Surveillance video has a wide range of applications in many fields, such as national economic construction, public security information construction, and widely adopted countries worldwide. The infrastructure of the video surveillance system has begun to take shape and is still rapid expansion. As thousands of surveillance cameras monitor and record round the clock, the amount of video data has been explosive growth. Finding the required information in a large number of surveillance videos is undoubtedly a needle in a haystack. Motion is a significant feature of video, and much meaningful visual information is contained in the movement. In many application scenarios, like road traffic monitoring, security for major events, guidance for military aircraft, and autonomous vehicle, people tend to pay more attention to the moving object. The analysis shows that the key frame extraction of surveillance video based on motion analysis has important practical significance. Hence, this book puts forward several key frame extraction methods of surveillance video around capturing the target motion state.