图像空洞填补算法及其FPGA硬件实现  

Holes filling algorithm and its hardware implementation on FPGA

作　　者：伍能举 徐栋 杨依忠[1] WU Nengju;XU Dong;YANG Yizhong(School of Electronic Science and Applied Physics,Hefei University of Technology,Hefei 230601,China)

机构地区：[1]合肥工业大学电子科学与应用物理学院,安徽合肥230601

出　　处：《合肥工业大学学报（自然科学版）》2021年第9期1216-1220,共5页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(61401137,61404043);安徽省科技重大专项资助项目(16030901007);安徽省科技攻关计划资助项目(1501021037)。

摘　　要：文章提出了一种有效的空洞填补算法,该算法首先通过深度梯度过渡减少前景与后景的边界空洞,然后根据空洞与周围像素远近关系设置不同填补权值并进行空洞填补;同时算法依据填补像素中不同的空洞数量给出不同填补方法。仿真结果表明,目标视图的峰值信噪比与均方误差分别达到38.64 dB和8.90,能够有效地消除目标视图中的空洞。运用Quartus Ⅱ软件,对该算法进行现场可编程门阵列(field programmable gate array,FPGA)硬件实现及测试;结果表明,该算法在硬件实现上也能够有效填补图像中的空洞。This paper presents an effective holes filling algorithm.A gradient transition depth map preprocessing method is adopted to reduce holes between foreground and background firstly.The weights of the filled pixels,which are used to holes filling,are set according to the distance between the hole pixels and their surrounding pixels.According to the number of holes in filled pixels,the algorithm uses different holes filling methods.Simulation results show that the peak signal-to-noise ratio(PSNR)and mean square error(MSE)of the target image are 38.64 dB and 8.90 respectively and holes in target image can be eliminated effectively.The algorithm is tested and verified by Quartus Ⅱ on field programmable gate array(FPGA)board.The experimental results show that the algorithm can fill image holes on hardware testing effectively.

关 键 词：空洞填补 深度梯度过渡 现场可编程门阵列(FPGA) 深度图像绘制 

分 类 号：TN409[电子电信—微电子学与固体电子学]