基于PLSA和BoW的高分遥感影像小型港口检测  

Small Harbor Detection Based on PLSA and BoW in High Resolution Remotely Sensed Imagery

作　　者：毕奇 童心 张济勇 许凯 张涵 秦昆[1] BI Qi;TONG Xin;ZHANG Jiyong;XU Kai;ZHANG Han;QIN Kun(School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China;Zhongnan Branch, Beijing North-Start Digital Remote Sensing Technology Co. Ltd., Wuhan 430070, China;State Grid Economic and Technical Research Institute Co. Ltd., Beijing 102209, China;Faculty of Information Engineering, China University of Geosciences, Wuhan 430070, China)

机构地区：[1]武汉大学遥感信息工程学院,武汉430079 [2]北京洛斯达数字遥感技术有限公司中南分公司,武汉430070 [3]国网经济技术研究院有限公司,北京102209 [4]中国地质大学(武汉)信息工程学院,武汉430070

出　　处：《应用科学学报》2019年第3期301-312,共12页Journal of Applied Sciences

基　　金：国家重点研发计划基金(No.2016YFB0502603);国家电网公司科技项目基金(No.JYYKJXM(2017)-011)资助

摘　　要：高分辨率遥感影像可以为小型港口的监管提供有效途径.针对小型港口形态多样、特征难以描述等问题,研究了一种基于概率潜在语义分析(probabilistic latent semantic analysis, PLSA)模型和词袋(bag of words, BoW)模型的小型港口检测方法.该方法首先提取水岸线以缩小搜索范围;然后将灰度直方图、归一化差分水体指数、分形维数特征引入PLSA模型生成特征描述集,将加速鲁棒特征向量引入BoW模型生成视觉词典;根据以上特征描述集和构建的小型港口样本库训练SVM分类器,利用22幅影像进行小型港口检测实验.实验结果表明,相比于只使用常见单一特征或单一模型,该方法的检测结果更佳,耗时更少.Remotely sensed high resolution imaging is an effective way to monitor small harbors along coastlines. Harbors vary a lot in features and are usually difficult to describe.This paper studies a harbor detection method based on probabilistic latent semantic analysis(PLSA) and bag of words(BoW). Firstly, coastlines are extracted to shrink searching areas. Then grey level histogram, normalized differential water index and fractal dimension features are introduced to PLSA model to generate feature description sets. Eigenvectors of speeded up robust features are introduced to BoW model to generate visual dictionary.Finally, after collecting samples of small harbors, support vector machine(SVM) classifiers are trained based on the above features. Based on the trained classifier, small harbor detection is implemented based on 22 images. Experiments show that the proposed method reaches the best accuracy with relatively low time cost than single-feature models and single PLSA or BoW model.

关 键 词：高分遥感影像 小型港口检测 概率潜在语义分析 词袋模型 水岸线提取 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]