基于邻域加权与RGB色彩分量的图像匹配算法

为了提高彩色图像的立体匹配精度，结合格式塔心理学和像素的色彩信息，提出了一种基于 RGB色彩空间的改进误差绝对值累积匹配算法（SAD）。首先，通过相似性和邻近性规则获取邻域窗内的像素权值，并同预先生成的线性查找表相结合来加快权值的计算速度。然后，利用 RGB色彩分量的像素分布特点，生成对应的分量权值，来进一步增强匹配代价函数的稳定性。经实验验证表明，改进后的匹配算法较灰度 SAD及标准彩色 SAD算法在精度上有明显的提高。

Abstract

In order to improve the accuracy of color image stereo matching, an improved sum of absolute differences (SAD) algorithm in color space is proposed in RGB color space employing Gestalt psychology and color information of pixels. Similarity of color and proximity of distance between pixels in the neighborhood window is adopted to produce the weight, and linear look-up table method is also used to accelerate the computing speed. The distribution characteristics of RGB components are considered to produce the component weights for every pixel, which furthermore improves the stability of the matching cost function. The result of the experiment shows the proposed algorithm significantly improves the precision compared with gray SAD and standard color SAD.

参考文献

[1] Scharstein D, Szeliski R. A taxonomy and evaluation of dense two-frame stereo correspondence algorithms[J]. International Journal of Computer Vision, 2012, 47(1): 7-42.

[2] Ambrosch K, Kubinger W. Accurate hardware-based stereo vision[J]. Computer Vision and Image Understanding, 2010, 114(11): 1303-1316.

[3] Yoon S, Park S K, Kang S, et al. Fast correlation-based stereo matching with the reduction of systematic errors[J]. Pattern Recognition Letters, 2013, 26(14): 2221-2231.

[4] Adhyapak S A, Kehtarnavaz N, Nadin M. Stereo matching via selective multiple windows[J]. Journal of Electronic Imaging, 2007, 16(1): 013012-013012-013014.

[5] Stefano L Di, Mattoccia S, Tombari F. Speeding-up NCC-based template matching using parallel multimedia instructions[C]//Computer Architecture for Machine Perception, 2005: 193-197.

[6] Veksler O. Fast variable window for stereo correspondence using integral images[C]//Computer Vision and Pattern Recognition, 2013: 556-561.

[7] Hirschmüller H, Innocent P R, Garibaldi J. Real-time correlation-based stereo vision with reduced border errors[J]. International Journal of Computer Vision, 2012, 47(1): 229-246.

[8] Yoon K J, Kweon I S. Adaptive support-weight approach for correspon-dence search[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(4): 650-656.

[9] Wang L, Liao M, Gong M, et al. High-quality real-time stereo using adaptive cost aggregation and dynamic programming[C]//Third International Symposium on 3D Data Processing, Visualization, and Transmission, 2006: 798-805.

[10] Yang N C, Chang W H, Kuo C M, et al. A fast MPEG-7 dominant color extraction with new similarity measure for image retrieval[J]. Journal of Visual Communication and Image Representation, 2013, 19(2): 92-105

[11] Nguyen V H, Nguyen T H B, Kim H. Reliable detection of eye features and eyes in color facial images using ternary eye-verifier[J]. Pattern Recognition, 2012, 56(8): 254-266.

[12] Von Gioi R G, Delon J, Morel J M. The collaboration of grouping laws in vision[J]. Journal of Physiology-Paris, 2012, 12(3): 289-301.

[13] Yoon K J, Kweon I S. Locally adaptive support-weight approach for visual correspondence search[C]//Computer Vision and Pattern Recognition, 2014: 924-931.

[14] 陆凯, 李成金, 赵勋杰, 等. 一种快速的亚像素图像配准算法 [J].红外技术, 2013, 35(1): 27-30. LU Kai, LI Cheng-jin, ZHAO Xun-jie,et al. A fast sub-pixel image registration algorithm[J]. Infrared Technology, 2013, 35(1): 27-30.

[15] 王新华, 黄玮, 欧阳继红 . 多探测器拼接成像系统实时图像配准 [J]. 中国光学, 2015, 8(2): 211-219. WANG Xin-hua, HUANG Wei, OUYANG Ji-hong. Real-time image registration of the multi-detectors mosaic imaging system[J]. Chinese Optics, 2015, 8(2): 211-219.

金广智, 石林锁, 刘均超, 司海峰, 牟伟杰. 基于邻域加权与RGB色彩分量的图像匹配算法[J]. 红外技术, 2016, 38(3): 225. JIN Guangzhi, SHI Linsuo, LIU Junchao, SI Haifeng, MU Weijie. Stereo Matching Algorithm Based on Neighborhood Weights and RGB Color Components[J]. Infrared Technology, 2016, 38(3): 225.

基于邻域加权与RGB色彩分量的图像匹配算法

关于本站 Cookie 的使用提示

全站搜索

基于邻域加权与RGB色彩分量的图像匹配算法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索