Video foreground image extraction method and device
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A foreground image and foreground pixel technology, applied in the field of computer vision, can solve problems such as difficulty in distinguishing foreground images and background images, and achieve the effect of sensitive judgment.
Active Publication Date: 2019-09-20
RUIJIE NETWORKS CO LTD
View PDF4 Cites 0 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0005] Embodiments of the present invention provide a video foreground image extraction method and device, which are used to solve the problem that it is difficult to distinguish the foreground image from the background image using the GMM model when the image foreground and image background gray levels are similar
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 more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0038] The invention provides a method for extracting video foreground images, referring to figure 2 As shown in , the method includes:
[0039] S101. Extract foreground pixels from a current frame image of a video stream according to a grayscale image-based GMM model to obtain a first set of pixels.
[0040] Specifically, refer to image 3 As shown in , step S101 includes steps S1011-S1014:
[0041] S1011. Convert the current frame image from an RGB image to a grayscale image.
[0042] S1012. Extract foreground pixels from the above grayscale image according to the GMM model.
[0043] The GMM model based on grayscale images is as follows:
[0044] Let I(x, y, t) represent the pixel gray value of the pixel point (x, y) at time t, then we have
[0045]
[0046] Among them, K is the number of Gaussian distribution, which is called the mixing coefficient of Gaussian mixture probability density; is the weighting coefficient of the i-th Gaussian component at time t, that...
Embodiment 2
[0090] Embodiments of the present invention provide a video foreground image extraction device 10, which is applied to the above video foreground image extraction method, referring to Figure 5 As shown, the device includes:
[0091] The extraction unit 101 is used to extract the foreground pixel points of the current frame image according to the GMM model based on the grayscale image to obtain the first set of pixel points;
[0092] The conversion unit 102 is used to determine the background reference image of the video stream, and convert the background reference image from the RGB color space to the HSV color space to obtain the background HSV image;
[0093] The conversion unit 102 is further configured to convert the current frame image of the video stream from the RGB color space to the HSV color space to obtain the current frame HSV image;
[0094] The calculation unit 103 is used to calculate the Euclidean distance of the HSV values of the pixels in the current fram...
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 video foreground image extraction method and device, relates to the field of computer vision, and is used for solving the problem that image segmentation is difficult to perform by using a Gaussian Mixture Model (GMM) when the image foreground is similar to the background grayscale of the image. The method comprises the steps that foreground pixel points are extracted from the present frame of image according to the GMM so that a first pixel point set is obtained; the background reference image of a video stream is determined, and the image is transformed to an HSV color space from an RGB color space so that a background HSV image is obtained; the present frame of image is transformed to the HSV color space from the RGB color space so that the present frame of HSV image is obtained; the Euclidean distance of the HSV value of the pixel points of the same coordinates in the present frame of HSV image and the background HSV image on an HS plane is calculated, and a set formed by the pixel points locating in the same coordinates in the present frame of image acts as a second pixel point set if the Euclidean distance is greater than a first threshold; and the foreground image is obtained according to the first pixel point set and the second pixel point set. The video foreground image extraction method and device are applied to foreground image extraction.
Description
technical field [0001] The invention relates to the field of computer vision, in particular to a method and device for extracting video foreground images. Background technique [0002] In order to extract the foreground image in the video, the background must be removed first. There are three typical methods: optical flow method, frame difference method and background subtraction method. Among them, the background subtraction method is the most widely used video foreground extraction method, but this method is sensitive to illumination and has poor adaptability. Therefore, in response to this shortcoming of the background subtraction method, the researchers proposed a background modeling model and a segmentation strategy, that is, a background subtraction method based on a Gaussian mixture model (English full name: Gaussian mixture model, English abbreviation: GMM). [0003] However, this algorithm is usually carried out in the hardware-oriented RGB color space. When the m...
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
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to view more 
Login to view more