Semi-U-Net: A Lightweight Deep Neural Network for Subject-Sensitive Hashing of HRRS Images

Kaimeng Ding,Shoubao Su,Tingting Jiang,Nan Xu

doi:10.1109/access.2021.3074055

Abstract

As a special case of perceptual hashing algorithm, subject-sensitive hashing can realize “subject-biased” integrity authentication of high resolution remote sensing (HRRS) images, which overcomes the deficiencies of existing integrity authentication technologies. However, the existing deep neural network for subject-sensitive hashing have disadvantages such as high model complexity and low computational efficiency. In this paper, we propose an efficient and lightweight deep neural network named Semi-U-net to achieve efficient subject-sensitive hashing. The proposed Semi-U-net realizes the lightweight of the network from three aspects: First, considering the general process of perceptual hashing, it adopts a semi-u-shaped structure, which simplify the model structure and prevent the model from extracting too much redundant information to enhance the robustness of the algorithm; Second, the number of model parameters and the computational cost are significantly reduced by using deep separable convolution in the entire asymmetric network; Third, the number of model parameters is further compressed by using the dropout layer several times. The experimental results show that the size of our Semi-U-Net model is only 5.38M, which is only 1/27 of MUM-net and 1/15 of MultiResUnet. The speed of the Semi-U-Net based subject-sensitive hashing algorithm is 88.6 FPS, which is 2.89 times faster than MultiResUnet based algorithm and 2.1 times faster than MUM-net Based Algorithm. FLOPs of Semi-U-net is only 1/28 of MUM-net and 1/16 of MultiResUnet.

Highlights

The extraction and analysis of earth surface features through high resolution remote sensing (HRRS) images has received extensive research, such as the buildings extraction [1]–[3], vegetation detection [4]–[6], urban expansion analysis [7]–[9] and detection of land cover changes [10]
In order to overcome the shortcomings of existing methods, such as excessive storage space and high computational complexity, we propose an effective lightweight deep neural network model Semi-U-net for subject-sensitive hashing of HRRS images
The integrity authentication process is implemented at the receiving end: the subject-sensitive hashing algorithm is used to generate the hash sequence of the HRRS image to be authenticated, and the generated hash value is compared with the received hash value

Summary

INTRODUCTION

The extraction and analysis of earth surface features through high resolution remote sensing (HRRS) images has received extensive research, such as the buildings extraction [1]–[3], vegetation detection [4]–[6], urban expansion analysis [7]–[9] and detection of land cover changes [10]. K. Ding et al.: Semi-U-Net: A Lightweight Deep Neural Network for Subject-Sensitive Hashing of HRRS Images. Cryptography methods are too sensitive to changes in the binary level of the data: As long as the data changes by one bit, it is regarded as data tampering This sensitivity is detrimental to the integrity authentication of HRRS image. In order to overcome the shortcomings of existing methods, such as excessive storage space and high computational complexity, we propose an effective lightweight deep neural network model Semi-U-net for subject-sensitive hashing of HRRS images. Combining the characteristics of subject-sensitive hash to change the structure of the neural network and reduce the redundancy of the network, which allows the model to avoid extracting too much redundant information to enhance the robustness of the algorithm.

RELATED WORK

SEMI-U-NET BASED SUBJECT-SENSITIVE HASHING ALGORITHM

EXPERIMENTS AND DISCUSSION

CONCLUSION AND FUTURE WORK