Detection of Objects Using the Yolo Neural Network Algorithm

The use of human labor in the clearance of mined areas in the post-war period can lead to great loss of life. Therefore, before the start of restoration work, it is advisable to use unmanned aerial vehicles to detect, clear land and hidden mines. The article proposes a model of a ground object recognition system based on a convolutional neural network with the YOLO-V5 architecture, which can be used as part of an artificial intelligence system of an unmanned aerial vehicle. The main purpose of our research is to study the terrain using an unmanned aerial vehicle (UAV), identification, recognition of existing objects, obtaining the exact coordinates of the object, calculating the area of the detected object and detecting a minefield. Today, the task of computer vision is not only to recognize images, but also to determine their location. In the article, YOLO was chosen as the detection algorithm. This algorithm is a modern real-time object detection system. The article deals with the issue of recognizing and determining the coordinates of mines using computer vision and the joint use of the YOLO neural network.