International Journal of Innovative Technology and Exploring Engineering (IJITEE) ISSN: 2278-3075, Volume-9 Issue-2, December 2019 3005 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Retrieval Number: B6649129219/2019©BEIESP DOI: 10.35940/ijitee.B6649.129219 Abstract: A printed circuit board without connecting with any components called as a bare PCB. Consider a PCB as a basic part which has been settled with more electronic units. In order to display the manufacturing process, the drawbacks have been taken by PCB individually. The reflection of this separation process impacts the performance of the circuits. Also, we have examined about classification methodologies as well as referential based PCB detection. From the input images, the needed and related information has been pulled out using image processing methodologies by the referential based PCB detection. Comparing with the un-defected PCB images, this was used to find out the defects. To meet the goal of the PCB defect detection, several feature extraction and pre-processing methods are derived in this article. The PCB defects have been classified by those features using the machine learning algorithms. Moreover, several types of machine learning algorithms are derived in this article. This paper helps the researchers for achieving a better solution for image processing and machine learning-based printed circuit board the defect classification. Index Terms: Image processing, printed circuit board, machine learning, defect classification, feature extraction. I. INTRODUCTION The basic structural unit of several contemporary electronic products are known as PCB (Printed Circuit Board). And, it was classified into 2 types. a) Bare PCB b) Assembled PCB The components are routed with the PCB board was called as printed circuit assembly (PCA) or printed circuit board assembly (PCBA). More Benefits of existing using PCB such as during assembly process and wiring, it reduced the likelihood of errors. Productivity of automation industrial gets increased. The appearance of more new methodologies were introduced due to increased productivity. Saved the considerable space. More complex and accurate. Although, inspecting a bare PCB is a main thing during the real PCB production process. Because of this inspection, they can find and categorize the defective circuit board earlier and left it for further proceedings. This removal helped to prevent greater loses as well as saved the cost. Also traditional detection methods contain the following. i) Electrical testing ii) X-ray methods iii) Visual Inspection. AOI is an automated inspection technology based on machine vision [1]. This modern automated inspection technology gets introduced which was named as, an Revised Manuscript Received on November 14, 2019. N. Munisankar, Department of Computer Science and Engineering, Annamalai University, Chidambaram, India. Dr. S. Nagarajan, Department of Computer Science and Engineering, Government college of engineering, Trichy, India. B. Narendra Kumar Rao, Department of Computer Science and Engineering, Sree Vidyanekethan Engineering college, Autonomous, Tirupati, India. automatic optical inspection (AOI). Due to the benefits of its high-tech and non-contact, it was generally used in PCB inspection industry [2]. The detection system was composed in the below terms as follows: a) An image processing device b) An image acquisition device c) A motion control device d) Camera. The major benefits of AOI are (i) Avoiding the limitation of weakness in artificial check. ii) Improving detection efficiency [3]. The main component of AOI was so expensive and it was very complex for several PCB manufacturers to accept [4]. So a cheaper automatic PCB inspection system [5] has to be developed as necessary. The alternative to find defect of PCB is vison-based detection methodology which is categorized into three types [6] as follows: The first one is reference comparison in which the defect area is identified based on the comparison process with the aid of template image. The advantage of reference comparison based defect detection is i) Easy to understand and ii) Intuitive. The major drawback of reference comparison based defect detection is in the photographing process, it was sensitive to the light environment. the scone one is high alignment accuracy required. The second one is non-reference comparison where the defects are discovered based on the design rules, this comparison method was used to check whether the traces and layout of the circuit board to be tested are reasonable. The major demerit of the non-reference comparison methodology easy to lose large defects and distortion characteristics. The third one is considering the advantages of reference based method and non-reference based method. However, the implementation of this hybrid comparison is very hard when compared with the other two methodologies and it requires large amount of computational complexity. No other way to provide more exact information for the further step such as repair, if the founded defect cannot be identified and categorized. So they need to analyse more about the defect classification in the defect detection process [3]. II. OVERVIEW OF PCB DEFECT DETECTION While PCB manufacturing process going on, two types of defects was happened. I. Functional Defects - It affected the PCB’s performance. II. Cosmetic Defects – It affected the PCB’s appearance [7]. PCB inspection algorithms are classified into three types. They are: 1. Referential based inspection methods – Contains model-based techniques as well as image comparison. Defect Detection in Printed Board Circuit using Image Processing N. Munisankar, S. Nagarajan, B. Narendra Kumar Rao