Wen-Yen Wu

Automated inspection of printed circuit boards through machine vision

Abstract This paper introduces the development of an automated visual inspection system for printed circuit boards (PCBs). It utilizes an elimination-subtraction method which directly subtracts the template image from the inspected image, and then conducts an elimination procedure to locate defects in the PCB. Each detected defect is subsequently classified into one of the seven defect types by three indices: the type of object detected, the difference in object numbers, and the difference in background numbers between the inspected image and the template. Finally, a 256 × 240 PCB image was tested to show the effectiveness of this system.

An adaptive method for detecting dominant points

Abstract In this paper, we propose an adaptive method for the polygonal approximation of a digitized curve. Instead of setting a fixed length of support region in advance, the new method will compute the suitable length of support region for each point to find the best approximated curvature. The dominant points are identified as the points with local maximum curvatures. In addition, the break point detection is conducted to reduce the computations. The experimental results show that the proposed method can approximate the curves effectively.

Elliptical object detection by using its geometric properties

Abstract A simple algorithm for ellipse detection in an image is proposed. The new method consists of three steps: (1) detect the elliptical center by using its symmetrical property; (2) identify the elliptical points by using a simple geometric property; and (3) estimate the five parameters ( x c , y c , A , B , θ ) by using a least sum of squares fitting method. The new method has been tested on both the synthetic and real images. The experimental results show that the method is reliable and accurate.

Dominant point detection using adaptive bending value

Abstract An efficient method for dominant point detection is proposed in this paper. The region of support for each point on curve is determined using bending value. The points with local maximum smoothing bending value can be located as the dominant points on the curve. The proposed algorithm needs no input parameter. The experimental results show that the new method is efficient and effective in detecting dominant points.

Corner detection using bending value

An efficient algorithm for corner detection is proposed in this paper. It utilizes the concept that the directions of the forward and backward vectors of a non-corner point will cancel each other to detect corners. The bending value is used to assess the degree of possibility of a point being a corner. The experimental results show that the method is efficient and effective in detecting corners from an image.

Performance evaluation of some noise reduction methods

Abstract In this paper, a quantitative evaluation of some noise reduction methods is presented. It first classifies the existing local noise reduction techniques by the use of neighbors and statistics and proposes some new filtering techniques. Subsequently, five performance evaluation criteria including processing time, mean distortion, mean busyness, stability, and correct processing ratio are discussed. By using these performance evaluation criteria, an experiment that involves four factors (test image, noise, mask size, and iteration) was designed to evaluate the performance of the local noise reduction methods. The results indicate that some new local noise reduction techniques have performance better than that of the standard existing techniques under certain criteria.

Two-dimensional object recognition through two-stage string matching

A two-stage string matching method for the recognition of two-dimensional (2-D) objects is proposed in this work. The first stage is a global cyclic string matching. The second stage is a local matching with local dissimilarity measure computing. The dissimilarity measure function of the input shape and the reference shape are obtained by combining the global matching cost and the local dissimilarity measure. The proposed method has the advantage that there is no need to set any parameter in the recognition process. Experimental results indicate that the hostage string matching approach significantly improves the recognition rates compared to the one-stage string matching method.

Automated post bonding inspection by using machine vision techniques

Inspection plays an important role in the semiconductor industry. In this paper, we focus on the inspection task after wire bonding in packaging. The purpose of wire bonding (W/B) is to connect the bond pads with the lead fingers. Two major types of defects are (1) bonding line missing and (2) bonding line breakage. The numbers of bonding lines and bonding balls are used as the features for defect classification. The proposed method consists of image preprocessing, orientation determination, connection detection, bonding line detection, bonding ball detection, and defect classification. The proposed method is simple and fast. The experimental results show that the proposed method can detect the defects effectively.

A dynamic method for dominant point detection

Detecting dominant points is an important step for shape representation. Most of dominant point detection methods attend to preset or find the region of support of each point. In this paper, we demonstrate an improved method for determining the region of support in the dominant point detection. Instead of setting the regions of support for points independently, the region of support dynamically depending on the previous region of support. The experimental results show that the proposed method is effective in detecting dominant points.

A system for automated BGA inspection

For the advantages of high I/O count and small size, ball grid array (BGA) package techniques have been developing very fast in recent years. In this paper, a vision-based algorithm was proposed. The BGA grayscale images captured by CCD camera, and some image processing methods were applied to segment and enhance the image of solder ball. Three kinds of feature size were employed here: area, centroid and roundness of solder ball. The defects were detected by matching these features to the template data. The inspection items included missing ball, extra ball, ball bridging, oversize or undersize, offset and deformation. In this research, the speed execution time of inspecting one BGA image is also taken into consideration