A pedal detection method and system based on image processing

Abstract

The present invention relates to a pedal detection system and method based on image processing, wherein the method includes the following steps: an image acquisition step, collecting images of pedals in a dropped and stowed state as training samples, and collecting images of pedals at the moment to be detected as test samples; Classifier training step, extracting the direction gradient histogram information of each image in the training sample to form a feature vector; select the SVM classifier kernel function, and use the extracted training sample to train the SVM classifier; pedal detection step, extracting the image of the test sample The directional gradient histogram information constitutes a feature vector, and the SVM classifier is used to determine the state of the pedal being down or retracted. The present invention collects pedal images, extracts direction gradient histogram information, forms feature vectors, uses SVM classification method feature vectors to classify, thereby real-time monitoring of pedal retractable state, prevents collision accidents with passing trains, and improves the safety margin of the motor vehicle maintenance platform. Spend.

Description

technical field [0001] The invention relates to the field of computer image detection, in particular to a pedal detection method and system based on image processing. Background technique [0002] The EMU maintenance platform is one of the key equipment in the EMU base and the EMU operation station, which provides convenience for the daily preparation, inspection, maintenance and repair of the EMU. In the equipment process design, methods such as hydraulic crossing boards are adopted to meet the needs of maintenance personnel on different heights of various types of EMUs. Therefore, pedals (also known as crossing boards) are set on the actual platform and the protective net to ensure the safety of the staff climbing to the top. There are many CRH series of EMUs in my country, and the height and contour shape of the car bodies of different models are quite different. In order to make the maintenance platform compatible with the contours of all EMUs, it is very necessary to s...

AI Technical Summary

Problems solved by technology

Due to the hardware status and system stability of the method, there are risks of false positives and false positives in individual cases, and the switch signal cannot visually display the real-time status of the pedals.

Images