A Secure and Efficient Remote Sensing Image Retrieval Method With Verifiable and Traceable in Cloud Environment
Zhaoyang Hou, Haowen Yan, Liming Zhang, Ronghua Ma, Qingbo Yan, Bingbing Yang · IEEE Transactions on Geoscience and Remote Sensing · 2025
Performing privacy-preserving remote sensing image retrieval-based tasks in cloud environments has gained widespread attention. This has raised concerns about the efficiency and accuracy of retrieval, the security of image content, the privacy of the retrieval process, the traceability of data distribution, and the authenticity of retrieval results. Therefore, in this article, we propose a secure and efficient remote sensing image retrieval method with verifiable and traceable in the cloud environment. The method applies the convolutional neural network (CNN) model to extract features. Then, it performs dimensionality reduction mapping based on spectral hashing with spectral rotation (SHSR) and clustering into meaningful clusters to improve retrieval efficiency and accuracy. Then, secure encrypted searchable indexes are generated based on the modified tuples of asymmetric scalar-product preserving encryption (ASPE) to realize privacy protection in the retrieval process. Image encryption is performed after making space for watermark embedding through prediction error (PE) labeling and pixel rearrangement to ensure the security of image content. The process allows flexible embedding and extraction of copyright information and user information, which facilitates copyright authentication and tracing of illegally distributed data. In addition, the authenticity of the retrieval results is verified by constructing a Merkel tree. The security analysis and performance evaluation of the experiments verify the accuracy, privacy, and usability of the proposed method to achieve secure and efficient remote sensing image retrieval.