Visual tracking platform and method for multiple application scenes

Abstract

The invention discloses a visual tracking platform and method for multiple application scenes. The method comprises the following steps: S1, parameters being set according to actual application scenes; s2, setting camera parameters; s3, acquiring information by a camera; s4, performing software and hardware optimization; S5, carrying out resource integration; Step S6, adjusting the camera. In the step S2, the set camera parameters comprise the number of frames and the resolution ratio. According to the invention, the application of a complex scene can be switched and debugged, different detections can be carried out on the same system platform, and the hardware cost and difficulty are reduced; a user only needs to carry out debugging design according to a scene currently needed by the user, so that the working efficiency is improved; according to the double-flow asynchronous algorithm, continuous calculation processing is independent, a new processing flow is started for each processing, the parallel capability of calculation hardware can be fully utilized, and mutual restriction of front and back data processing is reduced from a time domain; and the maximum multiplexing of the hardware module is realized.

Description

technical field [0001] The invention relates to a visual tracking system, in particular to a visual tracking platform and method for multiple application scenarios. Background technique [0002] Vision is one of the important ways for humans to perceive the world, and 80% of the external information that humans obtain comes from the visual system. Computational vision is based on the understanding of human vision, using imaging systems to replace human visual organs, and computing to replace the human brain to complete the processing and understanding of input images. At the same time, with the development of information technology and intelligent science, computer vision is one of the popular subjects in the field of artificial intelligence and one of the important technologies in the perception layer of the Internet of Things. [0003] As one of the hot topics in the field of computer vision, visual tracking technology is to detect moving targets, extract features, classi...

AI Technical Summary

Problems solved by technology

[0005] The existing image processing technology can basically recognize objects, face detection, face recognition, and object tracking. However, the current artificial intelligence system can only perform visual recognition and tracking for a single situation. It is difficult to switch and debug the application of complex scenarios. For example, the established scenario is to perform face recognition on users, but the actual application scenario needs to detect people without masks, and different system platforms need to be used for different detections. Increased hardware cost and development difficulty, users need to debug and design for different scenarios, reducing work efficiency

Images