A Method for Tree Image Segmentation Combined Adaptive Mean Shifting with Image Abstraction
2
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
Although huge progress has been made in current image segmentation work, there are still no efficientsegmentation strategies for tree image which is taken from natural environment and contains complexbackground. To improve those problems, we propose a method for tree image segmentation combining adaptivemean shifting with image abstraction. Our approach perform better than others because it focuses mainly onthe background of image and characteristics of the tree itself. First, we abstract the original tree image usingbilateral filtering and image pyramid from multiple perspectives, which can reduce the influence of thebackground and tree canopy gaps on clustering. Spatial location and gray scale features are obtained by stepdetection and the insertion rule method, respectively. Bandwidths calculated by spatial location and gray scalefeatures are then used to determine the size of the Gaussian kernel function and in the mean shift clustering. Furthermore, the flood fill method is employed to fill the results of clustering and highlight the region ofinterest. To prove the effectiveness of tree image abstractions on image clustering, we compared differentabstraction levels and achieved the optimal clustering results. For our algorithm, the average segmentationaccuracy (SA), over-segmentation rate (OR), and under-segmentation rate (UR) of the crown are 91.21%,3.54%, and 9.85%, respectively. The average values of the trunk are 92.78%, 8.16%, and 7.93%, respectively. Comparing the results of our method experimentally with other popular tree image segmentation methods, oursegmentation method get rid of human interaction and shows higher SA. Meanwhile, this work shows apromising application prospect on visual reconstruction and factors measurement of tree.
Abstract
Although huge progress has been made in current image segmentation work, there are still no efficientsegmentation strategies for tree image which is taken from natural environment and contains complexbackground. To improve those problems, we propose a method for tree image segmentation combining adaptivemean shifting with image abstraction. Our approach perform better than others because it focuses mainly onthe background of image and characteristics of the tree itself. First, we abstract the original tree image usingbilateral filtering and image pyramid from multiple perspectives, which can reduce the influence of thebackground and tree canopy gaps on clustering. Spatial location and gray scale features are obtained by stepdetection and the insertion rule method, respectively. Bandwidths calculated by spatial location and gray scalefeatures are then used to determine the size of the Gaussian kernel function and in the mean shift clustering. Furthermore, the flood fill method is employed to fill the results of clustering and highlight the region ofinterest. To prove the effectiveness of tree image abstractions on image clustering, we compared differentabstraction levels and achieved the optimal clustering results. For our algorithm, the average segmentationaccuracy (SA), over-segmentation rate (OR), and under-segmentation rate (UR) of the crown are 91.21%,3.54%, and 9.85%, respectively. The average values of the trunk are 92.78%, 8.16%, and 7.93%, respectively. Comparing the results of our method experimentally with other popular tree image segmentation methods, oursegmentation method get rid of human interaction and shows higher SA. Meanwhile, this work shows apromising application prospect on visual reconstruction and factors measurement of tree.
참고문헌 (36)
- Y. Boykov, "Interactive organ segmentation using graph cuts", Medical Image Computing and Computer-Assisted Intervention – MICCI 2000, pp.276~286 (2000), Springer
- Q. Zeng, "Algorithm based on marker-controlled watershed transform for overlapping plant fruit segmentation", Optical Engineering, 48(2), pp.027201~ (2009)
- H. T. Zhang, "Feature extraction for the stored-grain insect detection system based on image recognition technology", Transactions of the Chinese Society of Agricultural Engineering, 25(2), pp.126~130 (2009)
- C. S. Sharp, "A vision system for landing an unmanned aerial vehicle", Proceedings of IEEE International Conference on Robotics and Automation, pp.1720~1727 (2001)
- D. K Isenor, "Fingerprint identification using graph matching", Pattern Recognition, 19(2), pp.113~122 (1986)
- B. Wang, "New result on maximum entropy threshold image segmentation based on P system", Optik, 163, pp.81~85 (2018)
- N. Otsu, "A threshold selection method from gray-level histograms", IEEE Transactions on Systems, Man, and Cybernetics, 9(1), pp.62~66 (1979)
- Y. Xu, "Segmentation of building roofs from airborne LiDAR point clouds using robust voxel-based region growing", Remote Sensing Letters, 8(11), pp.1062~1071 (2017)
- D. Zhou, "Region growing for image segmentation using an extended PCNN model", IET Image Processing, 12(5), pp.729~737 (2018)
- L. Ding, "On the Canny edge detector", Pattern Recognition, 34(3), pp.721~725 (2001)
- Y. M. Luo, "Canny edge detection on NVIDIA CUDA", Proceedings of 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (2008)
- A. K. Cherri, "Optical symbolic substitution : edge detection using Prewitt, Sobel, and Roberts operators", Applied Optics, 28(21), pp.4644~4648 (1989)
- S. Vicente, "Graph cut based image segmentation with connectivity priors", Proceedings of 2008 IEEE Conference on Computer Vision and Pattern Recognition (2008)
- C. Rother, "GrabCut : interactive foreground extraction using iterated graph cuts", ACM Transactions on Graphics, 23(3), pp.309~314 (2004)
- K. He, "Mask R-CNN", IEEE Transactions on Pattern Analysis and Machine Intelligence, 42(1), pp.386~397 (2020)
- L. C. Chen, "DeepLab : semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs", IEEE Transactions on Pattern Analysis and Machine Intelligence, 40(4), pp.834~848 (2017)
- X. Li, "H-DenseUNet : hybrid densely connected UNet for liver and tumor segmentation from CT volumes", IEEE Transactions on Medical Imaging, 37(12), pp.2663~2674 (2018)
- S. Loutridis, "A two-dimensional wavelet transform for detection of cracks in plates", Engineering Structures, 27(9), pp.1327~1338 (2005)
- S. Balla-Arabe, "Architecture-driven level set optimization : From clustering to subpixel image segmentation", IEEE Transactions on Cybernetics, 46(12), pp.3181~3194 (2016)
- Z. Zhuge, "Fabric image segmentation algorithm based on Mean Shift", Journal of Textile Research, 28(10), pp.108~111 (2007)
- L. Lin, "Adaptive pixon represented segmentation(APRS)for 3D MR brain images based on mean shift and Markov random fields", Pattern Recognition Letters, 32(7), pp.1036~1043 (2011)
- J. Sun, "A fast MEANSHIFT algorithm-based target tracking system", Sensors, 12(6), pp.8218~8235 (2012)
- Y. Liu, "Dynamics of a mean-shift-like algorithm and its applications on clustering", Information Processing Letters, 113(1-2), pp.8~16 (2013)
- M. H. Jeong, "Adaptive mean-shift tracking with novel color model", Proceedings of IEEE International Conference Mechatronics and Automation, pp.1329~1333 (2005)
- H. Cho, "Image segmentation using linked mean-shift vectors and its implementation on GPU", IEEE Transactions on Consumer Electronics, 60(4), pp.719~727 (2014)
- K. Du, "Object tracking based on improved MeanShift and SIFT", Proceedings of 2012 2nd International Conference on Consumer Electronics, pp.2716~2719 (2012)
- G. B. Li, "Weighted fragments-based mean-shift tracking using color-texture histogram", Journal of Computer-Aided Design & Computer Graphics, 23(12), pp.2059~2066 (2011)
- C. Liu, "Adaptive image segmentation by using mean-shift and evolutionary optimization", IET Image Processing, 8(6), pp.327~333 (2014)
- A. Mayer, "An adaptive mean-shift framework for MRI brain segmentation", IEEE Transactions on Medical Imaging, 28(8), pp.1238~1250 (2009)
- J. Zhou, "An adaptive MeanShift segmentation method of remote sensing images based on multi-dimension features", Geomatics & Information Science of Wuhan University, 37(4), pp.419~418 (2012)
- Y. Wang, "Adaptive Mean Shift based image smoothing and segmentation", Acta Automatica Sinica, 36(12), pp.1637~1644 (2012)
- Y. F. Ge, "A tree image segmentation algorithm based on relative color indices", Journal of Nanjing Forestry University(Natural Science Edition), 2(4), pp.19~22 (2012)
- M. Zhao, "Tree image Segmentation method based on the fractional dimension", Transactions of the Chinese Society for Agricultural Machinery, 35(2), pp.72~75 (2004)
- S. H. Jiang, "Research segmentation methods of stumpage image based on Android System", M. S., pp.2015
- W. Ding, "Measurement methods of fruit tree canopy volume based on machine vision", Transactions of the Chinese Society for Agricultural Machinery, 47(6), pp.1~10 (2016)
- T. Yang, "Multiple trees contour extraction method based on Graph Cut algorithm", Journal of Nanjing Forestry University(Natural Sciences Edition), 42(6), pp.91~98 (2018)