Article,
A Model for Detecting Movement in Railway Infrastructure Using Deep Convolutional Neural Network
CENTRAL ASIAN JOURNAL OF THEORETICAL & APPLIED SCIENCES, (2022)
Abstract
The continuous monitoring of movements using artificial intelligence technique has become a necessary system to cob the level crossing problems faced around the railway infrastructure. This project work has developed a model for detecting movements using image processing method with convolutional neural network, an artificial intelligence technique in monitoring and detection of movements in the railway infrastructure. Images containing movements and non-movements in railway infrastructure were acquired from Google image search have been used in the implementation of the system and a deep convolutional neural network model has been developed. The proposed model was able to classify 11 out of 12 images correctly in the evaluation of the train dataset and the performance accuracy of the model recorded was 92% with a loss of 0.4% which resulted to the misclassification of one of the images. The analysis of the system showed that the system can be optimized to perform better.
