Similar frame detection method, system and device and readable storage medium

Abstract

The invention discloses a similar frame detection method. According to the invention, the similar frames are determined in a real matching manner; A similar frame detection model constructed on the basis of a deep learning algorithm is introduced to perform preliminary detection on each pair of frame point clouds belonging to similar frames; In the preliminary detection stage, most of frame pointcloud pairs of non-similar frames can be filtered out, so that the number of frame point cloud pairs which really adopt a matching mode to determine the similar frames is greatly reduced. Compared with a mode of completely judging similar frames based on matching, the time consumption required by similar frame detection based on the model is relatively short, so that the matching time consumptioncan be greatly reduced, and the effective utilization rate of computing resources is improved. The invention further discloses a similar frame detection system and device and a computer readable storage medium which have the above beneficial effects. 本申请公开了一种相似帧的检测方法，本申请在真正采用匹配的方式确定相似帧之前，引入了基于深度学习算法构建得到的相似帧检测模型来对是否属于相似帧的每对帧点云进行初步检测，在初步检测阶段将能够滤除绝大部分非相似帧的帧点云对，使得真正采用匹配的方式进行相似帧确定的帧点云对数得到了极大的减少。相比于全部基于匹配的方式进行相似帧的判断，基于模型的相似帧检测所需耗时本身较短，因此可极大的减少匹配耗时，提高对计算资源的有效利用率。本申请还同时公开了一种相似帧的检测系统、装置及计算机可读存储介质，具有上述有益效果。