Unlock instant, AI-driven research and patent intelligence for your innovation.

Frequency-time saliency combination-based moving target detection method

A moving target and detection method technology, applied in the field of pattern recognition and computer vision, can solve the problems of light mutation, ghosting, and large amount of calculation, and achieve the effects of suppressing ghosting problems, ensuring integrity, and high detection accuracy

Inactive Publication Date: 2017-09-01
HARBIN ENG UNIV
View PDF4 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the existing moving target detection method based on the mixed Gaussian model still has the ghost phenomenon caused by the sudden change of light, and the problem of large amount of calculation in a complex environment cannot be solved well.

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
  • Frequency-time saliency combination-based moving target detection method
  • Frequency-time saliency combination-based moving target detection method
  • Frequency-time saliency combination-based moving target detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0044] The present invention is described in more detail below in conjunction with accompanying drawing:

[0045] The present invention proposes a moving target detection method based on frequency-time saliency combination, and its flow is as follows figure 1 Specifically, the following steps are included:

[0046] Step 1: Obtain a video sequence containing multiple frames of images, and extract the first frame of images;

[0047] Step 2: Gaussian smoothing is performed on the image, and then the Gaussian smoothed image is converted to the Lab color space to obtain an image in the Lab color space;

[0048] The Gaussian smoothing process includes first averaging the pixel gray levels in the image neighborhood, and then performing weighted processing on each position pixel;

[0049] Step 3: Process the image of the Lab color space through the frequency tuning algorithm, and extract the salient moving object information; if the salient moving object information is extracted, pe...

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 present invention provides a frequency-time saliency combination-based moving target detection method. According to the method, an image is preprocessed; salient moving target information in the image is detected through a frequency tuning algorithm; a dynamic saliency graph is calculated; and the dynamic saliency graph is further detected through a Gaussian mixture model algorithm, so that a final detection result is obtained. The method of the invention integrates the frequency tuning method and the Gaussian mixture model method, and therefore, the calculation amount of the method is small, and the method can inhibit the ghosting of the moving object in the detection image.

Description

technical field [0001] The invention relates to the fields of pattern recognition and computer vision, in particular to a moving target detection method based on frequency-time saliency combination. Background technique [0002] As the main research content of video surveillance, moving target detection in video sequence images is a research hotspot in the field of computer vision, and it has an important role in promoting other research fields of computer vision. Moving object detection is the basis for realizing object tracking, intelligent monitoring, behavior analysis and other tasks. Therefore, detecting moving objects in video images has important research significance and application prospects. [0003] Moving object detection refers to distinguishing the moving object of interest from the background or other uninteresting moving objects in the video image sequence. Traditional moving target detection methods mainly include: optical flow method, frame difference met...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06T7/246G06T7/254
CPCG06T7/246G06T7/254G06T2207/30232G06V20/54G06V2201/08
Inventor 王辉高菁于立君董泽全魏智红张雪胡羽坤丁莹王正安
Owner HARBIN ENG UNIV