基于目标特征的植株深度图像修复
|
摘要
针对植株深度图像的像素错误和缺失、常见的滤波方法无法准确修复植株深度图像的问题,提出一种基于目标特征的植株深度图像修复方法。首先基于颜色和空间信息的图像分割算法对植株彩色图像进行目标分割,再检索每个目标的外轮廓,并对外轮廓进行多边形拟合;其次,基于目标区域搜索深度图像中具有正确深度值的像素作为目标区域采样点,并对叶片区域的图像进行归一化;最后,利用空间拟合法计算各目标区域的方程,修复区域内小面积错误和缺失的深度值,同时采用支持向量机和空间变换运算对大面积错误和缺失深度值的叶片区域进行修复。实验结果表明,该方法能够准确地修复植株深度图像中错误、缺失的深度数据,且能够有效地保护目标区域的边缘信息。
Considering the depth images of plants provided by depth camera are incomplete, and common filtering methods cannot inpaint the plant depth images accurately, we propose a plant depth images inpainting method which is based on target features. Firstly, targets of plant color images are segmented by using a color image segmentation algorithm based on color and spatial information,then the outer contour of each target is retrieved, and polygon for each outer contour is fitted.Secondly, the pixels with correct depth value in the depth images are searched to act as sampling points, and meanwhile the leaf maps are normalized. Finally, using spatial fitting method to calculate every target area's equation to correct the small area's depth pixels which need to be corrected. In the meantime, support vector machine and spatial transformation are used to get the accurate large area's depth pixels which need to be corrected. The experiments show that the proposed method achieves better performance for plant depth image inpainting, and protects targets' edge information.
Considering the depth images of plants provided by depth camera are incomplete, and common filtering methods cannot inpaint the plant depth images accurately, we propose a plant depth images inpainting method which is based on target features. Firstly, targets of plant color images are segmented by using a color image segmentation algorithm based on color and spatial information,then the outer contour of each target is retrieved, and polygon for each outer contour is fitted.Secondly, the pixels with correct depth value in the depth images are searched to act as sampling points, and meanwhile the leaf maps are normalized. Finally, using spatial fitting method to calculate every target area's equation to correct the small area's depth pixels which need to be corrected. In the meantime, support vector machine and spatial transformation are used to get the accurate large area's depth pixels which need to be corrected. The experiments show that the proposed method achieves better performance for plant depth image inpainting, and protects targets' edge information.
引文
[1]CAMPLANI M,SALGADO L.Adaptive spatiotemporal filter for low-cost camera depth maps[C]//2012 IEEEInternational Conference on Emerging Signal Processing Applications.New York:IEEE Press,2012:33-36.
[2]CAMPLANI M,SALGADO L.Efficient spatio-temporal hole filling strategy for kinect depth maps[EB/OL].[2018-10-11].https://www.spiedigitallibrary.org/conference-proceedings-of-spie/8290/1/Efficient-spatio-temporal-hole-filling-strategy-forKinect-depth-maps/10.1117/12.911909.full?SSO=1.
[3]MATYUNIN S,VATOLIN D,BERDNIKOV Y,et al.Temporal filtering for depth maps generated by Kinect depth camera[EB/OL].[2018-09-21].https://ieeexplore.ieee.org/document/5877202.
[4]LIU W,XUE H Y,GU Y,et al.Shape preserving RGB-D depth map restoration[C]//International Conference on Neural Information Processing 2014.Berlin:Springer,2014:150-158.
[5]CHEN C Y,CAI J F,ZHENG J M,et al.Kinect depth recovery using a color-guided,region-adaptive,and depth-selective framework[J].ACM Transactions on Intelligent Systems and Technology,2015,6(2):1-19.
[6]WASZA J,BAUER S,HORNEGGER J.Real-time preprocessing for dense 3-D range imaging on the GPU:Defect interpolation,bilateral temporal averaging and guided filtering[C]//2011 IEEE International Conference on Computer Vision Workshops.New York:IEEE Press,2011:1221-1227.
[7]徐欢,郑江滨.Kinect深度图像修复算法[J].微电子学与计算机,2013,30(11):98-103.
[8]吴倩,史晋芳,王德娇,等.结合彩色图像局部分割的Kinect深度图修复算法[J].计算机应用研究,2017,34(12):3852-3854,3884.
[9]LI H,LUO L J,VLASIC D,et al.Temporally coherent completion of dynamic shapes[J].ACM Transactions on Graphics,2012,31(1):1-11.
[10]SUNG M,KIM V G,ANGST R,et al.Data-driven structural priors for shape completion[J].ACMTransactions on Graphics,2015,34(6):1-11.
[11]沈跃,徐慧,刘慧,等.基于K-means和近邻回归算法的Kinect植株深度图像修复[J].农业工程学报,2016,32(19):188-194.
[12]高雅平,陈一民,邹一波,等.优化改进的深度图像空洞填补算法[J].计算机工程与设计,2016,37(11):3040-3044.
[13]叶齐祥,高文,王伟强,等.一种融合颜色和空间信息的彩色图像分割算法[J].软件学报,2004,15(4):522-530.
[14]SUZUKI S,BE K.Topological structural analysis of digitized binary images by border following[J].Computer Vision,Graphics,and Image Processing,1985,30(1):32-46.
[15]曾清红,卢德唐.基于移动最小二乘法的曲线曲面拟合[J].工程图学学报,2004,25(1):84-89.
[16]李知菲,陈源.基于联合双边滤波器的Kinect深度图像滤波算法[J].计算机应用,2014,34(8):2231-2234,2242.
[17]孟恬,全红艳.基于快速行进算法的深度图像增强算法[J].计算机应用与软件,2017(8):232-237.
[2]CAMPLANI M,SALGADO L.Efficient spatio-temporal hole filling strategy for kinect depth maps[EB/OL].[2018-10-11].https://www.spiedigitallibrary.org/conference-proceedings-of-spie/8290/1/Efficient-spatio-temporal-hole-filling-strategy-forKinect-depth-maps/10.1117/12.911909.full?SSO=1.
[3]MATYUNIN S,VATOLIN D,BERDNIKOV Y,et al.Temporal filtering for depth maps generated by Kinect depth camera[EB/OL].[2018-09-21].https://ieeexplore.ieee.org/document/5877202.
[4]LIU W,XUE H Y,GU Y,et al.Shape preserving RGB-D depth map restoration[C]//International Conference on Neural Information Processing 2014.Berlin:Springer,2014:150-158.
[5]CHEN C Y,CAI J F,ZHENG J M,et al.Kinect depth recovery using a color-guided,region-adaptive,and depth-selective framework[J].ACM Transactions on Intelligent Systems and Technology,2015,6(2):1-19.
[6]WASZA J,BAUER S,HORNEGGER J.Real-time preprocessing for dense 3-D range imaging on the GPU:Defect interpolation,bilateral temporal averaging and guided filtering[C]//2011 IEEE International Conference on Computer Vision Workshops.New York:IEEE Press,2011:1221-1227.
[7]徐欢,郑江滨.Kinect深度图像修复算法[J].微电子学与计算机,2013,30(11):98-103.
[8]吴倩,史晋芳,王德娇,等.结合彩色图像局部分割的Kinect深度图修复算法[J].计算机应用研究,2017,34(12):3852-3854,3884.
[9]LI H,LUO L J,VLASIC D,et al.Temporally coherent completion of dynamic shapes[J].ACM Transactions on Graphics,2012,31(1):1-11.
[10]SUNG M,KIM V G,ANGST R,et al.Data-driven structural priors for shape completion[J].ACMTransactions on Graphics,2015,34(6):1-11.
[11]沈跃,徐慧,刘慧,等.基于K-means和近邻回归算法的Kinect植株深度图像修复[J].农业工程学报,2016,32(19):188-194.
[12]高雅平,陈一民,邹一波,等.优化改进的深度图像空洞填补算法[J].计算机工程与设计,2016,37(11):3040-3044.
[13]叶齐祥,高文,王伟强,等.一种融合颜色和空间信息的彩色图像分割算法[J].软件学报,2004,15(4):522-530.
[14]SUZUKI S,BE K.Topological structural analysis of digitized binary images by border following[J].Computer Vision,Graphics,and Image Processing,1985,30(1):32-46.
[15]曾清红,卢德唐.基于移动最小二乘法的曲线曲面拟合[J].工程图学学报,2004,25(1):84-89.
[16]李知菲,陈源.基于联合双边滤波器的Kinect深度图像滤波算法[J].计算机应用,2014,34(8):2231-2234,2242.
[17]孟恬,全红艳.基于快速行进算法的深度图像增强算法[J].计算机应用与软件,2017(8):232-237.