Camera Image Registration Method for Autonomous Patrol Video Recognition Based on Multiscale Feature Fusion
Aiming at the problem of low registration accuracy caused by the structural similarity of autonomous patrol video recognition camera images,a method for autonomous patrol video recognition camera image registration based on multi-scale feature fusion is proposed.On the basis of image enhancement,Harris corner detection method is used to obtain the structural features of autonomous patrol video recognition camera images,and based on the feature vector construction results,a multi-scale fusion of image features is performed.Design an image registration transformation estimation model to achieve autonomous patrol video recognition camera image registration.The results show that the proposed method can effectively shorten the matching time and make the fusion results more accurate in feature match-ing,with good registration results.
NETL
NSTL
万方数据
