Building Smart Surveillance Solutions for Livestock Farms

Abstract

Farming is an important industry in the world. With the development of artificial intelligence, the intelligence of agriculture has become a trend. Intelligent monitoring of agricultural activities is an important part of it. However, due to difficulties in achieving a balance between quality and cost, the goal of improving the economic benefits of agricultural activities has not reached the expected level. Farm supervision requires intensive human effort and may not produce strong results. In order to achieve intelligent monitoring of agricultural activities to improve economic benefits, this paper combines UAV with Deep learning model, through the combination of UAV in the agricultural industry to detect and classify objects, to achieve independent agriculture without human intervention. In this paper, a highly reliable target detection and tracking system using UAV is developed, which will be proved to be cost-effective. The system uses Deep learning method to solve the target problem. The model uses the data collected from DJI Mirage 4 UAV to detect, track and classify different types of targets. The average map accuracy of this method for target detection and tracking is 90.98\%, the average accuracy is 89.76\%, and the average recall rate is 88.78\%. In addition, this paper also compares the performance of different Deep learning models, such as YOLOv7, FASTER-RCNN, SSD, MASK-RCNN, and compares them with key evaluation indicators, and finally determines a method through the development of YOLOv7-DeepSort model.