Machine vision anti-pinch system based on deep learning

Abstract

The invention provides a machine vision anti-pinch system based on deep learning, and belongs to the technical field of intelligent control. A convolutional neural network CNN is used, a learning table closing process is built into a convolutional neural network model, and the recognition accuracy is improved. The system makes full use of the characteristics, analyzes the occurrence of obstacles at the closing position of the learning table according to the obtained video images, and correctly controls the automatic closing action to achieve the anti-pinch purpose, thereby effectively overcoming the defects of a traditional anti-pinch mode. The method has the advantages of high precision, no contact and the like.

Description

technical field [0001] The invention belongs to the technical field of intelligent control, and relates to a machine vision anti-trap system based on deep learning. Background technique [0002] At present, there is a screen door between the waiting area of ​​the subway platform and the track area to ensure the safety of passengers while waiting. However, there is a certain gap between the train and the screen door. When the train is running, situations such as passengers being trapped, items left behind, bags, and clothes are prone to occur in this gap, and there are relatively large safety hazards and accident risks. For the safety detection between subway trains and screen doors, the existing method is mainly manual observation, but this method cannot accurately detect too long straight and curved platforms. The automatic detection method mainly uses laser beam detection and infrared light curtain detection, but laser beam detection and infrared light curtain detection b...

AI Technical Summary

Problems solved by technology

[0006] 2. Has a huge detection blind spot,

[0007] 3. Poor security

Images