Vision Based High Speed Sorting System
Automation and technology set the foundations of today's world because things are going towards intelligent and automated systems in the Fourth Industrial Revolution. Automation increases productivity and efficiency which humans lack when it comes to performing tasks for a longer duration.
2025-06-28 16:29:58 - Adil Khan
Vision Based High Speed Sorting System
Project Area of Specialization Mechatronics EngineeringProject SummaryAutomation and technology set the foundations of today's world because things are going towards intelligent and automated systems in the Fourth Industrial Revolution. Automation increases productivity and efficiency which humans lack when it comes to performing tasks for a longer duration. Industrial Automation is a major reason behind the success of various 1st world countries.
The Vision-based high-speed sorting system will make use of computer vision techniques a vision camera which will be attached with a microprocessor raspberry pi. The camera will detect the incoming carton box on the conveyor on basis of the pattern printed over it. The system will decide on the sorting direction under the signal of the raspberry pi. Mostly, Pneumatic sorting is used due to its low cost which is a major concern of any industry. Boxes after getting detected by the camera will be sorted by a Pneumatic sorting system which consists of a Cylinder, valves, and an end effector. The pneumatic sorting mechanism will be controlled by a programmable logic controller (PLC) which are rugged controller made for such applications. Here is a pictorial view of the project completed till now:
The objective is to develop a Vision-based high-speed sorting system indigenously. In large industries, various types of sorting systems are being used for segregation purposes. However, these systems have various drawbacks such as high initial cost, expensive maintenance, and non-availability of spare parts as they must be imported.
Therefore, we aimed to make a cost-effective sorting system that operates at nominal speed and to carry out its operations, and whose operating speed and capability can be improved in the future.
Project Implementation MethodThe methodology has been sub-divided into two sections.
- Detection Methodology
- Sorting Methodology
Detection Methodology
The first and most important part is pattern detection over the box. The system makes use of ARUCO Markers as a pattern to be detected. An Aruco marker is a synthetic square marker composed of a wide black border and an inner binary matrix that determines its identifier (id). These can conceptually be thought of as similar to a QR code — a 2D binary pattern that can be detected using computer vision algorithms. However, an Aruco marker only holds 4-12 bits of data, multiple orders of magnitude less than a QR code (a typical QR code can hold up to 3KB of data).
The camera will detect this pattern on the box when boxes pass in front of a camera and Raspberry Pi generates the signal and this signal is received at the PLC end. The signal contains the information about the direction of sorting.
Sorting Methodology
The sorting function is done by a pneumatic cylinder. The actuation of the cylinder to push the carton box or not is based on the signal received by the PLC. If a signal gives information to push the box, the PLC will send a signal to the solenoid valve to extend the cylinder when the box is in front of the cylinder. The decision that the box is in front of the cylinder is done by a proximity probe placed near the cylinder. The cylinder after sorting the box will come to its retracted position immediately. To summarise, The Raspberry Pi gives the signal to PLC to sort the box or not and the proximity probe tells the PLC when to sort the box.
This project is related to Industrial Automation, so it is important to discuss the impacts of Automation on society, the health of people, the safety of workers, and legal constraints. Major Advantages of Automation are higher production rates, more efficiency, reduced material waste, better quality, high safety, and reduced factory lead time.
EconomyThis project has numerous benefits associated with it. It will help the industry in doing the sorting process efficiently and effectively. It will reduce human labor and reduce the manufacturing lead time of the industry. Hence, it will help the industries deliver orders fast. This will help the industry grow.
Societal ImpactThe social impacts of Automation are numerous and different. In no way, Automation is consistently great or inadmissibly terrible for society. This project is not going to affect society negatively, it leaves no pollution. There is pneumatic actuation being performed, which may leave air as waste material.
HealthThis project is reducing worker interaction in the segregation process of any industry. It will avoid back pain in workers which usually arises due to picking and placing carton boxes manually on repeat.
SafetyThis product is meeting safety standards in Industry. It has proper vicinity to work in, over a conveyor belt where pneumatic actuation is being performed. It will not harm anyone until someone knowingly enters or interrupts the process.
Legal Constraints
There are no legal constraints related to the product. All equipment is easily available in the market. No explosives or patent product is being used.
As it is already explained that this sorting system is vision-based and it is based upon the following mechanisms. The technical detail of the components of those mechanisms is listed below;
1) Detection Mechanism:
- Vision Camera (FFY-U3-16S2M)
This is a monochromatic camera of 60 frames per second and can detect patterns printed on the carton boxes translating at the speed of 0.7 m/s efficiently.
- Raspberry Pi 4B
This microcontroller is being used as a liaison between the camera and the PLC. It is using Ubuntu (Linux) as its operating system. The vision camera is interfaced with Raspberry Pi and an algorithm is coded to grab the detection details of the camera to generate a signal at its output port.
2) Sorting Mechanism:
- Mitsubishi PLC
The PLC is being used to control the working of a pneumatic cylinder for sorting purposes based on a 5V signal coming from the Raspberry Pi. Relays are used as switches for generating either an ON or OFF signal. PLC programming was done according to the requirement of the project using a hand-held controller FX-10P.
- 5/2 Double Solenoid Valve
The valve is controlling the extraction and retraction of the cylinder based on the signal coming from the PLC. 24 V signal is provided by the PLC to the solenoid valve and it provides the pressure of 3-4 bar to the pneumatic cylinder.
- Pneumatic Cylinder (Double Acting)
The cylinder takes input from the solenoid valve and it extracts and retracts accordingly. The cylinder can sort carton boxes of 5 to 6 kilograms.
Final Deliverable of the Project HW/SW integrated systemCore Industry OthersOther IndustriesCore Technology OthersOther TechnologiesSustainable Development Goals Good Health and Well-Being for People, Decent Work and Economic Growth, Industry, Innovation and InfrastructureRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 78000 | |||
| Vision Camera (FFY-U3-16S2M) | Equipment | 1 | 35000 | 35000 |
| Raspberry Pi 4B | Equipment | 1 | 18000 | 18000 |
| Sorting Mechanism | Equipment | 1 | 5000 | 5000 |
| Conveyor Motor Assembly | Equipment | 1 | 10000 | 10000 |
| Miscellaneous Expenses | Miscellaneous | 1 | 10000 | 10000 |