科研之友
微信
新浪微博
Facebook
分享链接
摘要
Aiming to solve the problem of real time encryption transmission on 3D object collision detection process, the paper transforms the objects to be detected into two-dimensional mapping forms in geometry image, extracts the maximum points and minimum points of the original objects by AABB bounding box. The maximum points and minimum points list is mapped to another list-image which generates a gaussian scale-space, the octave images and the scale of the kernel are encrypted with the improved RC5 algorithm and transmitted to the receiver through the public channel; conducts quad tree partition on overlapping sub-elements, then the value and index number of the overlapping sub-nodes are encrypted in RC5 and transmitted to the receiver in batches. Through the encrypted transmission experiments of rigid objects and deformed objects in real-time collision detection, it is shown that the new algorithm is more robust with the comparison between the traditional encrypted transmission algorithm process, the stability and efficiency of the new algorithm is also showed.
