Computer Vision based Parts Inspection Machine
The project builds on the Computer Vision field and uses a combination of hardware and software to build a system of Parts Inspection machine that can inspect a part like a drill bit and determine, with accuracy and precision, the dimensions of the part. The project includes a hardware platform with
2025-06-28 16:30:53 - Adil Khan
Computer Vision based Parts Inspection Machine
Project Area of Specialization Internet of ThingsProject SummaryThe project builds on the Computer Vision field and uses a combination of hardware and software to build a system of Parts Inspection machine that can inspect a part like a drill bit and determine, with accuracy and precision, the dimensions of the part. The project includes a hardware platform with three movable cameras and a movable platform and a software code written in Python programming language to achieve this task.
Project ObjectivesThe objectives of this project are:
- Create and build a computer visions based inspection system and a measurement system for specific mechanical parts.
- To measure the parts with a stable non-varying level of accuracy.
- To make it accurate and efficient.
- Reduce manual labor and make the process automated.
A code will be designed that will take images of a mechanical part like a drill bit, and binarize the images. Next, it will take the image and remove noise and the background to get a clear image. The image will then be compared with the actual size of the drill bit (measured through instruments like the vernier caliper) and then will be scaled to the right amount. This way when the next part will be placed beneath the camera, it will binarize and quantify the data by multiplying it with the same base scaling factor to get the accurate pixel to mm measurement.
The system design includes hardware and software. Three cameras are used to capture various images that expose different dimensions of the mechanical part. Backlighting is used to illuminate the part placed on the V-shaped translucent surface,
Benefits of the ProjectThis will be a low-cost alternative to high-end methods used in most big industries. High-end industries with high precision requirements employ sophisticated methods for inspection, such as coordinate measuring machines (CMM) or X-ray computed tomography (CT). However, there are many small or medium-sized industries that have lower precision requirements for which the use of high-end inspection equipment is not feasible. Small or medium-sized industries rely on less expensive methods for their mechanical parts inspection. The simplest method is to use Go and No-Go gauges, however, it typically requires a human operator who only checks a single dimension per gauge, and thus it is inefficient and prone to human error. This will provide a sense of automation with a decrease in costs.
Technical Details of Final DeliverableThe final deliverables will be in the form of hardware and software. The software will be Python-based that will make use of the Computer Vision Library while the hardware will be composed of an acrylic base with a backlight illuminating it and three CMOS cameras, one on the top, one at the front and the last on the back. All of these will be configurable in their length and their distance would be variable to cater to different sizes of drill bits.
Final Deliverable of the Project HW/SW integrated systemCore Industry ManufacturingOther IndustriesCore Technology Internet of Things (IoT)Other TechnologiesSustainable Development Goals Industry, Innovation and InfrastructureRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 72500 | |||
| Steel Supports | Equipment | 2 | 300 | 600 |
| LED Lights | Equipment | 1 | 200 | 200 |
| Screws | Equipment | 40 | 10 | 400 |
| Wood Panel | Equipment | 1 | 1000 | 1000 |
| Acrylic Sheet | Equipment | 1 | 1300 | 1300 |
| CMOS Cameras | Equipment | 3 | 22000 | 66000 |
| Miscellaneous | Miscellaneous | 1 | 3000 | 3000 |