Self-adaptive resolution livestock video information processing method based on artificial intelligence

Abstract

The invention provides a self-adaptive resolution livestock video information processing method based on artificial intelligence. The method comprises the following steps: projecting the original image of each subarea on a BIM ground plane after projection transformation and splicing to obtain a panoramic top view of the culture area; and processing the livestock area image through a livestock detection neural network to obtain a livestock key point thermodynamic diagram, projecting the livestock key point thermodynamic diagram on the BIM ground plane after projection transformation and splicing to obtain a livestock movement amount and a livestock key point overlay diagram after certain post-processing, and adjusting the resolution of images shot by each camera according to the livestockmovement amount. Information in the BIM is visualized through the WebGIS and an information exchange module, real-time monitoring of the livestock movement condition of the whole livestock raising area is achieved, and livestock individuals with abnormal movement amount can be further observed and analyzed.

Description

technical field [0001] This application relates to the fields of artificial intelligence and video surveillance, in particular to an artificial intelligence-based adaptive resolution livestock video information processing method. Background technique [0002] In the livestock industry, the daily exercise of livestock is often closely related to its physical condition. When the amount of exercise is low, the animal may be sick; when the amount of exercise is large, the animal may be in estrus. It is extremely important for animal husbandry workers to grasp the physical condition of livestock in time, therefore, it is essential to monitor the movement of livestock in animal husbandry. [0003] The current monitoring method for the movement of livestock in the livestock area is to set multiple cameras in the livestock area to acquire images, and then splicing the images acquired by each camera to realize the monitoring of the livestock in the entire livestock area. [0004] C...

AI Technical Summary

Problems solved by technology

However, the over-the-horizon camera it uses cannot obtain information on the movement of individual livestock in the livestock area, and there is no means for further observation of animals with abnormal movements

Images