A QUALITY CONTROL SYSTEM PROTOTYPE FOR DETECTING KNOT DEFECTS IN THE WOODEN PANEL MANUFACTURING

Mugla Journal of Science and Technology A QUALITY CONTROL SYSTEM PROTOTYPE FOR DETECTING KNOT DEFECTS IN THE WOODEN PANEL MANUFACTURING Özgür KILIÇ, Department of Computer Engineering, Muğla Sıtkı Koçman University, Turkey, ( https://orcid.org/0000-0002-0970-2071) Delikanlı Mertcan SUSUZ, Department of Computer Engineering, Muğla Sıtkı Koçman University, Turkey, ( https://orcid.org/0000-0003-0279-9314) Barış Ethem SÜZEK*, Department of Computer Engineering, Muğla Sıtkı Koçman University, Turkey, ( Received: 22.10.2018, Accepted: 24.01.2019 *Corresponding author https://orcid.org/0000-0002-1521-4306) Research Article DOI: 10.22531/muglajsci.473338 Abstract Product quality has become a necessary goal for all manufacturers in today’s competitive market. Product defects, not detected, cause financial damages and reputation loss for the manufacturer. These defects can be due to quality of the inputs or misuse of the good quality inputs during the manufacturing process. This is also the case for wooden panel manufacturing where elements are the basic input. It is possible to reduce the loss of the manufacturer by using a method that minimizes the human error in the inspection of the elements. In this study, we, first, identified the quality control problems of the wooden panel manufacturers and basic steps in the automated element quality control. We then developed a prototype for the detection of knots, the most common defects in wooden panels. This prototype, with 80.0% true positive (with knot defect) and 82.0% true negative (without knot defect) rates, performs close to accuracy rates of a quality control inspector. The element image library created during the development of the system made publicly available for use in similar studies. This prototype is expected to be developed to detect other wood defects and to be applied in the wooden panel manufacturing. Keywords: Image processing, Defect detection, Wooden panel quality control, Knot defect AHŞAP PANEL ÜRETİMİNDE BUDAK KUSUR TESPİTİ İÇİN BİR PROTOTİP OTOMATİK ELEMENT KALİTE KONTROL SİSTEMİ Özet Ürün kalitesi yakalamak günümüzün rekabetçi pazarında üreticiler için olmazsa olmaz bir hedef olmuştur. Ürün kusurları, farkedilmezse, üreticinin ekonomik zararına ve itibar kaybına neden olur. Bu kusurlar ürünün girdilerinin kalitesinden veya kaliteli girdilerin uygun bir biçimde kullanılmamasından kaynaklanabilmektedir. Ana girdisi element olan ahşap panel üretimi için de bu durum söz konusudur. Elementlerin kontrol işlemi sırasında insan hatasını en aza indirecek bir yöntem ile üreticinin kaybını azaltmak mümkündür. Bu çalışmada ahşap panel üreticilerinin karşı karşıya olduğu kalite problemlerini ve üretiminde otomatik element kalite kontrolündeki temel adımları belirledik. Daha sonra, ahşap panellerde en yaygın kusur olan budakların tespiti için bir prototip geliştirdik. Bu prototip, %80,0 doğru pozitif(budak hatalı) ve %82,0 doğru negatif(budak hatasız) tespit oranları ile bir kalite kontrol denetçisine yakın doğruluk oranında çalıştı. Sistemin geliştirilmesi sırasında oluşturulan element görüntü kütüphanesi benzer çalışmalarda kullanılabilmesi için kamuyla paylaşıldı. Bu prototipin, diğer ahşap kusurlarını da tespit edecek şekilde geliştirilmesi ve ahşap panel üretiminde uygulanması öngörülmektedir. Anahtar Kelimeler: Görüntü işleme, Kusur tespiti, Ahşap panel kalite kontrolü, Budak kusuru Cite Kilic, O., Susuz, D.M., Suzek, B.E., (2019). “A Quality Control System Prototype for Detecting Knot Defects in the Wooden Panel Manufacturing”, Mugla Journal of Science and Technology, 5(1), 24-33. panels gained popularity among domestic and international consumers. For instance, wooden market production in Turkey has spiked up to an annual production capacity of 40.000 m³, a notable increase in its market share in an effort to meet the demand in the country [2]. 1. Introduction Wooden panel is defined as a product made from small sized wooden blocks, namely, elements which are first glued with each other on the narrow edges to form longer pieces of wood, which are then glued on the wide edges [1]. Due to their striking superiority in comparison to other equivalent products, wooden 24 Özgür Kılıç, Delikanlı Mertcan Susuz, Barış Ethem Süzek A Quality Control System Prototype for Detecting Knot Defects in the Wooden Panel Manufacturing Sizes of the elements used in wooden panels generally range from 5x5x25cm to 5x5x90cm and in each wooden panel, elements obtained from the same tree types are used. If the defects on the elements are overlooked during the manufacturing process then the final product, which is the wooden panel, will have poor quality and this will adversely affect the marketability of the wooden panel. Therefore, elements having defects should be eliminated through a rigorous quality control process. Existing element quality control in wooden panel manufacturing plants is performed manually by human inspectors. Therefore, they are prone to errors due to environmental and personal factors. Furthermore, the whole production process is prone to delays due to potential absence of quality control inspectors due to personal or health reasons. Although training new quality control inspectors is an option, this takes time and investment. Image processing techniques have been used in various fields to perform automated quality control without human intervention. These techniques are in use and becoming popular in the quality control of products such as textile [3-6], ceramic [7-9], steel [10,11], glass [12], electronic circuits[13], weld bead[14] , metal lids [15] and cylindrical parts [16] to keep up with high quality standards set by the competitive markets. There are also academic studies [17-22] on applying image processing techniques in the quality control of wooden surfaces. There are existing quality control systems in the wider wood industry. EasyScan [23] by Weinig for lumber quality control; SuperScan [24] by Grecon and ColourBrain [25] by Baumer for quality control of laminate and furniture panels, are such quality control systems. There is no automated quality control system specific to elements used in wooden panel manufacturing. One option is the adaptation of these quality control systems to element quality control purpose. However the investment to do so is forbiddingly high for the wooden panel manufacturers in Turkey, given their size. The other option is a custom element quality control system tailored towards this particular need of wooden panel industry. In this study, we, first, identified the quality control problems of the wooden panel manufacturers and basic steps in the automated element quality control. We then developed a prototype for the detection of knots, the most common defects in wooden panels. We envision this prototype will be extended to cover other type of wooden panel defects. Hence, this study aims to aid designing automated element quality control systems and improve the e (...truncated)