为了提高重点区域的监控安全性，改善图像整体加密速度慢的缺陷，立足于图像局部加密的快速处理方法设计，提出了基于图像感兴趣区域（region of interest，ROI）加密的概念。通过OSTU算法对红外图像进行ROI提取来确定加密区域；然后对Lorenz混沌序列进行数据处理确保序列的随机性，并利用处理后的Lorenz混沌序列对红外图像的ROI进行加密；最后与传统矩阵变换加密进行了详细的加密性能比较，并与其他3种加密算法进行了运算速度比较。实验结果表明，所提加密算法属于局部加密，与整体加密相比，其加密速度大幅提高，同时具有较高的密钥敏感性和较大的密钥空间。
In order to improve the security of monitoring system and the speed of full image encrypting, based on the fast image-block encryption method, the concept of Region of interest encryption for infrared image is advanced in this paper. The OSTU algorithm are used to extract the ROI for encryption, then a Lorenz sequence is processed in order to ensure the randomness. Based on the random Lorenz data sequence, the ROI encryption is processed. Compared to the other encryption methods, the Lorenz ROI encryption has better performance on operation speed, security and encryption effect.
[1] Teng L, Wang X Y, Meng J. A chaotic color image encryption using integrated bit-level permutation[J]. Multimedia Tools and Applications, 2018, 77(6):6883-6896.
[2] Gan Z H, Chai X L, Yuan K, et al. A novel image encryption algorithm based on LFT based S-boxes and chaos[J]. Multimedia Tools and Applications, 2018, 77(7):8759-8783.
[3] Hua Z Y, Jin F, Xu B X, et al. 2D logistic-sine-coupling map for image encryption[J]. Signal Processing, 2018, 149:148-161.
[4] 王聪丽,陈志斌,葛勇.利用Lorenz混沌系统实现红外图像加密的方案[J].计算机应用, 2015, 35(8):2205-2209. Wang C L, Chen Z B, Ge Y. Infrared image encryption scheme using Lorenz chaotic system[J]. Journal of Computer Applications, 2015, 35(8):2205-2209.(in Chinese)
[5] 王聪丽,吴微,张昊.一种红外图像DFT域加密算法[J].信息工程大学学报, 2021, 22(1):32-37. Wang C L, Wu W, Zhang H. Infrared image encryption method based on DFT[J]. Journal of Information Engineering University, 2021, 22(1):32-37.(in Chinese)
[6] Chaabouni I, Fourati W, Bouhlel M S. Using ROI with ISOM compression to medical image[J]. International Journal of Computational Vision and Robotics, 2016, 6(1/2):65.
[7] 刘刚,张晶,李月龙.基于最大内切圆算法的手掌静脉ROI提取[J].计算机科学, 2018, 45(8):264-267, 299. Liu G, Zhang J, Li Y L. Extraction of palm vein ROI based on maximal inscribed circle algorithm[J]. Computer Science, 2018, 45(8):264-267, 299.(in Chinese)
[8] 张琪昌,王洪礼,竺致文,等.分岔与混沌理论及应用[M].天津:天津大学出版社, 2005.
[9] Otsu N. A threshold selection method from gray-level histograms[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1979, 9(1):62-66.
[10] 吴成茂,田小平.三维不等长Arnold变换及其在图像置乱中的应用[J].计算机辅助设计与图形学学报, 2010, 22(10):1831-1840. Wu C M, Tian X P. 3-dimensional non-equilateral Arnold transformation and its application in image scrambling[J]. Journal of Computer-Aided Design&Computer Graphics, 2010, 22(10):1831-1840.(in Chinese)
