Authors: Shufan Yang KongFatt WongLin James Andrew Terrence Mak T Martin McGinnity
Publish Date: 2017/01/20
Volume: 30, Issue: 9, Pages: 2697-2708
Abstract
Using programmable systemonchip to implement computer vision functions poses many challenges due to highly constrained resources in cost size and power consumption In this work we propose a new neuroinspired image processing model and implemented it on a systemonchip Xilinx Z702c board With the attractor neural network model to store the object’s contour information we eliminate the computationally expensive steps in the curve evolution reinitialisation at every new iteration or frame Our experimental results demonstrate that this integrated approach achieves accurate and robust object tracking when they are partially or completely occluded in the scenes Importantly the system is able to process 640 by 480 videos in realtime stream with 30 frames per second using only one lowpower Xilinx Zynq7000 systemonchip platform This proofofconcept work has demonstrated the advantage of incorporating neuroinspired features in solving image processing problems during occlusion
Keywords: