Safety warning method and device

Abstract

The embodiment of the invention discloses a safety warning method and device. The method in the embodiment of the invention comprises the following steps: partitioning the acquired video data to obtain target video data; according to a preset first sampling strategy, sampling the target video data; analyzing the currently sampled image, when the image meets the set sampling adjustment condition, adjusting the first sampling strategy to a preset second sampling strategy, continuously sampling the target video data, and determining a first target image library; conducting image analysis on the images in the first target image library to obtain a first analysis result, and when the first analysis result meets the preset warning condition, giving out warning. The safety warning method and the device disclosed by the embodiment of the invention are not required to be monitored manually in real time, and are intelligent and efficient in monitoring video data processing.

Description

technical field [0001] The invention relates to the technical field of video surveillance, in particular to a safety alarm method and device. Background technique [0002] With the maturity and improvement of computer, network, information and communication, audio and video codec, streaming media and other technologies, driven by the huge demand in the security market, video surveillance technology is developing rapidly. [0003] In today's network era, most of the current video surveillance technology field obtains video data through the installation of video monitors, displays them on the monitor screen, and manually monitors the scene in the video, so as to take immediate measures to prevent the occurrence of adverse events. The situation requires 24 hours to arrange manual observation on the video surveillance display screen, which is labor-intensive and inefficient. Another situation is to arrange manual video surveillance during the day and record the surveillance vide...

AI Technical Summary

Problems solved by technology

[0003] In today's network era, most of the current video surveillance technology field obtains video data through the installation of video monitors, displays them on the monitor screen, and manually monitors the scene in the video, so as to take immediate measures to prevent the occurrence of adverse events. The situation requires 24 hours to arrange manual observation on the video surveillance display screen, which is labor-intensive and inefficient. Another situation is to arrange manual video surveillance during the day and record the surveillance video at night, so that in subsequent analysis, we can understand when an accident occurred. However, video files generally contain a large amount of information, and it takes a lot of manpower and material resources to check the monitoring records afterwards. The monitoring and alarming of adverse events in the above-mentioned prior art is not intelligent enough.

Images