Foreground detection method and system

A technology of foreground detection and foreground images, which is applied in image data processing, instruments, calculations, etc., can solve problems such as inability to eliminate false alarms, limited computing power of front-end camera processors, and weak contrast

Active Publication Date: 2015-06-10
SUZHOU KEDA TECH
View PDF6 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, the mixed Gaussian model recursively updates the background model according to the pixels of the current frame, which makes the errors in the modeling of the previous frame have a long-term impact on the background image
Moreover, the traditional mixed Gaussian background modeling cannot eliminate false alarms caused by rapid illumination changes, nor can it resist the noise impac

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
  • Foreground detection method and system
  • Foreground detection method and system
  • Foreground detection method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0080] Such as figure 1 As shown, this embodiment provides a foreground detection method, which specifically includes the following steps:

[0081] S11: Acquiring the current frame image.

[0082] S12: Calculate the local contrast of the current frame image. Specifically, the local contrast of the current frame image can be calculated in the following ways:

[0083] First, divide the current frame image into several m*n pixel blocks, where m and n are positive integers greater than 0;

[0084] Then, count the gray mean and gray variance of each pixel block;

[0085] Finally, the local contrast of each pixel block is obtained, and the local contrast is the quotient obtained by dividing the gray level variance of each pixel block by the gray level mean value.

[0086] In addition to using the above method to calculate the local contrast, the local contrast may also be calculated by other methods in the prior art.

[0087] S13: Referring to the relevant information of the pr...

Embodiment 2

[0102] Such as figure 2 As shown, this embodiment provides another foreground detection method. Compared with the above-mentioned embodiment 1, after the step of acquiring the current frame image and before calculating the local contrast of the current frame image, it also includes adaptive The process of noise processing to eliminate the impact of noise on foreground detection during imaging, the specific steps are as follows:

[0103] S101: Obtain the noise intensity of the current frame image;

[0104] S102: When the noise intensity is greater than the preset threshold, perform noise reduction processing on the current frame image. Specifically, a low-pass filter may be used for noise reduction processing, and an average filtering algorithm may be further selected for noise reduction.

[0105] In the foreground detection method provided in this embodiment, when the noise intensity of the current frame image is strong, that is, greater than a preset threshold, noise reduct...

Embodiment 3

[0120] Such as image 3 As shown, the present embodiment provides a foreground detection method, comprising the following steps:

[0121] S21: Acquiring the current frame image. Because the camera detects the foreground target in real time, it will acquire each frame of image collected in real time, and perform foreground detection on each frame of image to realize real-time tracking of the target.

[0122] S22: Calculate the noise intensity of the current frame image. Every time a frame of video image is acquired, the noise intensity of the frame image is calculated. The specific calculation process is as follows:

[0123] First, calculate the absolute value of the grayscale difference between the frame image and the previous frame image at each pixel point, which is denoted as D(x,y). If T0≤D(x,y)≤T1, then the pixel point (x ,y) is a noise point, otherwise it is not a noise point, wherein T0 and T1 are preset thresholds, T0

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

No PUM Login to view more

Abstract

The invention discloses a foreground detection method and a foreground detection system. The method comprises the following steps: obtaining the current frame image; calculating the partial contrast ratio of the current frame image; establishing a multi-gaussian background model based on the partial contrast ratio; detecting a portion of foreground image in the current frame image as foreground image sample according to the multi-gaussian background model; learning the background image according to the foreground image sample and the current frame image; performing foreground target detection for the current frame image according to the background image. The technical problem that most foreground detection methods cannot be realized due to that the calculating capability of the camera processor is limited can be solved. The foreground object with small size and weak contrast ratio can be completely detected at real time by adopting limited calculating capability by the foreground detection method and the foreground detection system.

Description

technical field [0001] The invention relates to the technical field of image and video processing of video surveillance. Specifically, it relates to a real-time foreground detection method and system adapted to small and weak targets. Background technique [0002] In the visual surveillance system, it is often necessary to detect, track, classify and analyze moving objects, and the accuracy of moving object detection directly affects the subsequent processing and operation. In order to adapt to complex and changeable scene changes, the most common method is to model the background, and then use the background model to detect the foreground target. The existing background modeling methods mainly include median method, mean method, kernel density estimation method, codebook model, mixed Gaussian model and so on. [0003] The Gaussian model is to use the Gaussian probability density function (normal distribution curve) to accurately quantify things, and decompose a thing into...

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
no application Login to view more
IPC IPC(8): G06T7/00
Inventor 陈建冲丁美玉晋兆龙陈卫东
Owner SUZHOU KEDA TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap