基于直方图区域生长的遥感图像阈值分割算法

doi:10.13474/j.cnki.11-2246.2021.0037

测绘通报 ›› 2021, Vol. 0 ›› Issue (2): 25-29.doi: 10.13474/j.cnki.11-2246.2021.0037

基于直方图区域生长的遥感图像阈值分割算法

刘思言^1,2, 李玲^1,2, 特日根^1,2,3, 李竺强^1,2, 马经宇^1,2, 朱瑞飞^1,2,3

1. 长光卫星技术有限公司, 吉林长春 130000;
2. 吉林省卫星遥感应用技术重点实验室, 吉林长春 130000;
3. 中国科学院长春光学精密机械与物理研究所, 吉林长春 130000

收稿日期:2020-03-31 修回日期:2020-06-15 出版日期:2021-02-25 发布日期:2021-03-09
通讯作者: 朱瑞飞。E-mail:18243188708@163.com
作者简介:刘思言(1992-),男,硕士,助理工程师,研究方向为遥感图像处理与深度学习。E-mail:liusiyan@charmingglobe.com
基金资助:
国家重点研发计划重点专项（2018YFB1004605）；吉林省重点科技研发项目（20180201109GX）

Threshold segmentation algorithm based on histogram region growing for remote sensing images

LIU Siyan^1,2, LI Ling^1,2, TE Rigen^1,2,3, LI Zhuqiang^1,2, MA Jingyu^1,2, ZHU Ruifei^1,2,3

1. Chang Guang Satellite Technology Co., Ltd., Changchun 130000, China;
2. Key Laboratory of Satellite Remote Sensing Application Technology of Jilin Province, Changchun 130000, China;
3. Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130000, China

Received:2020-03-31 Revised:2020-06-15 Online:2021-02-25 Published:2021-03-09

摘要/Abstract

摘要： 传统阈值分割算法从单阈值扩展到多阈值的过程中，时间复杂度会大幅度增加，并且由于遥感图像信息复杂，会导致分割效果降低。为了解决这些问题，本文提出了基于直方图区域生长的遥感图像阈值分割算法。在本文算法中，每一个灰度级均作为1个初始阈值，用256个阈值将直方图分割成256个原始小区域。为了减少阈值数目，本文将小区域合并成大区域，每一次合并都可视为一次区域的生长。在每次生长过程中，选取熵值H最小的区域作为直方图各区域中的主区域，并通过本文提出的预匹配策略将其与相邻区域合并。每一次区域生长后，阈值数目均减少1个。在整个过程中，最多只需要生长255次。算法的时间复杂度稳定在O（L）级别。最后通过单阈值和多阈值试验证明本文算法在运行时间和分割精度上均具有优势。

关键词: 图像分割, 直方图, 区域生长, 合并, 预匹配

Abstract: In the process of traditional thresholding algorithm from single-threshold to multi-threshold, the time complexity will increase greatly, and because of the complexity of remote sensing image information, the segmentation effect will be reduced. This paper creatively proposes the idea based on histogram region growing for remote sensing images. Each gray level is regarded as a threshold, so that the histogram is divided into 256 original small regions by the 256 thresholds. For reducing the number of thresholds, small regions are merged into large regions. Each merging can be regarded as the growth of a region. In each growth, the region with the smallest entropy H is selected as the main one in all regions of the histogram, and then it is merged with the adjacent region by the way of pre-judgment. After growing, the number of thresholds decreases. In the whole process, the growth times are only 255 at most, and the time complexity is stable at O(L). In this paper, single-threshold and multi-threshold experiments show that the algorithm has high accuracy in segmentation results, and has advantages in run time.

Key words: image segmentation, histogram, region growing, merge, pre-judgment

中图分类号:

P237

刘思言, 李玲, 特日根, 李竺强, 马经宇, 朱瑞飞. 基于直方图区域生长的遥感图像阈值分割算法[J]. 测绘通报, 2021, 0(2): 25-29.

LIU Siyan, LI Ling, TE Rigen, LI Zhuqiang, MA Jingyu, ZHU Ruifei. Threshold segmentation algorithm based on histogram region growing for remote sensing images[J]. Bulletin of Surveying and Mapping, 2021, 0(2): 25-29.

参考文献

[1] PUN T. A new method for grey-level picture thresholding using the entropy of the histogram[J]. Signal Processing, 1980, 2(3):223-237.
[2] OTSU N. A threshold selection method from gray-level histograms[J]. IEEE Transactions on Systems, Man, and Cybernetics, 2007, 9(1):62-66.
[3] 丁海勇, 王雨轩, 毛宇琼,等. 基于动态阈值区域分裂合并算法的高分辨率遥感图像分割研究[J]. 测绘通报, 2016(8):145-146.
[4] UGOLOTTI R, CAGNONI S. Multi-objective parameter tuning for PSO-based point cloud localization[C]//Advances in Artifical Life and Evolutionary Computation.[S.l.]:Springer, 2014.
[5] 申铉京, 刘翔, 陈海鹏. 基于多阈值Otsu准则的阈值分割快速计算[J]. 电子与信息学报, 2017, 39(1):144-149.
[6] SAHOO P K, KANUNGO P, MISHRA S. A fast valley-based segmentation for detection of slowly moving objects[J]. Signal Image & Video Processing, 2018,12:1265-1272.
[7] 袁玉珠. 改进分数阶达尔文粒子群的多Renyi熵图像分割算法[J]. 测绘通报, 2019(6):34-40.
[8] WANG L, CHEN G Q, SHI D, et al. Active contours driven by edge entropy fitting energy for image segmentation[J]. Signal Processing, 2018, 149:27-35.
[9] 张新明, 孙印杰, 郑延斌. 二维直方图准分的Otsu图像分割及其快速实现[J]. 电子学报, 2011,39(8):1778-1784.
[10] 林文杰,李玉,赵泉华. 结合MST划分和RHMRF-FCM算法的高分辨率遥感图像分割[J]. 测绘学报, 2019, 48(1):64-74.
[11] 杨绪业, 李傲雪, 徐帅婧,等. 基于最小类平均绝对偏差算法的遥感图像分割[J]. 中国激光,2014, 41(S1):S109011.
[12] CHANG H H, ZHUANG A H, VALENTINO D J, et al. Performance measure characterization for evaluating neuroimage segmentation algorithms[J]. NeuroImage, 2009, 47(1):122-135.
[13] TAHA A A, HANBURY A. Metrics for evaluating 3D medical image segmentation:analysis, selection, and tool[J]. BMC Medical Imaging, 2015, 15(1):29.
[14] MANIKANDAN S, RAMAR K, IRUTHAYARAJAN M W, et al. Multilevel thresholding for segmentation of medical brain images using real coded genetic algorithm[J]. Measurement, 2014, 47(1):558-568.
[15] 韦春桃, 赵平, 肖博林,等. 结合双树复小波纹理特征和MRF模型的遥感图像分割[J]. 测绘通报, 2019(10):40-45.

基于直方图区域生长的遥感图像阈值分割算法

Threshold segmentation algorithm based on histogram region growing for remote sensing images

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	陈智朗, 付振华, 朱紫阳, 王慧慧, 刘沁雯, 杨钰灵, 许耿然. 基于HRNet的高分辨率遥感影像建筑物变化信息提取[J]. 测绘通报, 2022, 0(5): 126-132.
[2]	张伯树, 张志华, 张洋. 改进的HRNet应用于路面裂缝分割与检测[J]. 测绘通报, 2022, 0(3): 83-89.
[3]	李少先. 利用双向布料模拟的机载建筑物点云提取[J]. 测绘通报, 2022, 0(3): 148-151.
[4]	宋锦阳, 钱建国, 汤圣君, 王伟玺, 李晓明. 三维点云室内平面要素提取与优化[J]. 测绘通报, 2021, 0(9): 15-20,27.
[5]	侯海倩, 唐超, 赵丽凤, 张昊. 轨道扣件病害自动化检测研究及应用[J]. 测绘通报, 2021, 0(8): 88-92.
[6]	刘茂华, 李曼雯. 融合高分影像和LiDAR数据的城市道路提取[J]. 测绘通报, 2021, 0(6): 1-5,60.
[7]	欧阳犬平, 周立, 王圣, 宋士林, 芦斌, 冯俊俊. 基于霍夫直线检测和区域生长的合成孔径声呐图像线目标提取[J]. 测绘通报, 2021, 0(11): 48-53.
[8]	李亚丽, 张松林, 韩杰. 基于小波系数分割的局部自适应阈值图像去噪[J]. 测绘通报, 2020, 0(5): 43-46,122.
[9]	戚爽, 喻光继. 多级存储优化的大规模全局光照快速计算[J]. 测绘通报, 2020, 0(3): 64-68.
[10]	张锦, 赖祖龙, 孙杰. Otsu法、区域生长法及形态学相结合的遥感图像海岸线提取[J]. 测绘通报, 2020, 0(10): 89-92.
[11]	陈世增, 李必军, 周继苗. 利用双灭点估计的车道线检测[J]. 测绘通报, 2020, 0(1): 16-20.
[12]	袁玉珠. 改进分数阶达尔文粒子群的多Renyi熵图像分割算法[J]. 测绘通报, 2019, 0(6): 34-40.
[13]	朱依民, 田林亚, 毕继鑫, 林松. 基于机载LiDAR数据的建筑物轮廓提取[J]. 测绘通报, 2019, 0(12): 65-70.
[14]	喻龙华, 王雷光, 吴楠, 廖声熙, 张怀清. 复杂地形DSM的地面点识别及DEM提取[J]. 测绘通报, 2018, 0(5): 59-64.
[15]	王永康, 汪云甲, 毕京学, 曹鸿基. 一种基于图像灰度直方图相似度计算的室内定位方法[J]. 测绘通报, 2018, 0(4): 63-67.