基于预测误差扩展的加密域点云模型可逆信息隐藏算法  

Reversible information hiding algorithm for cryptographic domain point cloud models based on prediction error expansion

作　　者：欧跃发 邓维 任帅[4] 程慧荣 OU Yuefa;DENG Wei;REN Shuai;CHENG Huirong(School of Mechanical and Marine Engineering,Beibu Gulf University,Qinzhou 535011;Guangxi Key Laboratory of Ocean Engineering Equipment and Technology,Beibu Gulf University,Qinzhou 535011;School of Computer Science&Engineering,Guilin University of Aerospace Technology,Guilin 541004;School of Information Engineering,Chang’an University,Xi’an 710064)

机构地区：[1]北部湾大学机械与船舶海洋工程学院,广西钦州535011 [2]北部湾大学广西海洋工程装备与技术重点实验室,广西钦州535011 [3]桂林航天工业学院计算机科学与工程学院,广西桂林541004 [4]长安大学信息工程学院,西安710064

出　　处：《重庆邮电大学学报（自然科学版）》2024年第6期1128-1139,共12页Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)

基　　金：国家自然科学基金项目(61702050);广西高校中青年教师科研基础能力提升项目(2024KY0447);2022年度广西职业教育教学改革研究项目(GXGZJG2022B006)。

摘　　要：在现代云计算环境中,对于医疗诊断和三维地质模型等应用场景,保证数据安全的同时实现数据的无损恢复尤为重要。针对这一挑战,提出了基于预测误差扩展的加密域点云模型可逆信息隐藏算法。利用新颖的混沌系统对点云模型进行加密,以确保模型内容在上传到云端存储时的安全性;通过贪心算法对模型顶点进行有效分类,并计算每个顶点的预测误差;利用预测误差模长的扩展技术,使秘密信息安全地嵌入到点云模型中;在接收端,通过对比预测误差模长实现秘密信息的准确提取,并无损恢复原始点云模型。实验验证显示,相比现有技术,提出的算法在嵌入性能上有着显著提升,平均嵌入率较对比算法分别提高了0.284和0.298 bit/vertex,明显优化了信息嵌入的效率和安全性;保证了点云模型的无损恢复能力,实现了对算法可逆性的要求;在确保模型内容安全的同时,有效提高了秘密信息的嵌入性能和提取精度。该算法对于医疗诊断、地质探索等需要高度数据安全和完整性恢复的应用场景,提供了新的解决方案。In the contemporary cloud computing environment,ensuring data security while achieving lossless recovery of data is especially critical for applications such as medical diagnosis and three-dimensional geological modeling.This paper proposes an encryption-domain reversible information hiding algorithm for point cloud models based on prediction error expansion,aimed at addressing this challenge.The essence of the algorithm lies in initially encrypting the point cloud model with a novel chaotic system to secure the model content before uploading it to cloud storage.Subsequently,the model vertices are effectively classified using a greedy algorithm,and the prediction error for each vertex is calculated.By extending the magnitude of the prediction error,secret information is securely embedded into the point cloud model.At the receiving end,the secret information is accurately extracted by comparing the magnitudes of prediction errors,and the original point cloud model is losslessly recovered.Experimental verification has demonstrated that,compared to existing technologies,the proposed algorithm significantly improves embedding performance,with an average embedding rate increase of 0.284 bits per vertex(bpv)and 0.298 bpv,respectively,thus markedly optimizing the efficiency and security of information embedding.More importantly,the algorithm ensures the lossless recovery capability of the point cloud model,fulfilling the requirement for algorithm reversibility.In conclusion,the proposed encryption-domain reversible information hiding algorithm based on prediction error expansion for point cloud models not only secures the content of the model but also effectively enhances the performance of secret information embedding and precision of extraction.This algorithm offers a novel solution for application scenarios such as medical diagnosis and geological exploration that demand high data security and integrity restoration.

关 键 词：信息隐藏 三维点云模型 预测误差扩展 贪心算法 

分 类 号：TP309.7[自动化与计算机技术—计算机系统结构] TN918.91[自动化与计算机技术—计算机科学与技术]