基于联合双边滤波器的Kinect深度图像滤波算法

doi:10.11772/j.issn.1001-9081.2014.08.2231

计算机应用 ›› 2014, Vol. 34 ›› Issue (8): 2231-2234.DOI: 10.11772/j.issn.1001-9081.2014.08.2231

• 第五届中国数据挖掘会议(CCDM 2014)论文 • 上一篇下一篇

基于联合双边滤波器的Kinect深度图像滤波算法

李知菲,陈源

浙江师范大学数理与信息工程学院，浙江金华321004

收稿日期:2014-04-29 修回日期:2014-05-08 出版日期:2014-08-01 发布日期:2014-08-10
通讯作者: 李知菲
作者简介:李知菲(1978-)，男，黑龙江哈尔滨人，讲师，硕士研究生，主要研究方向：图像处理、模式识别、虚拟现实；陈源(1989-)，女，浙江杭州人，硕士研究生，主要研究方向：图像处理、模式识别。
基金资助:
浙江省教育厅科研项目;浙江省科技计划一般科研项目

Kinect depth image filtering algorithm based on joint bilateral filter

LI Zhifei,CHEN Yuan

College of Mathematics, Physics and Information Engineering, Zhejiang Normal University, Jinhua Zhejiang 321004, China

Received:2014-04-29 Revised:2014-05-08 Online:2014-08-01 Published:2014-08-10
Contact: LI Zhifei

摘要/Abstract

摘要：

针对Kinect镜头采集的深度图像一般有噪声和黑洞现象，直接应用于人体动作跟踪和识别等系统中效果差的问题，提出一种基于联合双边滤波器的深度图像滤波算法。算法利用联合双边滤波原理，将Kinect镜头同一时刻采集的深度图像和彩色图像作为输入，首先，用高斯核函数计算出深度图像的空间距离权值和RGB彩色图像的灰度权值；然后，将这两个权值相乘得到联合滤波权值，并利用快速高斯变换替换高斯核函数，设计出联合双边滤波器；最后，用此滤波器的滤波结果与噪声图像进行卷积运算实现Kinect深度图像滤波。实验结果表明，所提算法应用在人体动作识别和跟踪系统后，可显著提高在背景复杂场景中的抗噪能力，识别正确率提高17.3%，同时所提算法的平均耗时为371ms，远低于同类算法。所提算法保持了联合双边滤波平滑保边的优点，由于引入彩色图像作为引导图像，去噪的同时也能对黑洞进行修补，因此该算法在Kinect深度图像上的去噪和修复效果优于经典的双边滤波算法和联合双边滤波算法，且实时性强。

Abstract:

Usually the depth image obtained by Kinect camera contains noise and black holes, so the effect is poor if it is directly applied into human motion tracking and recognition system. To solve this problem, an efficient depth image filtering algorithm based on joint bilateral filter was proposed. The principle of joint bilateral filtering was used in the proposed algorithm, and the depth and color images were captured by Kinect camera at the same time as the input. Spatial distance weight value of depth image and grayscale weight value of RGB color image were computed by Gaussian kernel function. Then these two weight values were multiplied to get the weight value of joint bilateral filter. A joint bilateral filter was designed by replacing the Gaussian kernel function with fast Gaussian transform. Finally, this filtered result was convolved with the noisy image to filter the Kinect depth image. The experimental results show that the proposed algorithm can significantly improve the robustness to noise in the human motion tracking and identification system and increase the recognition rate by 17.3%. The average running time of the proposed algorithm is 371ms, and is much lower than similar other algorithms. The proposed algorithm keeps the advantages of joint bilateral filter. Since the color image is introduced into the algorithm, the proposed algorithm can well repair the black holes while reducing the noise. The proposed algorithm is better than traditional bilateral filter and joint bilateral filter in denoising and repairing holes for the Kinect depth image, and it has higher real-time performance.

中图分类号:

TP391.413

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.

LI Zhifei CHEN Yuan. Kinect depth image filtering algorithm based on joint bilateral filter[J]. Journal of Computer Applications, 2014, 34(8): 2231-2234.

参考文献

［1］SHEN X, ZHANG H, GAO Z, et al. A robust behavior recognition algorithm based on sole depth information ［J］. Journal of Optoelectronics · Laser, 2013, 24(8): 1613-1618.(申晓霞,张桦,高赞,等.一种鲁棒的基于深度数据的行为识别算法［J］.光电子·激光,2013,24(8):1613-1618.)
［2］ZHENG L, LUO J, XU K. Implementation on mocap system based on Kinect ［J］. Journal of Jilin University: Engineering and Technology Edition, 2013, 43(S1): 249-255.(郑立国,罗江林,许舸.基于Kinect的动作捕捉系统的实现［J］.吉林大学学报:工学版,2013,43(增刊1):249-255.)
［3］YANG X, ZHANG Q, ZHOU Y. Human motion tracing algorithm based on Kinect depth information ［J］. Journal of Beijing Information Science and Technology University, 2013, 28(1): 33-37.(杨晓敏,张奇志,周亚丽.基于Kinect深度信息的人体运动跟踪算法［J］.北京信息科技大学学报:自然科学版,2013,28(1):33-37.)
［4］WANG K, XIONG X, REN Z. Highly efficient mean filtering algorithm ［J］. Application Research of Computers, 2010, 27(2): 434-438.(王科俊,熊新炎,任桢.高效均值滤波算法［J］.计算机应用研究,2010,27(2):434-438.)
［5］LEI C, LIU J, ZHANG M. New median filter algorithm based on adaption ［J］.Computer Engineering and Applications, 2008, 44(2): 60-62.(雷超阳,刘军华,张敏.一种基于自适应的新型中值滤波算法［J］.计算机工程与应用,2008,44(2):60-62.)
［6］DARIBO I, TILLIER C, PESQUET-POPESCU B. Distance dependent depth filtering in 3D warping for 3DTV ［C］// MMSP 2007: Proceedings of IEEE 9th Workshop on Multimedia Signal Processing. Piscataway: IEEE Press, 2007: 312-315.
［7］EDELER T, OHLIGER K, HUSSMANN S, et al. Time-of-flight depth image denoising using prior noise information ［C］// ICSP2010: Proceedings of IEEE 10th International Conference on Signal Processing. Piscataway: IEEE Press, 2010: 119-122.
［8］NEWCOMBE R, HILLIGES O, KIM D, et al. Kinect fusion: real-time dense surface mapping and tracking ［C］// Proceedings of the 2011 10th IEEE International Symposium on Mixed and Augmented Reality. Piscataway: IEEE Press, 2011: 127-136.
［9］IZADI S, KIM D, HILLIGES O, et al. Kinect fusion: real-time 3D reconstruction and interaction using a moving depth camera ［C］// Proceedings of the 24th Annual ACM Symposium on User Interface Software and Technology. New York: ACM, 2011: 559-568.
［10］TOMASI C, MANDUCHI R. Bilateral filtering for gray and color images ［C］// Proceedings of the Sixth International Conference on Computer Vision. Washington, DC: IEEE Computer Society, 1998: 839-846.
［11］ZHANG Z, WANG W. A modified bilateral filtering algorithm ［J］. Journal of Image and Graphics, 2009, 14(3): 443-447.(张志强,王万玉.一种改进的双边滤波算法［J］.中国图象图形学报,2009,14(3):443-447.)
［12］PETSCHNINGG G,SZELISKI R, AGRAWALA A, et al. Digital photography with flash and no-flash image pairs ［J］. ACM Transactions on Graphics, 2004, 23(3): 664-672.
［13］WANG X, JIN X, XU X, et al. Joint bilateral image interpolation ［J］. Journal of Image and Graphics, 2011, 16(12): 2117-2123.(王孝通,金鑫,徐晓刚,等.联合双边图像插值［J］.中国图象图形学报,2011,16(12):2117-2123.)
［14］YOSHIZAWA S, BELYAEV A, YOKOTA H. Fast Gauss bilateral filtering ［J］. Computer Graphics Forum, 2010, 29(1): 60-74.

[1]	温静, 宋建伟. 基于多级全局信息传递模型的视觉显著性检测[J]. 计算机应用, 2021, 41(1): 208-214.
[2]	许长青, 陈振杰, 侯仁福. 融合不精准先验知识的Landsat 8 OLI影像深度学习分类方法[J]. 计算机应用, 2020, 40(12): 3550-3557.
[3]	陈利霞, 班颖, 王学文. 基于张量核范数与3D全变分的背景减除[J]. 计算机应用, 2020, 40(9): 2737-2742.
[4]	邓茜文, 冯子亮, 邱晨鹏. 基于近红外与可见光双目视觉的活体人脸检测方法[J]. 计算机应用, 2020, 40(7): 2096-2103.
[5]	王肖, 魏嘉旺, 袁玉波. 基于特征部位圆形域的人脸图像修复方法[J]. 计算机应用, 2020, 40(3): 847-853.
[6]	刘颖, 王凤伟, 刘卫华, 艾达, 李芸, 杨凡超. 基于亮度分区模糊融合的高动态范围成像算法[J]. 计算机应用, 2020, 40(1): 233-238.
[7]	王文杰, 乔志伟, 牛蕾, 席雅睿. 自适应步长非局部全变分约束迭代图像重建算法[J]. 计算机应用, 2020, 40(1): 245-251.
[8]	严经纬, 李强, 王春茂, 谢迪, 王保青, 戴骏. 面部运动单元检测研究综述[J]. 计算机应用, 2020, 40(1): 8-15.
[9]	王书朋, 赵瑶. 基于自适应分割的多曝光图像融合算法[J]. 计算机应用, 2020, 40(1): 252-257.
[10]	马伟苹, 李文新, 孙晋川, 曹鹏霞. 基于粗精立体匹配的双目视觉目标定位方法[J]. 计算机应用, 2020, 40(1): 227-232.
[11]	王鑫, 张鑫, 宁晨. 基于多特征降维和迁移学习的红外人体目标识别方法[J]. 计算机应用, 2019, 39(12): 3490-3495.
[12]	曹昀炀, 王涛. 耦合先验拉普拉斯坐标的半监督图像分割算法[J]. 计算机应用, 2019, 39(9): 2695-2700.
[13]	王克强, 张雨帅, 王保群. 基于Retinex理论的多曝光图像融合算法[J]. 计算机应用, 2019, 39(7): 2087-2092.
[14]	李文俊, 陈斌, 李建明, 钱基德. 基于深度神经网络的表面划痕识别方法[J]. 计算机应用, 2019, 39(7): 2103-2108.
[15]	朱锴, 付忠良, 陈晓清. 基于卷积神经网络的超声图像左心室分割方法[J]. 计算机应用, 2019, 39(7): 2121-2124.

基于联合双边滤波器的Kinect深度图像滤波算法

Kinect depth image filtering algorithm based on joint bilateral filter

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics