Research on Key Frame Extraction of Digital Image Based on Unsupervised Clustering Algorithm

At present, video monitoring technology is widely used in agriculture, transportation, behavior analysis and other fields. The mode of manually browsing video to obtain information can no longer meet the demand. Combining key frame extraction with image processing and analysis can greatly improve the automation and intelligence level of management and decision-making. Aiming at the problems existing in the current digital key frame tracking, in order to obtain more ideal digital image key frame tracking results, this paper proposes a digital image key frame tracking method based on adaptive unsupervised clustering algorithm, which uses clustering algorithm to classify digital image key frames according to their features, and uses adaptive unsupervised learning algorithm to track digital image key frames. The simulation results show that, the method in this paper achieves an ideal key-frame tracking effect of digital image, and the tracking error of key-frame of digital image is much smaller than other key-frame tracking methods, which improves the tracking efficiency of key-frame of digital image. The comparative results verify the effectiveness and superiority of unsupervised clustering algorithm for key-frame tracking of digital image.