Infant behavior monitoring and alarming system and method

Abstract

The invention provides an infant behavior monitoring and alarming system and method, belonging to the field of computer vision. The system comprises a control center host and a plurality of network cameras. According to the method, the plurality of network cameras are arranged in space, and whether a human body is close to a dangerous area or not is judged by processing and analyzing video imagescollected by the network cameras. According to the invention, monitoring of children is realized by using a human joint motion feature tracking technology; when parents work or are out of home, the parents can turn on computers or mobile phones at any time to check real-time images at home, communicate with children face to face, know family conditions, remotely care family members and remotely assist in handling emergencies, so the parents can work more securely.

Description

technical field [0001] The invention belongs to the field of computer vision, and in particular relates to a child behavior monitoring and alarm system and method. Background technique [0002] Driven by a wide range of application requirements, there are more and more researches on capturing and analyzing human motion through video. However, at present, there are still many difficulties in directly and automatically obtaining human motion data from video. The main difficulty is the high complexity of the human body itself and its motion and the inherent information loss in the imaging process of the camera; moreover, the human body may be frequently occluded or self-occluded in the video, and the camera used to acquire the video has different time and space The resolution of the above is limited, and these objective factors also make it still a difficult technical challenge to track and obtain human motion data from video sequences. These difficulties are also the technical...

AI Technical Summary

Problems solved by technology

[0002] Driven by a wide range of application requirements, there are more and more researches on capturing and analyzing human motion through video. However, at present, there are still many difficulties in directly and automatically obtaining human motion data from video.

The main difficulty is the high complexity of the human body itself and its motion and the inherent information loss in the imaging process of the camera; moreover, the human body may be frequently occluded or self-occluded in the video, and the camera used to acquire the video has different time and space The resolution of the above is limited, and these objective factors also make it still a difficult technical challenge to track and obtain human motion data from video sequences. These difficulties are also the key technology of human motion analysis research.

Images