Super-resolution reconstruction method and device for video stream
A super-resolution reconstruction and super-resolution technology, which is applied in image analysis, image enhancement, instruments, etc., can solve problems such as difficult to achieve real-time performance, singleness, and poor processing effect
Active Publication Date: 2020-08-04
北京集朗半导体科技有限公司
View PDF22 Cites 4 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004]
However, the inventor found in the process of implementing the present invention that the existing technology uses a single technology for super-resolution reconstruction for each video frame image of the video stream, and this processing method is ineffective or difficult to meet the real-time requirements
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 approach
[0148] In one embodiment, according to the motion state of the image block corresponding to the pixel in the video frame image and the marker value of the texture richness, the weight of the zoom result of the pixel point and the superstructure corresponding to the video frame image are determined. The weight of the super-resolution reconstruction result of the pixel in the resolution reconstruction result. The implementation of fusion processing is as follows:
[0149] For any pixel in the enlarged image of the video frame image, make the pix_SR_flag of each pixel equal to the SR_flag corresponding to the image block in the video frame image, and then each pixel in the KxL window size of the pixel The pix_SR_flag of the point is averaged as the weight w of the super-resolution reconstruction result of the pixel in the super-resolution reconstruction result corresponding to the video frame image. As mentioned above, SR_flag represents the motion state and texture richness of ...
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 super-resolution reconstruction method and device for a video stream, and the method comprises the steps: dividing a target video frame image in a video stream into a plurality of image blocks, and detecting the motion states of the image blocks; counting the proportion of the image blocks in the target video frame image, wherein the motion state of the image blocks is astatic state or slow motion; determining a reconstruction frame rate according to the proportion, and determining each to-be-reconstructed video frame image according to the reconstruction frame rate;based on a preset super-resolution network model, performing super-resolution reconstruction on each to-be-reconstructed video frame image to obtain a super-resolution reconstruction result of each to-be-reconstructed video frame image; amplifying the video stream frame by frame by adopting a preset scaling algorithm to obtain an amplification result of each video frame image of the video stream;and fusing the super-resolution reconstruction result of each to-be-reconstructed video frame image and the amplification result of each video frame image to obtain a super-resolution reconstructionresult of the video stream. According to the method, super-resolution reconstruction of the video stream is realized.
Description
technical field [0001] The invention relates to the technical field of image processing, in particular to a method and device for super-resolution reconstruction of video streams. Background technique [0002] Super-resolution algorithm refers to recovering high-resolution images from low-resolution images. A high-resolution image has a higher pixel density, which provides more detail. Single-image super-resolution algorithms are mainly divided into three categories: interpolation-based methods, reconstruction-based methods, and learning-based methods. [0003] At present, the more popular one is the deep neural network-based method in the learning-based method. In the super-resolution method based on deep neural network, CNN (Convolutional Neural Network, Convolutional Neural Network) is mainly used. If the resolution of the input image is MxN, after CNN processing, an image enlarged by s times is obtained. From the 3-layer super-resolution reconstruction convolutional ...
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): G06T3/40G06T5/50G06T7/20
CPCG06T3/4053G06T5/50G06T7/20G06T2207/10016G06T2207/20021
Inventor 韩睿郭若杉李晨庄重李瑞玲刘壮
Owner 北京集朗半导体科技有限公司