General bridge crane safety bottom event recognition method based on machine vision

Abstract

The application discloses a general bridge crane safety bottom event identification method based on machine vision and relates to the technical field of safety detection of hoisting machinery. The machine vision collection system is used to collect images of the metal structure of the crane in full coverage, and image preprocessing and a convolutional neural network classifier are used to automatically identify and quantify visual features such as cracks, deformation and welding defects. These features are mapped into specific bottom events in an accident tree analysis model. The size, quantity, position and development trend of specific defects are calculated through image analysis, and these quantitative indexes are converted into the occurrence probability of bottom events and input into the accident tree model, so that the accident tree model can reflect the real health condition of the structure at present, and the minimum cut set and the top event probability are calculated in real time to dynamically reveal the current failure path, thereby providing accurate target orientation for preventive maintenance and changing safety supervision from post-response to pre-warning.

Images