Remote sensing image has the characteristics of multi-temporal, multi-semantic and multi-spectral, and it plays an important role in different industries and military. It is a significant issue that the efficiency and accuracy of ciphertext query on the huge amounts of remote sensing image have direct impact on the remote sensing big data in its general applicability and real-time performance. For ciphertext, secure-retrieval on remote sensing big data is one of the important standards on the usability of remote sensing image. This paper first proposes an encryption/query scheme on remote sensing image including encryption algorithm and query algorithm, and then constructs an image query scheme on the encrypted remote sensing image based on the improved Henon mapping. Based on the imaging principle and multi-band remote sensing image characteristics, the paper takes the improved Henon mapping technique to encrypt the gray value of remote sensing image for each band. According to the characteristics of "remote sensing big data", the process extracts the feature vector by the statistical gray values for each band, and then queries the target remote image based on matching algorithm. Finally, experiments are carried out on encrypting and processing the big data of Landsat 8 remote sensing satellites, the experimental results show that the presented algorithms effectively improve the security and accuracy on the usage of remote sensing image. Meanwhile, the scheme has a low computational complexity and communication cost.