基于分水岭和强度分层的机器人视觉认知方法
Robot Visual Cognition Method Based on Watershed and Intensity Stratification
DOI: 10.12677/CSA.2019.98177, PDF,  被引量    国家自然科学基金支持
作者: 王天琪*, 贺乃宝, 高 倩, 赵俊杰：江苏理工学院，江苏 常州
关键词: 视觉伺服；分水岭；强度分层；图像矩；Visual Servo； Watershed； Strength Stratification； Image Moment
摘要: 应用图像矩进行工件识别时，对于随机放置或重叠的工件，由于手眼相机拍摄的原始图像中不可避免地存在各类噪声和灰度值相同但不属于同一工件的情况，因而无法直接计算目标工件的图像矩。针对这类问题，本文提出了一种基于分水岭分割和强度分层算法的视觉认知方法。该方法首先通过手眼相机获取实时图像，由专用的分水岭算法分割工件图像；将分割后的图像转换为灰度图像；然后根据不同的灰度值，将分割出来的工件图像进行强度分层，得到不同强度的切片，即各个工件的二值图像。最后使用基于边界的方法计算其中一个工件的图像矩。实验结果表明，该方法能够满足特定的视觉伺服作业。
Abstract: When the image moment is applied for workpiece recognition, for the randomly placed or overlapping workpiece, because of the unavoidable existence of all kinds of noise and gray value in the original image taken by the hand-eye camera, but not the same workpiece, therefore, the image moments of the target workpiece cannot be calculated directly. In this paper, a visual cognitive method based on watershed segmentation and intensity delamination is proposed. Firstly, the real-time image is obtained by hand-eye camera, and the workpiece image is segmented by a special watershed algorithm; the segmented image is converted into gray image; then, according to different gray values, the segmented workpiece image is stratified with intensity. Different strength slices are obtained, that is, binary images of each workpiece. Finally, an example is given to calculate the image moments of one of the workpieces. Experimental results show that the proposed method can meet specific visual servo operations.
文章引用：王天琪, 贺乃宝, 高倩, 赵俊杰. 基于分水岭和强度分层的机器人视觉认知方法[J]. 计算机科学与应用, 2019, 9(8): 1576-1583. https://doi.org/10.12677/CSA.2019.98177

参考文献

[1] Hutchinson, S., Hager, G.D. and Corke, P.I. (1996) A Tutorial on Visual Servo Control. IEEE Transactions on Robotics and Automa-tion, 12, 651-670. [Google Scholar] [CrossRef] 
[2] Espiau, B., Chaumette, F. and Rives, P. (1992) A New Approach to Visual Servoing in Robotics. IEEE Transactions on Robotics and Automation, 8, 313-326. [Google Scholar] [CrossRef] 
[3] Tahri, O. and Chaumette, F. (2005) Point-Based and Region-Based Image Moments for Visual Servoing of Planar Objects. IEEE Transactions on Robotics, 21, 1116-1127. [Google Scholar] [CrossRef] 
[4] Lin, F., Dong, X.X., Chen, B.M., Lum, K.Y. and Lee, T.H. (2012) A Robust Real-Time Embedded Vision System on an Unmanned Rotorcraft for Ground Target Following. IEEE Transactions on Industrial Electronics, 59, 1038-1049. [Google Scholar] [CrossRef] 
[5] Tahri, O., Araujo, H., Chaumette, F. and Mezouar, Y. (2013) Robust Im-age-Based Visual Servoing Using Invariant Visual Information. Robotics and Autonomous Systems, 61, 1588-1600. [Google Scholar] [CrossRef] 
[6] 叶国强, 等. 结合学习特征的图像矩视觉伺服方法[J]. 华南理工大学学报, 2017, 2(45): 99-106.
[7] 毛优新, 等. 基于图像矩与神经网络的机器人四自由度视觉伺服[J]. 控制理论与应用, 2009, 26(10): 1162-1166.
[8] 赵栋杰, 张宾, 王学雷, 郭洪红, 张松兵. 基于图像矩的室内喷雾机器人自动对靶研[J]. 农业机械学报, 2016, 47(12): 22-29.
[9] Mayer, S.B. (1990) Morphology Segmentation. Journal of Visual Communication and Image Representation, 1, 21-46. [Google Scholar] [CrossRef] 
[10] 高丽, 杨树元, 夏杰, 等. 基于标记的Watershed图像分割新算法[J]. 电子学报, 2006 , 34(11): 2018-2023.
[11] Canny, J. (1986) A Computational Approach to Edge Detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-8, 679-698. [Google Scholar] [CrossRef] 
[12] Soille, P. (1999) Morphological Image Analysis Principles and Applications. Springer Verlag, Berlin, 123-140. [Google Scholar] [CrossRef] 
[13] Soille, P. (2000) Morphological Image Analysis Applied to Crop Field Mapping. Image and Vision Computing, 18, 1025-1032. [Google Scholar] [CrossRef] 
[14] 方红萍, 方康玲, 刘新海. 自适应H-minima的改进分水岭堆叠细胞分割方法[J]. 计算机应用研究, 2016, 33(5): 1587-1590.
[15] Papakostas, G.A., Kara-kasis, E.G. and Koulouriotis, D.E. (2008) Efficient and Accurate Computation of Geometric Moments on Gray-Scale Image. Pattern Recognition, 41, 1895-1904. [Google Scholar] [CrossRef] 
[16] Philips, W.A. (1993) New Fast Algorithm for Moment Compution. Pattern Recognition, 26, 1619-1621. [Google Scholar] [CrossRef]