基于动态阈值哈希的大规模遥感影像快速内容检索方法

doi:10.13474/j.cnki.11-2246.2019.0247

测绘通报 ›› 2019, Vol. 0 ›› Issue (8): 34-38,53.doi: 10.13474/j.cnki.11-2246.2019.0247

基于动态阈值哈希的大规模遥感影像快速内容检索方法

强永刚¹, 肖志峰², 陈欢欢¹, 闫丽阳²

1. 中国科学技术大学计算机科学与技术学院, 安徽合肥 230027;
2. 武汉大学测绘遥感信息工程国家重点实验室, 湖北武汉 430079

收稿日期:2018-11-01 修回日期:2019-02-15 出版日期:2019-08-25 发布日期:2019-09-06
作者简介:强永刚(1982-),男,博士生,研究方向为机器学习和图像理解。E-mail:ygqiang@mail.ustc.edu.cn
基金资助:
高分专项青年创新基金（GFZX04061502）

Content retrieval of large-scale remote sensing images based on dynamic threshold hashing

QIANG Yonggang¹, XIAO Zhifeng², CHEN Huanhuan¹, YAN Liyang²

1. College of Computer Science and Technology, University of Science and Technology of China, Hefei 230027, China;
2. State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, China

Received:2018-11-01 Revised:2019-02-15 Online:2019-08-25 Published:2019-09-06

摘要/Abstract

摘要： 随着我国遥感对地观测技术的快速发展，接收和存档的遥感影像数据量呈指数级增长，传统的检索方法难以在超大的遥感影像数据量上进行快速内容检索，造成遥感影像检索技术缺乏突破性进展，使得我国遥感影像利用率和利用效率受到限制。本文提出了一种创新的哈希索引方法，该方法根据特征向量的空间分布情况动态生成向量的哈希编码，可对高维的遥感影像特征向量进行低维编码，大大降低了检索计算量，可显著提高大规模遥感影像库内容检索的准确率和效率。在天地图数据集的检索试验表明本文提出方法在准确度和检索效率上均有显著提升，有较大的应用潜力。

关键词: 遥感影像检索, 哈希算法, 特征索引, 降维

Abstract: With the rapid development of remote sensing earth observation technology in China, the amount of remote sensing image data received and archived has increased exponentially. The traditional retrieval methods are difficult to retrieve the large amount of remote sensing image data quickly, resulting in the lack of breakthrough in remote sensing image retrieval technology, the utilization ratio and utilization efficiency of remote sensing images in China are very limited. In this paper, an innovative hash index method is proposed, which generates the hash codes dynamically according to the spatial distribution of the feature vectors. This method can encode the feature vectors of high-dimensional remote sensing images in low dimensions, greatly reduces the amount of retrieval computation and significantly improves the retrieval accuracy and efficiency of large-scale remote sensing image database. The retrieval experiments on the sky map data set show that the proposed method has a significant improvement in accuracy and retrieval efficiency, and has a great application potential.

Key words: remote sensing image retrieval, hash algorithm, feature index, dimensionality reduction

中图分类号:

P237

强永刚, 肖志峰, 陈欢欢, 闫丽阳. 基于动态阈值哈希的大规模遥感影像快速内容检索方法[J]. 测绘通报, 2019, 0(8): 34-38,53.

QIANG Yonggang, XIAO Zhifeng, CHEN Huanhuan, YAN Liyang. Content retrieval of large-scale remote sensing images based on dynamic threshold hashing[J]. 测绘通报, 2019, 0(8): 34-38,53.

参考文献

[1] YANG Y, NEWSAM S. Geographic image retrieval using local invariant features[J]. IEEE Transactions on Geoscience and Remote Sensing, 2013, 51(2):818-832.
[2] 徐长勇,周焰,李德仁. 基于内容的遥感图像检索综述[J]. 武汉理工大学学报(信息与管理工程版), 2003, 25(5):8-12.
[3] 李峰,胡岩峰,曾志明,等. 一种遥感影像基于内容检索模型的研究与设计[J]. 光子学报, 2004, 33(12):1522-1525.
[4] SCHRÖDER M, REHRAUER H, SEIDEL K, et al. Interactive learning and probabilistic retrieval in remote sensing image archives[J]. IEEE Transactions on Geoscience and Remote Sensing, 2000, 38(5):2288-2298.
[5] SHYU C R, KLARIC M, SCOTT G J, et al. GeoIRIS:geospatial information retrieval and indexing system-content mining, semantics modeling, and complex queries[J]. IEEE Transactions on Geoscience and Remote Sensing, 2007, 45(4):839-852.
[6] MAHESWARY P, SRIVASTAVA N. Retrieval of remote sensing images using colour and texture attribute[J]. International Journal of Computer Science and Information Security, 2009, 4(1):1-5.
[7] SHAO Z F, ZHOU W X, CHENG Q M. Remote sensing image retrieval with combined features of salient region[J]. International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, 2014, 40(6):83.
[8] 胡屹群,周绍光,岳顺,等. 利用视觉词袋模型和颜色直方图进行遥感影像检索[J]. 测绘通报, 2017(1):53-57.
[9] INDYK P, MOTWANI R. Approximate nearest neighbors:towards removing the curse of dimensionality[C]//Thirtieth ACM Symposium on Theory of Computing.[S.l.]:ACM, 1998:604-613.
[10] ANDONI A, INDYK P. Near-optimal hashing algorithms for approximate nearest neighbor in high dimensions[C]//Proceedings of the 47th Annual IEEE Symposium on Foundations of Computer Science. Berkeley:IEEE, 2006:459-468.
[11] WAN J, WANG D, HOI S C H, et al. Deep learning for content-based image retrieval:a comprehensive study[C]//Proceedings of the ACM International Conference on Multimedia. Orlando:ACM, 2014:157-166.
[12] ZHOU B, LAPEDRIZA A, XIAO J, et al. Learning deep features for scene recognition using places database[C]//Advances in Neural Information Processing Systems.[S.l.]:CAM, 2014:487-495.
[13] BABENKO A, SLESAREV A, CHIGORIN A, et al. Neural codes for image retrieval[C]//Proceedings of ECCV 2014.[S.l.]:Springer International Publishing, 2014:584-599.
[14] HARRINGTON P. Machine learning in action[M].[S.l.]:Manning, 2012.
[15] RAGINSKY M, LAZEBNIK S. Locality-sensitive binary codes from shift-invariant kernels[C]//Proceedings of Advances in Neural Information Processing Systems.[S.l.]:CAM, 2009:1509-1517.
[16] GONG Y, LAZEBNIK S, GORDO A, et al. Iterative quantization:a procrustean approach to learning binary codes for large-scale image retrieval[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(12):2916-2929.
[17] 曹玉东,刘艳洋,贾旭,等. 基于改进的局部敏感哈希算法实现图像型垃圾邮件过滤[J]. 计算机应用研究, 2016, 33(6):1693-1696.
[18] 夏立超,蒋建国,齐美彬. 基于改进谱哈希的大规模图像检索[J]. 合肥工业大学学报(自然科学版), 2016, 39(8):1049-1054.

基于动态阈值哈希的大规模遥感影像快速内容检索方法

Content retrieval of large-scale remote sensing images based on dynamic threshold hashing

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 4

编辑推荐

Metrics

本文评价

[1]	窦世卿, 陈治宇, 徐勇, 郑贺刚, 苗林林, 宋莹莹. 基于多特征融合与典型降维方法的高光谱影像分类[J]. 测绘通报, 2022, 0(4): 32-36,50.
[2]	徐益峰, 陈昱, 程宝银, 张蒙. 二维哈希算法在地理空间数据批量分幅中的应用[J]. 测绘通报, 2021, 0(9): 140-144.
[3]	宋广钦, 杜正舜, 贺智. 高光谱图像波段选择的改进二进制布谷鸟算法[J]. 测绘通报, 2019, 0(4): 43-48.
[4]	汪少华, 杨婷. 判别流形学习算法的高光谱数据降维与树种识别[J]. 测绘通报, 2018, 0(1): 55-61.