American Journal of Engineering and Applied Sciences 6 (3): 263-273, 2013 ISSN: 1941-7020 © 2014 A. TeohOng et al ., This open access article is distributed under a Creative Commons Attribution (CC-BY) 3.0 license doi:10.3844/ajeassp.2013.263.273 Published Online 6 (3) 2013 (http://www.thescipub.com/ajeas.toc) Computer Machine Vision Inspection on Printed Circuit Boards Flux Defects 1Ang TeohOng, 2Zulkifilie Bin Ibrahim and 3Suzaimah Ramli 1Control Ez Technology Sdn. Bhd, No.4-4, Jalan SP2/2, Serdang Perdana, 43300 Sri Kembangan, Selangor, Malaysia 2Department of Electrical Engineering, Faculty of Electrical Engineering, Universiti Teknikal Malaysia Melaka (UTeM), Hang Tuah Jaya, 76100, Durian Tunggal, Melaka, Malaysia 3Department of Computer Science, Faculty of Defense Science and Technology, Universiti Pertahanan Nasional Malaysia, Kem Sg Besi, 57000 Kuala Lumpur, Malaysia Received 2013-07-08, Revised 2013-08-06; Accepted 2013-08-07 ABSTRACT The new visual inspection systems techniques using real time machine vision replace the human visual manual inspection on PCB flux defects, which brings harmful effects on the board which may come in the form of corrosion and can cause harm to the assembly. In short, it brings improvement in Printed Circuit Boards (PCB) production quality, principally concerning the acceptance or rejection of the PCB boards. To develop new algorithm in image processing which detects flux defect at PCB board during re-flow process and achieve good accuracy of the PCB quality checking. The machine will be designed and fabricated with the total automation control system with mechanical PCB loader/un- loader, pneumatic system handler with vacuum cap, vision inspection station and final classification station (accept or reject). The image processing system is based on shape (pattern) and color image analysis techniques with Matrox Imaging Library. The shape/texture of the PCB pins is analyzed by using pattern matching technique to detect the PCB flux defect area. The color analysis of the flux defect in a PCB boards are processed based on their red color pixel percentage in Red, Green and Blue (RGB) model. The red color filter band mean value of histogram is measured and compared to the value threshold to determine the occurring on the PCB flux defects. The system was tested with PCB boards from factory production line and achieved PCB board flux defects sorting accuracy at 86.0% based on proposed pattern matching technique combined with red color filter band histogram. Keywords: Machine Vision, Pattern Matching Technique, RGB Color Model, Red Color Filter Band, Threshold, Histogram 1. INTRODUCTION visualize by human eye because it’s liquid and transparent. The result, obtained based on the proposed The main disadvantage of manual inspection of PCB technique is possibly be applied in automated PCB defects are human errors, inconsistent grading and labour manufacturing process. intensive. The inspection process can be automated by The machine will be designed and fabricated with “PC-Based Machine Vision”. To identify the fatal the Total Automation Control System with mechanical defects the system uses a connectivity approach, it finds PCB loader/un-loader, pneumatic system handler with any type of error like: PCB board printing and labeling, vacuum cap, vision inspection station and final scratches, marking on components, components classification station (accept or reject). orientation and others. For this research, the main defect Computer Vision techniques were used to develop focus is on PCB flux defect, which is very difficult to an automatic visual inspection of PCB boards, which Corresponding Author: Ang TeohOng, Control Ez Technology Sdn. Bhd, No.4-4, Jalan SP2/2, Serdang Perdana, 43300 Sri Kembangan, Selangor, Malaysia Science Publications 263 AJEAS Ang TeohOng et al . / American Journal of Engineering and Applied Sciences 6 (3): 263-273, 2013 intends to evaluate the various PCB board defects. classified in two categories: the one with excessive The fault detection strategy refers to the use of copper and missing copper. The incomplete etching referential inspection methods, in which the reference process leaves unwanted conductive materials and is a board artwork or a manufactured board without forms defects like short-circuit, extra hole, protrusion, errors. The PCB defects are normally grouped in two island and small space. Excessive etching leads to categories, the fatal defects (reject units) and no open-circuit and thin pattern on PCB. In addition, defects (accept units). The system identifies the fatal some other defects may exist on bare PCB, i.e. defects using an image comparison technique, missing holes, scratch and cracks. et al subtracting the reference board image from the tested Mar . (2009) proposed a front-end system for board image. This project has been designed to the automatic detection, localization and segmentation combine all aspects of engineering, including of solder joint defects. An illumination normalization mechanical, electrical, electronic, communications approach is applied to an image, which can effectively and software engineering, into one development and efficiently eliminate the effect of uneven product. The machine will be able to inspect and illumination while keeping the properties of the separate the PCBs defects board from good ones. processed image the same as in the corresponding image under normal lighting conditions. Consequently 1.1. Literature Review special lighting and instrumental setup can be reduced Numerous PCB inspection algorithms have been in order to detect solder joints. In the segmentation proposed in the literature to date. approach, the PCB image is transformed from an RGB Greenberg et al . (2006), this invention discloses a color space to a YIQ color space for the effective method for Printed Circuit Board (PCB) inspection, detection of solder joints from the background. et al including providing a multiplicity of PCBs placed on Zeng . (2011) proposed an algorithm of an inspection panel, defingng each non-indentical allocating and identifying PCB board components based PCB in terms of geometry and features which are to on color distribution of solder joints, the proposed be inspected. approach analyzes the color distribution patterns of Leta et al . (2005), This study uses some computer solders joints under three layers of ring-shaped LEDs. vision techniques to measure parts and discusses Chuhan and Bhardwaj (2011) used image common difficulities of automated inspection. The subtraction method to detect PCB Bare defects which parts conformity analysis using a non-contact involves loading a reference image, buffering the measurement system has been adopted specially to reference image for subtraction operation and small objects, where accurate instruments like inspecting the PCB error; inspected image is XORed Coordinate Measuring Machine (CMM) is used. with reference image. The obtained defected area Tsai and Tsai (2002), in this study a rotation- undergoes “particle analysis” which is used to analyze invariant pattern-mathcing scheme for detecting the defects in terms of area, size, orientation and objects in complex color images is proposed. The percentage. Kaushik and Ashraf (2012) also tested on complexity and computational load for matching image subtraction method to detect PCB defects under colored objects in arbitary orientations are reduced different noise levels. significantly by the 1-D color ring-protection Bhardwaj (2012) proposed an Automated Optical representation. Inspection (AOI) algorithm for PCB inspection Singh and Bharti (2012), The on-line or automatic system. The system performs measurement and visual inspection of PCB is basically a very first detection of holes defect that occur during PCB examination before its electronic testing. This etching process during manufacturing. One image inspection consists of mainly missing or wrongly feature is measured from the examined image and is placed components in the PCB. If there is any missing used for detection of defects. Different steps and electronic component then it is not so damaging the methodology are used for defect detection. At first, an PCB. image is selected for inspection from the location Mashohor et al . (2004), this study presents the first where captured image are stored. To enhance the prototype of automating a low-cost printed circuit (PCB) image for detection filters are used. Low pass filter inspection on physical defects through the development and highlight detail filter are used and grayscale of a technique for image detection using Genetic image converted into binary one by applying Algorithm (GA). threshold. For the detection of defects algorithms are Sundaraj (2009), during etching process, the applied on the enhanced image. Advanced anomalies occurring on bare PCB could be largely Morphology is used to remove the boarder object in Science Publications 264 AJEAS Ang TeohOng et al . / American Journal of Engineering and Applied Sciences 6 (3): 263-273, 2013 the binary image. This process removes unwanted mechanical handler with electro-pneumatic system. portion of image. On the resultant image Particle The machine vision constructed with these Analysis functions i.e., circle detection and Particle componenets is shown in Fig. 1 along with block measurement are applied. Output of these functions diagram of inspection system in Fig. 2. gives the desired results in terms of specified parameters. 2.1. Camera Specification 2. MATERIALS AND METHODS