·234· 智能系统学报 第13卷 5结束语 mensional Otsu image segmentation method and fast recurs- ive realization[J].Journal of electronics and information 本文提出的基于直线截距直方图倒数灰度熵 technology,2010,32(5):1100-1104. 和ABC的多阈值分割方法，以城区遥感图像像素 [7]RANJANI JJ,THIRUVENGADAM S J.Fast threshold se- 的二维联合信息为基础建立直线截距直方图，以此 lection algorithm for segmentation of synthetic aperture 将阈值搜索空间转化为一维。另一方面，以倒数灰 radar images[J].IET radar,sonar and navigation,2012,6(8): 度熵作为城区遥感图像的阈值选取准则函数，并采 788-795. 用ABC算法对最优阈值的搜寻进行优化，大大减 [8]张金矿，吴一全.基于Tent映射CPSO的二维斜分指数嫡 少了方法的运行时间。本文方法是对基于嫡理论的 阈值分割.信号处理，2010,26(5)：703-708. 城区遥感图像分割技术的进一步扩展和补充。实验 ZHANG Jinkuang,WU Yiquan.Image thresholding based 结果表明，与KFCM聚类分割法、基于PSO的指数 on 2-D oblique exponent entropy method and Tent map chaotic particle swarm algorithm[J].Signal processing, 嫡阈值分割法等进行对比，在主观视觉效果和客观 2010,26(5):703-708 评价指标两个方面，证明了本文方法的准确性和实 [9]SARKAR S,DAS S,CHAUDHURI S S.A multilevel col- 时性。本文提出的方法，已应用于地物识别中的城 or image thresholding scheme based on minimum cross en- 区、建筑物分割，取得了极佳的分割效果。 tropy and differential evolution[J].Pattern recognition let- 参考文献： ters,2015,54:27-35 [10]MALYSZKO D,STEPANIUK J.Adaptive multilevel [1]SIRMACEK B,UNSALAN C.Urban area detection using rough entropy evolutionary thresholding[J].Information local feature points and spatial voting[J].IEEE geoscience sciences,2010,180(7):1138-1158 and remote sensing letters,2010,7(1):146-150 [11]NIAZMARDI S,NAEINI AA,HOMAYOUNI S,et al. [2]SIRMACEK B,UNSALAN C.Using local features to Particle swarm optimization of kernel-based fuzzy C- measure land development in urban regions[J].Pattern re- means for hyperspectral data clustering[J].Journal of ap- cognition letters,2010,31(10):1155-1159. plied remote sensing,2012,6(1):063601. [3]朱江洪，李江风，叶菁.利用决策树工具的土地利用类型 [12]KAPUR J N,SAHOO P K,WONG A K C.A new method 遥感识别方法研究).武汉大学学报：信息科学版，2011， for gray-level picture thresholding using the entropy of the 36(3):301-305. histogram[J].Computer vision,graphics,and image pro- ZHU Jianghong,LI Jiangfeng,YE Jing.Land use informa- cessing,.1985,29(3:273-285. tion extraction from remote sensing data based on decision [13]CAO L,SHI Z,CHENG E K W.Fast automatic multilevel tree tool[J].Geomatics and information science of Wuhan thresholding method[J].Electronics letters,2002,38(16): university,2011,36(3):301-305. 868-870 「4]陈洪，陶超，邹峥嵘，等.一种新的高分辨率遥感影像城区 [14]吴一全，孟天亮，吴诗婳，等.基于二维倒数灰度嫡的河 提取方法[J】.武汉大学学报：信息科学版，2013,38(9)： 流遥感图像分割[.华中科技大学学报：自然科学版， 1063-1067 2014,42(12):70-74,80 CHEN Hong,TAO Chao,ZOU Zhengrong,et al.Automat- WU Yiquan,MENG Tianliang,WU Shihua,et al.Remote ic urban area extraction using a Gabor filter and high-resolu- sensing images segmentation of rivers based on two-di- tion remote sensing imagery[].Geomatics and information mensional reciprocal gray entropy[J].Journal of Huazhong science of Wuhan university,2013,38(9):1063-1067. university of science and technology:nature science,2014, [5]李丽，柴文婷，梅树立.基于自适应全局阈值融合标记的 42(12):70-74,80. 遥感图像建筑群分割[J].农业机械学报，2013,44(7)： [15]陈恺，陈芳，戴敏，等.基于萤火虫算法的二维嫡多阈值 222-228 快速图像分割).光学精密工程，2014,22(2：517-523. LILi,CHAI Wenting,MEI Shuli.Segmentation of remote CHEN Kai,CHEN Fang,DAI Min,et al.Fast image seg- sensing images based on adaptive global threshold and fused mentation with multilevel threshold of two-dimensional en- markers[].Transactions of the Chinese society for agricul- tropy based on firefly algorithm[J].Optics and precision tural machinery,2013.44(7):222-228. engineering,2014,22(2:517-523. [6陈琪，熊博莅，陆军，等.改进的二维Ots图像分割方法 「16]罗希平，田捷.用最大嫡原则作多阈值选择的条件迭代 及其快速实现[J.电子与信息学报，2010,32(5)： 算法.软件学报，2000,11(3：379-385. 1100-1104 LUO Xiping,TIAN Jie.The ICM algorithm for multi-level CHEN Qi,XIONG Boli,LU Jun,et al.Improved Two-Di- threshold selection by maximum entropy criterion[J].5 结束语 本文提出的基于直线截距直方图倒数灰度熵 和 ABC 的多阈值分割方法，以城区遥感图像像素 的二维联合信息为基础建立直线截距直方图，以此 将阈值搜索空间转化为一维。另一方面，以倒数灰 度熵作为城区遥感图像的阈值选取准则函数，并采 用 ABC 算法对最优阈值的搜寻进行优化，大大减 少了方法的运行时间。本文方法是对基于熵理论的 城区遥感图像分割技术的进一步扩展和补充。实验 结果表明，与 KFCM 聚类分割法、基于 PSO 的指数 熵阈值分割法等进行对比，在主观视觉效果和客观 评价指标两个方面，证明了本文方法的准确性和实 时性。本文提出的方法，已应用于地物识别中的城 区、建筑物分割，取得了极佳的分割效果。 参考文献： SIRMACEK B, UNSALAN C. Urban area detection using local feature points and spatial voting[J]. IEEE geoscience and remote sensing letters, 2010, 7(1): 146–150. [1] SIRMAÇEK B, ÜNSALAN C. Using local features to measure land development in urban regions[J]. Pattern re￾cognition letters, 2010, 31(10): 1155–1159. [2] 朱江洪, 李江风, 叶菁. 利用决策树工具的土地利用类型 遥感识别方法研究[J]. 武汉大学学报: 信息科学版, 2011, 36(3): 301–305. ZHU Jianghong, LI Jiangfeng, YE Jing. Land use informa￾tion extraction from remote sensing data based on decision tree tool[J]. Geomatics and information science of Wuhan university, 2011, 36(3): 301–305. [3] 陈洪, 陶超, 邹峥嵘, 等. 一种新的高分辨率遥感影像城区 提取方法[J]. 武汉大学学报:信息科学版, 2013, 38(9): 1063–1067. CHEN Hong, TAO Chao, ZOU Zhengrong, et al. Automat￾ic urban area extraction using a Gabor filter and high-resolu￾tion remote sensing imagery[J]. Geomatics and information science of Wuhan university, 2013, 38(9): 1063–1067. [4] 李丽, 柴文婷, 梅树立. 基于自适应全局阈值融合标记的 遥感图像建筑群分割[J]. 农业机械学报, 2013, 44(7): 222–228. LI Li, CHAI Wenting, MEI Shuli. Segmentation of remote sensing images based on adaptive global threshold and fused markers[J]. Transactions of the Chinese society for agricul￾tural machinery, 2013, 44(7): 222–228. [5] 陈琪, 熊博莅, 陆军, 等. 改进的二维 Otsu 图像分割方法 及其快速实现[J]. 电子与信息学报, 2010, 32(5): 1100–1104. CHEN Qi, XIONG Boli, LU Jun, et al. Improved Two-Di- [6] mensional Otsu image segmentation method and fast recurs￾ive realization[J]. Journal of electronics and information technology, 2010, 32(5): 1100–1104. RANJANI J J, THIRUVENGADAM S J. Fast threshold se￾lection algorithm for segmentation of synthetic aperture radar images[J]. IET radar, sonar and navigation, 2012, 6(8): 788–795. [7] 张金矿, 吴一全. 基于 Tent 映射 CPSO 的二维斜分指数熵 阈值分割[J]. 信号处理, 2010, 26(5): 703–708. ZHANG Jinkuang, WU Yiquan. Image thresholding based on 2-D oblique exponent entropy method and Tent map chaotic particle swarm algorithm[J]. Signal processing, 2010, 26(5): 703–708. [8] SARKAR S, DAS S, CHAUDHURI S S. A multilevel col￾or image thresholding scheme based on minimum cross en￾tropy and differential evolution[J]. Pattern recognition let￾ters, 2015, 54: 27–35. [9] MAŁYSZKO D, STEPANIUK J. Adaptive multilevel rough entropy evolutionary thresholding[J]. Information sciences, 2010, 180(7): 1138–1158. [10] NIAZMARDI S, NAEINI A A, HOMAYOUNI S, et al. Particle swarm optimization of kernel-based fuzzy C￾means for hyperspectral data clustering[J]. Journal of ap￾plied remote sensing, 2012, 6(1): 063601. [11] KAPUR J N, SAHOO P K, WONG A K C. A new method for gray-level picture thresholding using the entropy of the histogram[J]. Computer vision, graphics, and image pro￾cessing, 1985, 29(3): 273–285. [12] CAO L, SHI Z, CHENG E K W. Fast automatic multilevel thresholding method[J]. Electronics letters, 2002, 38(16): 868–870. [13] 吴一全, 孟天亮, 吴诗婳, 等. 基于二维倒数灰度熵的河 流遥感图像分割[J]. 华中科技大学学报: 自然科学版, 2014, 42(12): 70–74, 80. WU Yiquan, MENG Tianliang, WU Shihua, et al. Remote sensing images segmentation of rivers based on two-di￾mensional reciprocal gray entropy[J]. Journal of Huazhong university of science and technology: nature science, 2014, 42(12): 70–74, 80. [14] 陈恺, 陈芳, 戴敏, 等. 基于萤火虫算法的二维熵多阈值 快速图像分割[J]. 光学精密工程, 2014, 22(2): 517–523. CHEN Kai, CHEN Fang, DAI Min, et al. Fast image seg￾mentation with multilevel threshold of two-dimensional en￾tropy based on firefly algorithm[J]. Optics and precision engineering, 2014, 22(2): 517–523. [15] 罗希平, 田捷. 用最大熵原则作多阈值选择的条件迭代 算法[J]. 软件学报, 2000, 11(3): 379–385. LUO Xiping, TIAN Jie. The ICM algorithm for multi-level threshold selection by maximum entropy criterion[J]. [16] ·234· 智 能 系 统 学 报 第 13 卷