A method of background subtraction and foreground detection
A technology of background difference and foreground, applied in the field of computer vision, can solve the problem of low model accuracy, achieve the effect of improving accuracy, eliminating noise interference, and improving computing efficiency
Active Publication Date: 2019-07-23
WUHAN UNIV
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0023] Aiming at the problem of low accuracy of the existing models based on statistical methods, the present invention proposes a new Methods of background subtraction and foreground detection
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
[0033] In order to facilitate those of ordinary skill in the art to understand and implement the present invention, the present invention will be described in further detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the implementation examples described here are only used to illustrate and explain the present invention, and are not intended to limit this invention.
[0034] please see figure 1 , a method of background difference and foreground detection provided by the present invention, constructing a Minoan vector for saving reference data and statistical information of each pixel; defining the dynamics of each position of the background image, setting the dynamic level series , according to the dynamic class division rules, the dynamics are assigned to a certain dynamic class; the dynamic KDE algorithm is used to obtain the optimal bandwidth; an adaptive threshold is designed for each pixel, and the value of the thresh...
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 background difference and foreground detection method. A Mino vector is constructed, and is used for storing reference data and statistic information of every pixel. A dynamic feature of every position of a background image is defined, and a dynamic grade number is set, and the dynamic feature belongs to one dynamic grade according to a dynamic grade division rule. An optical bandwidth is acquired by adopting a dynamic KDE algorithm. A self-adapting threshold value is designed for every pixel, and the valuing of the threshold value is related to the dynamic feature of every pixel. By adopting a Tetris updating scheme, the bottom layer of the reference data of the Mino vector is eliminated, and therefore noise interferences are eliminated, and the quality of the reference data is improved. The validity of the novel detection method in foreground detection application is proved by an experimental result, and by comparing with the novel method, one of three indexes is better than every other method, and the other two indexes are slightly inferior to the novel method.
Description
technical field [0001] The invention belongs to the technical field of computer vision, in particular to a method for accurate and efficient background difference and foreground detection modeling. Background technique [0002] In computer vision applications, an accurate background subtraction model is a fundamental and important step for foreground extraction. Typically, an initial background model is generated by assembling and studying a series of consecutive images before observing and evaluating an image to be classified. However, this method of selecting a fixed reference window may be effective for static backgrounds, but neglects the research on dynamic backgrounds with a dynamic range. For the size of the fixed reference window, setting an upper limit can avoid missing the dynamic background, but it will increase the calculation burden, especially with complex calculation components (such as Gaussian). Through the analysis of the dynamic background, it is found t...
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): G06T7/00G06T7/194
CPCG06T7/0002
Inventor 袁志勇张贵安童倩倩
Owner WUHAN UNIV