Video saliency detection method based on Bayesian fusion

Abstract

The invention discloses a video saliency detection method based on Bayesian fusion and mainly solves the problem that an existing video saliency detection method cannot detect small targets. The implementation scheme is that 1) extracting a video sequence static boundary probability saliency map, a color mean value saliency map and a color contrast saliency map, and carrying out weight fusion to generate a static saliency map; 2) extracting a video sequence dynamic boundary probability saliency map, a PCA prior saliency map and a background prior saliency map, and carrying out weight fusion togenerate a dynamic saliency map; and 3) carrying out fusion on the static saliency map and the dynamic saliency map through a Bayesian model to obtain a video sequence saliency map. Compared with a conventional video saliency algorithm, the method enhances characteristic space and time representation ability, reduces influence of complex backgrounds on the detection effect, can detect the small targets in a video effectively, and can be used for early-stage pretreatment of video target tracking and video segmentation.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to a video saliency detection method, which can be used for target tracking, object recognition and video segmentation. Background technique [0002] When the computer is dealing with complex scene problems, the complexity of the background makes some existing methods unable to deal with the scene better. Studies have found that the human visual system can easily understand various complex scenes, so the working principle of the human system can be used for reference when dealing with problems related to complex scenes. Scholars have conducted in-depth research and reasoning on the mechanism of human visual attention selection and obtained the theory of visual saliency. This theory believes that the human visual system only processes some parts of the image in detail, while almost turning a blind eye to the rest of the image. Based on this theory, relevant scholars i...

AI Technical Summary

Problems solved by technology

[0009] The above-mentioned video salient region extraction methods all add motion features to the salient model in the image domain, their time domain structure is too simple, and the relationship between the sequences is weak. Although the video salient region can be extracted, the accuracy and efficiency are relatively low. Low

Images