Method and device for video image processing
A video image and variance technology, applied in the field of image processing, can solve problems such as loss of time series information, difficulty in real-time processing, lack of strategy guidance for spatiotemporal shape change characteristics, etc.
Active Publication Date: 2021-12-10
南京智谱科技有限公司
View PDF7 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The calculation of optical flow method is complicated, and it is difficult to realize real-time processing. Factors such as noise, multiple light sources, shadows and occlusions will seriously affect the calculation results of optical flow field distribution; video feature extraction methods based on deep learning include three-dimensional convolution method, dual-stream The three-dimensional convolution method realizes the capture of time and space information through 3D CNN, but its computational complexity is large, there are many network parameters, and it lacks strategic guidance on the characteristics of spatiotemporal shape changes
The two-stream convolution method uses two CNN networks, one to process spatial information and the other to process time domain information. Although the number of network model parameters is reduced compared to the three-dimensional convolution method, it relies on the optical flow extraction of the pre-video, and for long For video, temporal information is lost in feature learning
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
[0059] In the following description, references to "some embodiments" describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or a different subset of all possible embodiments, and Can be combined with each other without conflict.
[0060] In the following description, the terms "first\second\third, etc." or module A, module B, module C, etc. are only used to distinguish similar objects, or to distinguish different embodiments, not Representing a specific ordering of objects, it is understood that where permitted the specific order or sequence can be interchanged so that the embodiments of the invention described herein can be practiced in an order other than that illustrated or described herein.
[0061] In the following description, the involved reference numerals representing steps, such as S110, S120, etc., do not mean that this step must be executed, and the order of the previous and subsequent steps can be interc...
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 present invention provides a method and device for video image processing. The processing method includes: acquiring the current video image sequence; obtaining the current depth feature matrix according to the video image sequence; using a mixed Gaussian model according to the current depth feature matrix Obtain the current spatio-temporal morphological change feature matrix, wherein the time distribution of the depth features at any spatial position in the depth feature matrix is represented by using a mixed Gaussian model; video image processing is performed according to the current spatio-temporal morphological change feature matrix. The method and device of the invention have the advantages of small calculation amount and low model complexity when explicitly mining the characteristics of spatio-temporal form changes.
Description
technical field [0001] The present application relates to the field of image processing, in particular to a video image processing method and device. Background technique [0002] With the advent of the digital age, video, as the most widely used media form today, has gradually surpassed text and pictures, which makes video understanding even more important. Compared with images, videos have more one-dimensional timing information. How to make good use of timing information in video is a key issue worth studying. [0003] However, the traditional frame difference method is sensitive to environmental noise, and the selection of the threshold is quite critical. For relatively large moving objects with consistent colors, there may be holes inside the object, and the moving objects cannot be completely extracted. It only works when the camera is still. The calculation of optical flow method is complicated, and it is difficult to realize real-time processing. Factors such as n...
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): G06T7/246G06T7/55G06T7/73G06K9/62G06N3/04
CPCG06T7/246G06T7/55G06T7/73G06T2207/10016G06N3/045G06F18/22
Inventor 周凯来陈林森李昀谦祖永祥王远卓陈文龙李晗黄奥成张梦雅
Owner 南京智谱科技有限公司