A fast detection method of image region of interest based on EEG and eye-mover

Abstract

The invention discloses a fast detection method of an image region of interest based on EEG and an eye mover, belonging to the technical field of cognitive neuroscience and image processing, which utilizes the EEG and the eye mover to realize the fast detection of an image region of interest which may contain a target object. The method utilizes the mixed mode of EEG and eye movement meter, takeseye movement signal as switch signal, makes full use of the information characteristics of EEG and eye movement meter, and combines with image segmentation technology to quickly and efficiently detectthe region of interest where the object is located.

Description

technical field [0001] The invention discloses a method for quickly detecting a region of interest in an image, in particular to a method for rapidly detecting a region of interest in which a target object is located by comprehensively detecting an eye movement signal and an electroencephalogram signal, and belongs to a technology intersecting cognitive neuroscience and image processing field. Background technique [0002] Realizing the rapid detection of the region of interest containing the target in the image has important application value in the fields of national defense, security, exploration and so on. Most of the existing target detection methods only rely on image recognition methods, which need to process the entire image, resulting in an increase in the amount of calculation and a longer processing time. However, for some images containing noise or low resolution, it cannot be detected by image processing alone or may easily cause false detection. [0003] When...

AI Technical Summary

Problems solved by technology

At present, the scheme of using EEG signals combined with eye movement signals to obtain the target area is to collect eye movement data and EEG signals when the user watches the screen in real time, and jointly analyze the eye movement trajectory and EEG signals to reveal the user's perception of the image. Which areas in the image are of interest, so as to determine the position of the target object in the image, this scheme has the following defects: (1) The eye tracker always outputs the ROI, so it lacks sufficient shielding effect for the special case where no ROI exists, reducing the (2) Use the ratio of the power spectrum estimates of the three frequency bands of the EEG to represent the degree of interest, and cluster the eye track point data corresponding to the time when the user is more interested Analysis to determine the image target area, the calculation of detecting the area of ​​interest is complex, which increases the calculation amount of the human-machine hybrid detection method and reduces the detection efficiency

Images