Steel hoisting ropes play an important role in metallurgical equipment, ensuring reliability and efficiency of lifting operations. One of the key features of their operation is the high level of contamination typical of metallurgical operations. Metallurgical processes are often accompanied by dust, metal chips and other abrasive particles that can significantly degrade ropes, causing wear and corrosion. To maintain the efficient operation of equipment it is necessary to monitor the condition of hoisting ropes in real time, which makes the task of improving automatic systems for monitoring the condition of ropes urgent. The paper reviews the methods of optical control of defects in hoisting steel ropes, the advantages and limitations of different approaches are considered. The aim of the work is to justify the effectiveness of the authors' developed method of analyzing rope defect images using neural networks in relation to the method based on the discrete Fourier transform. It is revealed that one of the most promising in terms of technical and economic efficiency of inspection methods is the application of vision system with image processing based on convolutional neural network technology, which allows to effectively detect defects in complex and changing operating conditions, such as metallurgical and mining production, where the background of the image may be non-uniform, and the distance between the camera and the rope varies.

Keywords: lifting ropes, vision systems, optical control methods, fast Fourier transform, hidden Markov models, convolutional neural networks