Crack Detection Drone
Modern Structural Health Monitoring requires techniques that revolutionize the process without involving the risk of human life or adding any extra labor costs. Maintenance of structures demands great input in terms of workforce and costs. To eliminate this requirement, the industry requires an inno
2025-06-28 16:30:59 - Adil Khan
Crack Detection Drone
Project Area of Specialization Mechatronics EngineeringProject SummaryModern Structural Health Monitoring requires techniques that revolutionize the process without involving the risk of human life or adding any extra labor costs. Maintenance of structures demands great input in terms of workforce and costs. To eliminate this requirement, the industry requires an innovative, modern and automated technique to make maintenance and health monitoring of structures, economically viable and result-wise, accurate disregarding any fear of human error. A group of three students take upon the challenge of developing a new technique that involves the use of an Unmanned Aerial Vehicle (UAV) Drone prototype, coupled with Deep Neural Networks (DNN) (to detect cracks in structures) and Image Processing (To enhance cracks and obtain crack parameters). A drone is assembled carrying a high definition (HD) camera connected to a Raspberry Pi that uses deep learning to detect cracks and take pictures of structures that are damaged. The picture is then processed to obtain crack parameters (width, length, area). Not only this technique guarantees safer and easier scheduled maintenance, its running cost is quite low too. This technique involves no human intervention (apart from operating the drone) so accuracy in identifying cracks is far greater than the human eye. Not only this, a drone will be able to access the areas that are inaccessible by humans for example, bridges, overhead trusses and structures etc. All three students working on the project are majoring in Mechanical Engineering, so this was quite a task for them to get in the field of neural networks and deep learning to gain knowledge and be able to understand this field (that is alien to them) and perfectly implement it into their project.
Project ObjectivesTo develop a new modern, automatic and time-saving method for Structural Health Monitoring (SHM). To solve the problem, a drone will be assembled that will monitor and take images of structures with cracks by identifying them. A program will be developed that will identify and obtain crack parameters.
Project Implementation MethodAssemble a drone on the F450 Drone frame with the techincal specifications according to our need and budget i.e. payload, range etc. To identify cracks, we use Deep Neural Network with a dataset of sample crack pictures to help identify cracks. A MATLAB code will run then if a crack is detected to identify the crack parameters. All this will be possible onboard the flight through a Raspberry Pi onboard controller with real time processing. The data gathered will then be transmitted to a work station.
Benefits of the ProjectIndustry use for Civil structures, Structural Health Monitoring and various structure maintenance procedures. Reduce human life risk involved in maintenance procedures and make scheduled maintenance easier. Reduce maintenance costs in the long run.
Technical Details of Final DeliverableA drone capable of detecting cracked surface with the help of 5mp camera and a Raspberry pi 4. After doing so the program will run on matlab in order to determine the parameters of the crack with the helpmof image processing. The drone is also equipped with a pixhawk powered by a 2200 mAh lipo battery. The pixhawk is controlled by a 6 channel Radio Controller.
Final Deliverable of the Project Hardware SystemCore Industry OthersOther IndustriesCore Technology NeuroTechOther TechnologiesSustainable Development Goals Good Health and Well-Being for People, Industry, Innovation and InfrastructureRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 51255 | |||
| F450 | Equipment | 1 | 1450 | 1450 |
| Pix Hawk | Equipment | 1 | 8000 | 8000 |
| Telemetry | Equipment | 1 | 4000 | 4000 |
| Radio Controller | Equipment | 1 | 9500 | 9500 |
| Propeller | Equipment | 4 | 200 | 800 |
| Esc | Equipment | 4 | 1600 | 6400 |
| Power Distribution board | Equipment | 1 | 300 | 300 |
| Safety Switch | Equipment | 1 | 1200 | 1200 |
| Buzzer | Equipment | 1 | 200 | 200 |
| Power Module | Equipment | 1 | 2500 | 2500 |
| Motors | Equipment | 4 | 450 | 1800 |
| Vibration Damper | Equipment | 1 | 1200 | 1200 |
| Lipo Battery 2200 mAh | Equipment | 1 | 2500 | 2500 |
| Raspberry pi 4 | Equipment | 1 | 9000 | 9000 |
| Pi Camera | Equipment | 1 | 2000 | 2000 |
| Landing gears | Equipment | 1 | 405 | 405 |