A Machine Vision Quality Control System for Industrial Acrylic Fibre Production
- PDF / 1,203,196 Bytes
- 8 Pages / 600 x 792 pts Page_size
- 54 Downloads / 200 Views
A Machine Vision Quality Control System for Industrial Acrylic Fibre Production Paulo Heleno INETI-DOP, Estrada do Pac¸o do Lumiar, 22, 1649-038 Lisbon, Portugal Email: [email protected]
Roger Davies INETI-DOP, Estrada do Pac¸o do Lumiar, 22, 1649-038 Lisbon, Portugal Email: [email protected]
´ Bento A. Brazio Correia INETI-DOP, Estrada do Pac¸o do Lumiar, 22, 1649-038 Lisbon, Portugal Email: [email protected]
˜ Dinis Joao INETI-DOP, Estrada do Pac¸o do Lumiar, 22, 1649-038 Lisbon, Portugal Email: [email protected] Received 1 August 2001 This paper describes the implementation of INFIBRA, a machine vision system used in the quality control of acrylic fibre production. The system was developed by INETI under a contract with a leading industrial manufacturer of acrylic fibres. It monitors several parameters of the acrylic production process. This paper presents, after a brief overview of the system, a detailed description of the machine vision algorithms developed to perform the inspection tasks unique to this system. Some of the results of online operation are also presented. Keywords and phrases: quality control, machine vision, textile industry.
1.
INTRODUCTION
Quality control is becoming increasingly more important in today’s manufacturing industries. Customers demand cheap but high quality products. For an efficient and successful production, manufacturers must rely on a robust quality control system. These systems can prevent product rejection by detecting all defective products and simultaneously avoid false defect detections. Also, an automated quality control system integrated into the manufacturing process can detect production deviations early and thus avoid waste. Compared to the traditional quality control tasks performed by humans, automated quality control systems offer various advantages, including the ability to work in hazardous environments 24 hours a day, and in some tasks perform quicker measurements with higher accuracy and consistency than humans. Additionally, when the automated quality control system is computer-controlled, the results of monitoring can easily be integrated into the manufacturer’s statistical process control.
Machine vision quality control systems play a growing role in modern manufacturing quality control systems. One of the reasons explaining this growth is that quick, accurate noncontact measurement and complex feature analysis can now be performed by low-cost machine vision systems. 2.
PROBLEM DESCRIPTION
The production of acrylic fibres on the manufacturer’s production lines begins with the extrusion of a liquid polymer through spinnerets (thin metal diaphragms perforated by thousands of small holes) into a high-temperature water bath. The solidifying fibres form six parallel, flattened bundles, or tapes, each approximately of 110 mm wide and spaced approximately 15 mm apart. The tapes are drawn through the production line machinery where they are progressively stretched and dried over a series of heated metal cylinders. The width of each tape and the in
Data Loading...
