利用SVM相关反馈和语义挖掘的遥感影像检索

¹武汉大学中国南极测绘研究中心,武汉市珞喻路129号430079;²香港中文大学太空与地球信息科学研究所,香港新界沙田;³武汉大学测绘遥感信息工程国家重点实验室,武汉市珞喻路129号430079;⁴黄冈市城乡规划局,黄冈市黄州区西湖三路23号438000

基金项目: 国家海洋局极地专项“测绘遥感技术在极地环境考察与评估中的应用”资助项目(JDZX20110008);;武汉大学青年教师资助项目(3101004)

详细信息

SVM-relevance-feedback and Semantic-extraction-based RS Image Retrieval

¹ Chinese Antarctic Center of Surveying and Mapping,Wuhan University,129 Luoyu Road,Wuhan 430079,China;² Institute of Space and Earth Information Science,the Chinese University of Hong Kong,Shatin,Hong Kong;³ State Key Laboratory of Information Engineering in Surveying,Mapping and Remote Sensing, Wuhan University,129 Luoyu Road,Wuhan,430079,China;⁴Huanggang City Planning Bureau,23 West Lake Road,Huanggang 438000,China

Funds: 国家海洋局极地专项“测绘遥感技术在极地环境考察与评估中的应用”资助项目(JDZX20110008);;武汉大学青年教师资助项目(3101004)

摘要: 针对语义鸿沟问题,将基于语义特征挖掘模型与支持向量机相关反馈方法相结合,建立了基于支持向量机相关反馈的人机交互遥感影像语义检索系统。实验结果表明,该方法利用高层语义特征及人机交互反馈信息缩小了语义鸿沟,提高了影像检索的精度。
- 影像检索 /
- 语义挖掘模型 /
- 相关反馈 /
- SVM
Abstract: The semantic gap between high-level human perception and low-level image features becomes the bottleneck in content-based remotely sensed image retrieval technology.To solve this problem,in this research,a human machine interaction(HMI) remotely sensed image retrieval system is built that combines semantic mining model and SVM-based relevance feedback method.The experiments indicate that this method can well narrow semantic gap and improve retrieval precision and recall.
- image retrieval /
- semantic mining model /
- relevance feedback /
- SVM