Moving Object Detection in High Dynamic Scene for Visual Prosthesis

Visual prosthesis that applies electrical stimulation to restore functional vision has a promising prospect. However, visual percepts generated by current visual prosthesis are low resolution with unruly color and restricted grayscale. This severely restricts the ability of prosthetic implant to complete visual tasks in daily scenes. At present, some studies use existing image processing techniques to improve the perceptions of objects in prosthetic vision. However, most of these studies detect the static objects and optimize the visual percepts in general dynamic scenes. This greatly limits the application of visual prosthesis in high dynamic scenes. In this study, a novel moving object detection model is proposed to automatically extract the moving objects in high dynamic scene. In this model, the optical flow, color and luminance are applied to form different dimensional feature space, and construct the moving object map according to the visual saliency in different feature channel. The proposed method can uniformly highlight the moving object and keep good boundaries in high dynamic scene. The simulation results indicate that the proposed moving object model can effectively detect the moving object.