Image monitoring method based on computer vision

Abstract

The invention discloses an image monitoring method based on computer vision. The image monitoring method comprises the following steps: acquiring image data; processing the image data by using a processor so as to as least obtain an image feature of an image in a monitored region; performing three-dimensional reconstruction based on the image feature by using the processor, thereby acquiring three-dimensional motion information of an object in the monitored region; and judging whether the three-dimensional motion information satisfies a preset condition by using the processor; if the three-dimensional motion information satisfies the preset condition, sending out a prompt signal or a control signal. Through the image monitoring method based on the computer vision provided by the invention,the cooperative operation values of various equipment resources are deeply mined and optimized, and the method is especially suitable for developing an efficient and energy-saving intelligent video monitoring software platform and hardware product suitable for safety production of the coal mine, the automatic speed-regulation of the coal mine conveyor belt, the automatic identification of the coal transportation and sale behavior, and the advance pre-control on the intrusion of the dangerous region personnel and the abnormal detection can be automatically identified, the mine production efficiency and the safety protection level are improved, and the energy-saving and consumption reduction can be realized.

Description

technical field [0001] The invention relates to a video monitoring method, in particular to an image monitoring method based on computer vision. Background technique [0002] With the continuous development of science and technology, the application of industrial video surveillance systems is becoming more and more common, and its application mode has developed to a complex system consisting of multiple cameras, multiple monitors and a central control room with superior configuration performance. Moreover, general industrial video surveillance has the characteristics of wide frequency bandwidth, low loss, and anti-electromagnetic interference. Therefore, the surveillance cameras are distributed to production units or business departments, and then effectively monitored by the central controller, which can be reflected in a concentrated, real-time, and vivid manner. Detected objects, even if a large amount of rich information is obtained. [0003] Especially for the coal ind...

AI Technical Summary

Problems solved by technology

It can be said that there is still a lack of a unified theory for image surveillance methods based on computer vision, and for targeted image surveillance methods, even if they are applicable in some scenarios, they may not be applicable to other scenarios.

Images