Railways provide the economical and most convenient mode of passenger transport for both long-distance and suburban travel. Additionally, a large portion of transportation in India relies on the railway network. However, accidents remain a major concern, particularly with railway track crossings and unidentified cracks in railway tracks. Approximately 60% of accidents occur at railway track crossings or are caused by cracks in railway tracks, leading to the loss of precious lives and economic setbacks. Therefore, there is a need to consider new technology that is robust, efficient, and reliable for both automatic gate closure systems and crack detection in railway tracks. In this research, for the automatic gate control system, two Infrared (IR) sensors are used to detect the entry and exit of the train. Using the IR data, a Servo Motor is employed to open and close the railway gate. Next, for railway crack detection, a novel Artificial Intelligence (AI) model is designed using the DenseNet, with a residual layer added at the end. The global and local features obtained from the DenseNet and residual layer are combined and passed to a classification layer for crack detection. This model shows excellent results in crack detection, leading to its deployment on the cloud. Crack images are captured by a camera and sent to the cloud along with latitude and longitude details using the Global Positioning System (GPS). If a crack is detected, the Global System for Mobile Communication (GSM) module is activated, sending an alert message with location details to the concerned person. Firebase Cloud is used for data visualization and AI model deployment. By integrating AI and Internet of Things (IoT) technology, the proposed methodology enhances the safety of the railway system and helps prevent accidents.
Automatic Railway Gate Crossing Control And Track Crack Detection System Using IoT
07.01.2025
865195 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch