Elevator asynchronous door opening recognition system and method based on deep learning

Abstract

The invention relates to an asynchronous door opening recognition system and method based on deep learning, and the system comprises: an image correction module which is used for carrying out the illumination adaptive correction and image angle adaptive correction of an image; a model training module which comprises a sill groove target detection model training unit and an asynchronous door opening classification model training unit and is used for pre-training a model; a detection and identification module which comprises a sill groove detection unit and a non-synchronous door opening identification unit; and an alarm issuing and processing module which is used for issuing an alarm for the asynchronous door opening fault in the modes of voice, light and the like. In the scheme, the video acquisition mode acquires an elevator sill groove in real time, image illumination and angle correction are carried out, a sill groove detection model is loaded, the elevator door opening state and amplitude are detected, self-adaptive scaling processing is carried out after the landing door sill groove is detected, and an asynchronous door opening recognition model is loaded to quickly recognize asynchronous door opening faults. The asynchronous door opening fault of the elevator can be efficiently and accurately detected and recognized in real time.

Description

technical field [0001] The present invention relates to the field of elevator fault identification, in particular to a deep learning-based elevator asynchronous door opening identification system and method. Background technique [0002] In recent years, the number of elevators in my country has increased rapidly. According to relevant statistics, by the end of 2019, the total number of elevators in the country has exceeded 7.09 million, and the number of elevators in possession, annual output, and annual growth are all the first in the world. At the same time, the aging problem of elevators is becoming more and more serious. There are more than 400,000 old elevators that have been used for more than 10 years in the country, and more than 100,000 elevators that have been used for more than 15 years. The resulting huge safety problems. In 2019 alone, there were 33 elevator accidents and 29 deaths across the country. 87% of the accidents were caused by falling, shearing, and e...

AI Technical Summary

Problems solved by technology

However, there are many types of elevator door systems and the causes of failures are complex, resulting in poor stability and reliability of the fault monitoring device, and the false alarm rate of the monitoring system remains high, making it difficult to put into practical industrial applications.

Images