Monitoring method of bursaphelenchus xylophilus based on multi-feature CRF by UAV image

doi:10.13474/j.cnki.11-2246.2019.0223

Abstract

Abstract: The use of UAV remote sensing technology for forest survey can obtain low-cost, high-resolution, high-time-density remote sensing data, especially for small-scale forest pest monitoring. In this paper, we use the small unmanned aerial vehicle as the image acquisition platform, acquire the RGB image, and study the extraction method of red-attacked pine based on the high resolution image. According to the image characteristics, the color and texture features of the image are extracted, and the CRF method is adopted to identify red-attacked pine. The feasibility and effectiveness of the multi-feature conditional random field method in the monitoring of bursaphelenchus xylophilus is verified by comparing the extraction results of multiple classification methods.

Key words: UAV image, feature selection, image classification, CRF, bursaphelenchus xylophilus

CLC Number:

P237

LIU Jincang, WANG Chengbo, CHANG Yuanfei. Monitoring method of bursaphelenchus xylophilus based on multi-feature CRF by UAV image[J]. 测绘通报, 2019, 0(7): 78-82.

References

[1] 王斌,范冬林.深度学习在遥感影像分类与识别中的研究进展综述[J].测绘通报,2019(2):99-102,136.
[2] SUTTON C, MCCALLUM A. An introduction to conditional random fields[J]. Foundations and Trends® in Machine Learning, 2012, 4(4):267-373.
[3] SHOTTON J, WINN J, ROTHER C, et al.TextonBoost for image understanding:multi-class object recognition and segmentation by jointly modeling texture, layout, and context[J]. International Journal of Computer Vision, 2009, 81(1):2-23.
[4] KOHLI P, LADICKY L U, TORR P H S. Robust higher order potentials for enforcing label consistency[J]. International Journal of Computer Vision, 2009, 82(3):302-324.
[5] WEGNER J D, HANSCH R, THIELE A, et al. Building detection from one orthophoto and high-resolution InSAR data using conditional random fields[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2011, 4(1):83-91.
[6] ZHONG Y F, ZHAO J, ZHANG L P. A hybrid object-oriented conditional random field classification framework for high spatial resolution remote sensing imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(11):7023-7037.
[7] ZHONG P, WANG R. Learning conditional random fields for classification of hyperspectral images[J]. IEEE Transactions on Image Processing, 2010, 19(7):1890-1907.
[8] 杨耘,隋立春.条件随机场建模的大尺度空间上下文的高分辨率遥感图像分类[J].测绘通报,2014(8):21-24.
[9] 李娟,姚翰文.我国松材线虫病疫区现状及管理对策[J].中国森林病虫,2019(2):26-28,35.
[10] 李丰毅,余光柱,陈宪超.基于航拍数据的松材线虫病区域检测[J].电脑知识与技术,2018(29):239-241.
[11] 徐誉远,胡爽,王本洋.无人机遥感在我国森林资源监测中的应用动态[J].林业与环境科学,2017(1):97-101.
[12] 陈军丽,黄睿.利用多属性剖面概率融合的高光谱影像分类[J].遥感信息,2019(2):69-74.
[13] 徐芳.基于Gabor纹理描述的遥感影像分割[J].测绘科学,2013,38(1):116-117,124.
[14] HE Y, LUO Y P, HU D C. Unsupervised texture segmentation based on geodesic active regions[J]. Journal of Software, 2007, 18(3):592.
[15] DALAL N, TRIGGS B, SCHMID C. Human detection using oriented histograms of flow and appearance[C]//Proceedings of 2006 European lonference on Computer Vision. Heidelberg:Springer, 2006(3952):428-441.