Image area clustering method, image area clustering device, outline searching method and outline searching device

Abstract

The invention discloses an image area clustering method, an image area clustering device, an outline searching method and an outline searching device. After a difference value point image is obtained by comparing the current input image with a background image, stacking predetermined grid models on the difference value point image; calculating the energy characteristics of each edge of grids in the grid models which are stacked on the difference value point image; then, determining the edges with the larger energy characteristics as candidate edges; remaining the partial candidate edges whichcan form the closed outline; deleting the rest candidate edges; obtaining the outline of a specific area formed by pixel points with larger background difference in the current input image, correspondingly, obtaining the outline of the specific area; and clustering the pixel points of the specific area at the same time; and moreover, because the clustering mode is realized by the closed outline, the dispersed noise points can not be detected as points in the specific area by mistake.

Description

technical field [0001] The present invention relates to image contour detection technology, in particular to a clustering method and device for image regions, and a contour search method and device. Background technique [0002] In the existing video surveillance device, the video in the monitoring scene is usually captured by a still camera, and then the moving object detection and tracking are performed on the continuous multi-frame images of the video, so as to distinguish the moving objects in the continuous multi-frame images from the background image. for analysis. [0003] In the detection and tracking of moving objects, foreground detection is the first step, and its accuracy directly affects the performance of the entire device. The processing process of the existing foreground detection is as follows: first compare the current input image with the background image, and obtain the value difference between each pixel in the current input image and the corresponding ...

AI Technical Summary

Problems solved by technology

[0005] However, although the above clustering process can obtain each foreground area in the current input image, it cannot obtain the contour features of the specific area represented by each foreground area.

Moreover, the above-mentioned clustering process is also easy to determine discrete noise points as foreground points, so that the noise area is easily detected as the foreground area by mistake, and it is easy to determine the foreground points with a similar color to the background as the background, thereby making the foreground detection Not very accurate

Images