Automated Industrial Inspection solution for plastic bottle defects
Industrial automation is the major revolution all across the world. Many of the manual processes in manufacturing products has now become automated and has increased its growth rate. But in our country many industries are still using conventional methods for manufacturing product where manual inspec
2025-06-28 16:25:14 - Adil Khan
Automated Industrial Inspection solution for plastic bottle defects
Project Area of Specialization Artificial IntelligenceProject SummaryIndustrial automation is the major revolution all across the world. Many of the manual processes in manufacturing products has now become automated and has increased its growth rate. But in our country many industries are still using conventional methods for manufacturing product where manual inspection is the biggest area which needs to be automated as they are consuming a great deal of time. Due to manual inspection, there exists a great probability of human error which results in shipment of faulty products into the market and descend the customer’s satisfaction. Our proposed project will automate the inspection process by using computer vision and AI based algorithms which can increase the inspection capability without human integrity at a low cost budget.
The proposed project will target the bottle inspection of hair care product from one of the leading consumer product industry of Pakistan. The inspection process will include the following defect detection.
- Dent on bottle
- Skew Label
- Cap Distortion
- Batch Number reading
The main objective of this project is to develop prototype solution for the industry which will reduce the inspection time and give results with high accuracy rate.
- To develop a cost efficient system/application solution as compared to importing expensive machineries that would only detect singular defect.
- To optimized the inspection process by the proposed automated inspection solution to all the industrial sector across the Pakistan which eventually have great impact on nation’s economy.
- Data acquisition: Data Acquisition leads to the phase of Generating or gathering information by detecting defects from the sample bottles.
- ROI extraction: Region of Interest (ROI) extraction indicates a process where we find the specific area where all the detecting algorithms will get implemented.
- Pre –Processing: The aim of pre-processing is an improvement of the image data that suppresses undesired distortions or enhances some image features relevant for further processing and analysis task
- Apply detection Algorithms: we would apply some efficient detection algorithms to identify defects and flaws in the manufactured bottles. Here are some areas where detection algorithms will be used.
- Dent on bottle: An Algorithm will be applied while scanning the whole bottle through cam/sensors to detect the scratches and dents on the bottle and also analyze the density of dents so to classify them as acceptable or unacceptable samples.
- Skew Label: An algorithm will be developed to identify the label’s alignment and to check whether the brand label is printed properly or not.
- Cap Distortion: An algorithm will be developed to pick out the bottles with unsealed caps or the bottles with loose caps, Deformed caps or the caps with undesired designed will be detected.
- Batch Number reading: An algorithm will be made to specifically read and analyze the batch number so that if a bottle found defected, the whole batch would be checked again to ensure the quality, the algorithm also detect flaw in the format of the batch number.
- Result: The All four detection algorithms will work concurrently and efficiently. A report of sufficient appropriate data regarding the scratches & Dents will be generated through which we analyze the flaws in manufacturing process and make it better.
Our proposed project will automate the inspection process by using computer vision and AI based algorithms which can increase the inspection capability without human integrity at a low cost budget.
Technical Details of Final Deliverable- Acquired Images will be processed on software Environment to detect the multiple defects by using Ai and classical methods of image processing.
- The proposed solution will be administered through interactive GUI/Smart app for reporting.
- The Complete Solution would help to elevate the level of quality assurance throughout the manufacturing sector at a much cheaper cost compared to importing expensive machineries that would only detect singular defect.
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 69000 | |||
| Chamber | Equipment | 1 | 15000 | 15000 |
| Camera | Equipment | 1 | 30000 | 30000 |
| motor | Equipment | 1 | 4000 | 4000 |
| controller | Equipment | 1 | 5000 | 5000 |
| illuminator | Equipment | 1 | 5000 | 5000 |
| printing | Miscellaneous | 1 | 5000 | 5000 |
| extra circuitry | Miscellaneous | 1 | 5000 | 5000 |