Ferrographic image multi-abrasive-particle identification method based on single-stage detection model yolov3

Abstract

The invention discloses a ferrographic image multi-abrasive-particle identification method based on a single-stage detection model yolov3. The invention provides a ferrographic image multi-abrasive-particle identification method. The method comprises the following steps: firstly, improving a backbone network of a yoov3 model; adding a spatial pyramid pooling module; replacing an IOU loss functionof an original yov3 model with a GIOU loss function, expanding a yoto layer scale is expanded, and the like. And by improving a yov3 backbone network, the problems that in the ferrographic abrasive particle recognition process, the recognition rate of similar abrasive particles is low, and the omission ratio of small abrasive particles is high are effectively solved. According to the ferrographicimage multi-abrasive-particle identification method provided by the invention, the BN layer and the convolution layer of the model are fused, so that the network structure is simplified, the additional calculation amount brought by module addition is limited, and the real-time performance of the lowest detection speed of the yolov3 model is ensured. According to the ferrographic image multi-abrasive-particle identification method provided by the invention, an intelligent identification result can be obtained only by inputting the multi-abrasive-particle picture into the trained module, more manual operations are not needed, and intelligent identification is realized.

Description

technical field [0001] The invention relates to the technical field of ferrographic analysis, in particular to a method for identifying multi-abrasive grains in a ferrographic image based on a single-stage detection model yolov3. Background technique [0002] Modern production requires mechanical equipment to have high reliability. For large-scale continuous working mechanical equipment, if a failure occurs and the equipment is shut down, it will cause major economic losses. A large number of research data show that most of the failures of mechanical equipment are caused by wear and tear. In the 1970s, the ferrography diagnostic technology was developed. By extracting the lubricating oil in the mechanical system, the wear particles (wear particles) in the oil were quantitatively and qualitatively analyzed, so as to monitor the wear state of the equipment and diagnose the fault, that is, to analyze the wear particles. The size, shape, color, particle size distribution and ab...

AI Technical Summary

Problems solved by technology

The two-stage detection is based on the candidate region recommendation to generate the region of interest for target detection, and then use CNN to classify. This type of algorithm has high detection accuracy, but due to the heavy network, the algorithm consumes a long calculation time

Images