Video service system, video service device and extraction method for key frame thereof

Abstract

The invention discloses an extraction method for a key frame, applied to extraction of the key frame of a video data flow in a video service system. The extraction method is characterized by comprising the following steps: obtaining a motion vector of each frame in the video data stream, and obtaining a feature vector set of the motion vector of each frame; extracting the key frame by determining whether directions and amplitudes of the motion vectors corresponding to the feature vector sets of two adjacent front and back frames are changed. The invention also provides a video service device and the video service system. The invention can efficiently extract a frame with suddenly changed speed, lower the error rate and the complexity for extracting the key frame, and reduce the calculation amount.

Description

technical field [0001] Embodiments of the present invention relate to the field of communication technologies, and in particular, to a video service system, a video service device and a method for extracting key frames thereof. Background technique [0002] When people observe the objective world, they are often most interested in abnormal events and obtain a lot of information from abnormal changes. For an object, an abnormal state means that the motion state of the object has changed significantly, such as from static to motion, from motion to static, the direction of motion changes, or the speed of motion changes significantly. Similarly, when people watch a video scene, they pay attention to the change of the scene, and the change of the video scene is the reflection of the significant change in the motion state of the object in the scene. The scene change also includes the switching of the scene. It can be considered that the switching of the scene is that the object i...

AI Technical Summary

Problems solved by technology

[0005] However, in the process of realizing the present invention, the inventor found that there are at least the following problems in the prior art: first, prior art one needs to extract the contour information of all moving objects, and perform operations on the contour information, because the algorithmic process of extracting the contour information It is very complicated, thus causing a large amount of calculation in the prior art; secondly, the prior art one is to extract key frames that change from static to motion or from motion to static, but cannot extract

[0007] However, in the process of realizing the present invention, the inventors found that there are at least the following problems in the prior art: first, the prior art 2 does not consider the directionality problem, that is, the prior art 2 cannot reflect a moving object with a uniform speed but a changing direction of motion, Therefore, the corresponding key frame cannot be extracted; secondly, when there are multiple moving objects in the kth frame, the motion vector information of individual moving objects is very large, and the motion vector information of individual moving objects is very small, after the average value is calculated , may cause the perceived motion energy value of the kth frame to be very small, which cannot well reflect the changes of the kth frame, which will lead to misjudgment of the key frame, that is, the kth frame cannot be selected as the key frame

Images