surface target detection, comprising the following steps:
[0040] The first step: the local surface of the image can be developed;
[0041] Such as figure 1 As shown, the local expandable processing is performed on the mixed noise image, and the evolution of the local surface of the image to the expandable surface is to obtain 8 kinds of distances from the current point to the tangent plane {d i , the minimum absolute value of i=1,2,...8}|d m After |, according to U(i,j)=U(i,j)+d m Complete pixel point U i,j update, complete a U i,j The local extensibility of ; use the same method to update the pixels in other sub-domains to realize the extensibility of the entire image, suppress the salt and pepper noise and low-densit...
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 technical field of image processing, and particularly discloses a mixed noise suppression algorithm suitable for unmanned aerial vehicle water surface target detection, which comprises the following steps: firstly, suppressing salt-and-pepper noise and low-density Gaussian noise in an image by locally expanding the image to obtain a preliminary noise reduction image; dividing the preliminary noise reduction image into a high-contrast image and a low-contrast image, processing the high-contrast image in the preliminary noise reduction image by using bilateral filtering, and performing parallel filtering on the guided image and the denoised high-contrast image by using a bilateral filtering core; then, preparing wavelet contraction of a transform domain by extracting a low-contrast signal and performing short-time Fourier transform, processing a low-contrast image, and keeping details such as edges and textures of the image while Gaussian noise of the low-contrast image is suppressed; and finally, iteratively suppressing salt and pepper noise and Gaussian noise in the mixed noise image to obtain a final noise-reduced image.
Description
technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a mixed noise suppression algorithm suitable for the detection of unmanned aerial vehicle surface targets. Background technique [0002] With the continuous development of Internet of Things, big data, cloud computing and other technologies, more and more video-based application technologies have entered people's lives, such as face recognition systems, license plate recognition systems, text recognition systems and other target detection systems. In the field of target detection, the quality of acquired images is a key factor affecting the accuracy of target recognition. Compared with the target detection system with a relatively fixed imaging environment, the imaging environment of the UAV-based surface target detection system is volatile, and the collected images are easily affected by various types of noise, especially the mixed noise composed of Gaussian...
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 Applications(China)
IPC IPC(8): G06T5/00G06T5/50
CPCG06T5/50G06T2207/20028G06T2207/20056G06T5/70
Inventor 马凤颖王满利张长森
Owner HENAN POLYTECHNIC UNIV