Computer Vision-based Smart Railway Track Crack Detecting System

The Pakistan Railways (PR) network is comprised of 7,791 route kilometers, 7,346km of broad gauge, and 445 km of meter gauge. There are almost 633 stations in the network all across Pakistan. Pakistan Railways carries 65 million passengers annually and operates 228 mail, exp

Project Title

Computer Vision-based Smart Railway Track Crack Detecting System

Project Area of Specialization

Electrical/Electronic Engineering

Project Summary

The Pakistan Railways (PR) network is comprised of 7,791 route kilometers, 7,346km of broad gauge, and 445 km of meter gauge. There are almost 633 stations in the network all across Pakistan. Pakistan Railways carries 65 million passengers annually and operates 228 mail, express, and passenger trains daily. Over the last few decades, Pakistan Railways has not been on track because of the multi-faceted challenges of inadequate human capital, and poor infrastructure.

Cracks in rails were diagnosed to be the principal purpose of derailments in the past, but there were no reasonably-priced automatic answers to be had for trying out purposes. Although, the railway has an upkeep branch which consumes a whole lot of manpower and money.

The proposed project utilizes current innovation by using Arduino as well as image processing.

The fault data can be remotely transmitted to the railway safety management center using a GSM module. It incorporates imperfection level and area data obtained by the embedded GPS receiver and by using image processing. We monitor the track health condition more frequently using an automated crack detection system. The proposed framework focuses on implementing a MATLAB model to detect railway track cracks through image processing techniques. Gabor filter is used to analyze the depth of the cracks. By using Arduino, the system utilizes a global positioning system (GPS) system along with a global system for mobile communication (GSM) module, Arduino nano, and infra-red (IR) sensors for transmitting and receiving messages during a fault condition. This work proposes a practical answer for the issue of cracks on the tracks of railway identification using an IR sensor that utilizes its transmitter and receiver for tracing the faulty section of the track which at that point retouched quickly with the goal that numerous lives will be spared.

Project Objectives

The major goal of the task is to layout the railway crack detection using IR, and Ultrasonic sensors. The ultrasonic sensor is used to inspect the crack inside the rail line by measuring the gap from track to sensor. The ultrasonic approach is the best approach which inspecting cracks on a railway track. The targets of this task are to lessen existence losses, excessive damages, and manpower. It additionally to enhance our railway system.

Project Implementation Method

The proposed project utilizes current innovation by using Arduino as well as image processing. The fault data can be remotely transmitted to the railway safety management center using a GSM module. It incorporates imperfection level and area data obtained by the embedded GPS receiver and by using image processing. We monitor the track health condition more frequently using an automated crack detection system. The proposed framework focuses on implementing a MATLAB model to detect railway track cracks through image processing techniques. Gabor filter is used to analyze the depth of the cracks. By using Arduino, the system utilizes a global positioning system (GPS) system along with a global system for mobile communication (GSM) module, Arduino nano, and infra-red (IR) sensors for transmitting and receiving messages during a fault condition. This work proposes a practical answer for the issue of cracks on the tracks of railway identification using an IR sensor that utilizes its transmitter and receiver for tracing the faulty section of the track which at that point retouched quickly with the goal that numerous lives will be spared. As an added advantage we also focus to create a separate model which will predict the internal cracks through thermal image processing techniques. Thus, the efficient design and implementation of railway track crack detection are being evaluated and simulated using MATLAB. The ultrasonic sensors are fixed at the vehicle, and once detected the message is automatically sent to the control room. Then an alert is given to the driver and an emergency brake control is applied.

Benefits of the Project

Reduces the need for manpower. Provides real-time monitoring of and detection of any crack or objects on railway tracks and in case of detection sends information with the location to the receiver side using a GPS module via GSM.

Technical Details of Final Deliverable

In our project, there are set of ultraviolet sensors gadgets suited for the front aspect of the automobile. This activates or deactivates the GSM transmitter unit whilst it encounters any cracks inside the tune. The sensors withinside the circuit are used to experience the cracks which arise whilst the programmed fee changes. When the automobile is Powered On, it actions alongside the version tune. The sensors constantly display the response of the tracks. In a normal state the motor, transmission is in the initial stage. When the battery strength components of the Arduino then it begins off evolved the motor in the forwarding route and serial transmission is used to transmit the messages to the microcontroller. When the crack is detected with the aid of using the UV sensors the automobile stops without delay and the GSM module transmits the textual content message to the predefined range with the assistance of a SIM card this is injected into the module and the location of the crack is dispatched at the side of the message comprising the GPS module, therefore, this allows the operator find the location of crack without problems on the time of maintenance. We also use an alternate method by using the computer vision-based method proposed for analyzing the rail of railway systems. There is image processing based on several techniques in the literature. The rails are determined by means of image processing techniques using the proposed method. This project also concentrates on the use of ultrasonic sensors which are used to detect humans or any object being pursued on the track. If any crack or objects occur in the track means longitude and latitude of the place are messaged to the nearest station GPS module. The recording and sending of coordinates are done by the GPS module. The activities of various devices are controlled by a Raspberry Pi 3.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

IT

Other Industries

Transportation , Security , Telecommunication

Core Technology

Internet of Things (IoT)

Other Technologies

Artificial Intelligence(AI), Big Data

Sustainable Development Goals

Good Health and Well-Being for People, Industry, Innovation and Infrastructure, Sustainable Cities and Communities

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com