Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/476153| Title: | Cap and fill level inspection for plastic bottle using vision system |
| Authors: | Leila Yazdi (P53662) |
| Supervisor: | Anton Satria Prabuwono, Prof. Dr. |
| Keywords: | Level Plastic bottle Vision system Computer vision |
| Issue Date: | 10-May-2012 |
| Description: | Automated Visual Inspection System (AVIS) has a strong ability to improve manufacturing quality control by means of inspecting products automatically instead of usual manual inspections. In other words, AVIS automatically tends to make a suitable decision in process and result’s classification according to the images of the products via image processing and Artificial Intelligence (AI) techniques. Since bottling is one of the most common packaging styles in the food, medical and other chemical product, this research focuses on the visual inspection of the bottles. The intention of this research is to develop and design a real time automated visual inspection system to check the cap and to detect fill level of plastic bottles. Whereas the previous studies have exerted these two kinds of inspections in separated systems, this study combines both of inspections in one AVIS. The objectives of this research are: first, to develop a Feature Extraction (FE) algorithm to detect cap and fill level of plastic bottle. Second, to compare and evaluate Mamdani, Sugeno and production rule fuzzy logic classification methods in order to realize the appropriate one related to this subject. Finally to develop a real time AVIS for detection of cap and fill level while bottles move on a conveyor belt. A webcam and two LEDs in a dark background with a suitable conveyor speed are applied to capture images of plastic bottles. The edges and the average of distances between edges are two significant parameters, obtained by FE algorithm. This algorithm is capable to be used for both cap and fill level detection in the same time. The results demonstrate that 96.25%, 95.41%, 94.58% accuracy for Mamdani, Sugeno and production rule classifier respectively.,Master/Sarjana |
| Pages: | 95 |
| Call Number: | TA1637.Y349 2012 3 |
| Publisher: | UKM, Bangi |
| Appears in Collections: | Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| ukmvital_74627+Source01+Source010.PDF Restricted Access | 2.43 MB | Adobe PDF |  View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.