Computer vision in artificial intelligence is dedicated to the development of automated systems that, using deep and machine learning algorithms and primarily convolutional neural networks and sophisticated software methods, can extract visual data from images and videos and analyze and interpret them in the same way as humans do. Computer vision requires huge amounts of information on which a machine learning system with the help of pre-programmed algorithms can automatically learn and repeat data analysis until it achieves a satisfactory solution in tasks such as calling objects in an image and does not begin to perceive an image like a human. The lecture and the practical part will give examples of deep convolutional neural networks with emphasis on the YOLO family that include the analysis of images taken by drone during search and rescue operations, and images taken at night with thermal Flir camera to detect the movement of people at night in the protected area.