Yih-Chih Chiou

Intelligent segmentation method for real-time defect inspection system

To extract desired flaws from various types of images, the integration of different segmentation methods is required. In this study, we present an intelligent method for automatic selection of a proper image segmentation method upon detecting a particular flaw type. The new method is capable of choosing the most suitable method from four segmentation methods currently available. The automatic selection procedures start from the pre-segmentation of an image to be examined. Then, the predetermined features are extracted from the original, foreground, and background images. After that, a suitable segmentation method will be selected using a classifier based on six features. Finally, the image is re-segmented by the selected segmentation method to discover flaws. The proposed method has been tested using 1676 defective images. The results show a significant reduction in misclassification rate from about 44% to 13.96%.

The feature extraction and analysis of flaw detection and classification in BGA gold-plating areas

In this paper, the measurements along with color image segmentation to detect all possible defects in BGA (ball grid array) type PCB (printed circuit boards) were presented. We use feature extraction and analysis as well as BPN (back-propagation neural) network classification to classify the detected defects. There are variable defects to be detected and classified including stain, scratch, solder-mask, and pinhole. The experimental results show that the proposed algorithm is successful in detecting and classifying the defects on gold-plating regions. The recognition speed becomes faster and the system becomes more flexible in comparison to the previous system. The proposed method, using unsophisticated and economical equipment, is also verified in providing highly accurate results with a low error rate.

Machine Vision-Based Automatic Raw Fish Handling and Weighing System of Taiwan Tilapia

This study proposes a vision-based automatic raw fish handling system to speed up fish cleaning and weighing. The proposed fish weighing system used a camera to capture projected images of fishes. Applying image processing techniques, physical properties of fishes, such as length, width, perimeter and area were obtained. Followed by regression analysis, weight-length, weight-height, weight-perimeter and weight-area relationships were derived. Analysis results of fifty tilapias show that coefficient of determination of the regression equation relating weight and area is 0.9303. The high value suggests that a tilapias weight is highly correlated with its projected area. Therefore, use a tilapias area to estimate its weight is justifiable.

Vision-based automatic tool wear monitoring system

The vision-based of automated tool wear monitoring systems are very important and efficient for unmanned machining systems. This research is use the machine vision inspection technique to automatic tool wear monitoring measurement of different coated drills. The tool wear images are captured using a machine vision system incorporating with an effective extract vertex algorithm based on subpixel edge detector and Gaussian filter is presented. Finally, Statically Process Control (SPC) technique is applied to detect vertices. The results show that the proposed algorithm is an effective method for the different coated drilling factor is recognized to make the most significant contribution to the over all performance. The TiAlN-coated drilling has the least wear rate amongst these coated drilling cutters and has the longest tool life in this experiment.

Automatic texture inspection in the classification of papers and cloths with neural networks method

Purpose – The purpose of this paper is to present an automatic inspection method of colors and textures classification of paper and cloth objects.Design/methodology/approach – In this system, the color image is transformed from RGB model to other suitable color model with one of the components being chosen as the gray‐level image for extracting textures. The gray‐level image is decomposed into four child images using wavelet transformation. Two child images capable of detecting variations along columns and rows are used to generate 0° and 90° co‐occurrence matrices, respectively. Some of the distinguishable texture features are derived from the two co‐occurrence matrixes. Finally, the test image is classified using neural networks. Nine color papers and eight color cloths are used to test the developed classification method.Findings – The results show that recognition rate higher than 97.86 percent can be achieved if color and texture features are both used as the inputs to the networks.Originality/value ...

HYBRID REGISTRATION OF CORRESPONDING MAMMOGRAM IMAGES FOR AUTOMATIC DETECTION OF BREAST CANCER

Mammogram registration is a critical step in automatic detection of breast cancer. Much research has been devoted to registering mammograms using either feature-matching or similarity measure. However, a few studies have been done on combining these two methods. In this research, a hybrid mammogram registration method for the early detection of breast cancer is developed by combining feature-based and intensity-based image registration techniques. Besides, internal and external features were used simultaneously during the registration to obtain a global spatial transformation. The experimental results indicates that the similarity between the two mammograms increases significantly after a proper registration using the proposed TPS-registration procedures.

An Effective Drilling Wear Measurement based on Visual Inspection Technique

The purpose of this research is to use the visual inspection technique for the automatic tool wear measurement of different coated drills. The tool wear images with the different coated drilling are captured using a machine vision system incorporating with an effective vertex detection algorithm based on subpixel edge detector and Gaussian filter is presented. The results show that the proposed algorithm is an effective method for the different coated drilling factor is recognized to make the most significant contribution to the over all performance. All drilling tests were carried out under dry cutting conditions without any coolant being used, The TiAlN-coated drilling has the least wear rate amongst these coated drilling cutters and has the longest tool life in this experiment

Investigation into Micro Machining Cutting Parameters of PMMA Polymer Material Using Taguchi’s Method

This paper applied Grey-Taguchi method to optimize the micro-drilling of PMMA polymer with multiple performance characteristics. The four parameters being optimized are coating layer, feed rate, spindle speed, and depth of cut. The performance of the drilling process was evaluated by two performance characteristics, namely drill wear and surface roughness. The orthogonal array, grey relational analysis, and analysis of variance were used to study the two performance indices. The optimal combination of parameters was determined by using the grey relational grade, a performance index formed by combining the two performance characteristics. The experimental results show that TiAlN-coating drills generate least wear and thus possess the longest tool life and the best holes quality. Finally, confirmation experiments were conducted to confirm the validity of the results.

Taguchi Analysis of Milling Wear Automatic Monitoring System Based on Machine Vision Technique

This study proposes a tool wear monitoring system based on machine vision technique. The tool wear of single edge rhombus micro-end-mills with mill parameters (Side Clearance Angle, different coating layer, feed rate and spindle speed) in milling 6061 aluminum alloy was experimentally investigated in this study. A L 9 (34) orthogonal array, analysis of variance (ANOVA) and signal-to-noise (S/N) were determined to know the level of importance of the machining parameters. Using Taguchi method for design of a robust experiment, the interactions among factors are also investigated. The experimental results indicate that Side Clearance Angle and coating layer are recognized to make the most significant contribution to the overall performance. The correlation was obtained by multi-variable nonlinear regression and compared with the experimental results. The experimental results showed that Using TiCN-coated micro-end-mill and setting side clearance angle at 12 degrees, spindle speed at 6000 rpm and feed rate at 0.0125 mm/rev minimized the wear on micro-end mills and maximized tool life. The confirmation tests demonstrated a feasible and an effective method for the evaluation of tool wear in milling of 6061 aluminum alloy.

Segmentation of unevenly illuminated line scanned images

Purpose – Though many segmentation methods have been published, few of them are developed especially for line scanned images. An ill‐illuminated line scanned (IILS) image tends to have a uniform intensity distribution in column direction while non‐uniform intensity distribution in the row direction. So, it is improper to segment IILS images using either a pixed threshold or threshold surface. In view of this, the purpose of this paper is to develop a segmentation method that is suitable for segmented IILS images.Design/methodology/approach – To obtain satisfactory segmentation results, the illumination variation across the column of a line scanned image was taken into account and a column‐based segmentation method was developed. The method first calculates each columns standard deviation. Then a threshold value is automatically assigned to each column based on the derived values. Finally, by assembling each columns threshold value, a so‐called threshold line is formed. The method is threshold‐line segmen...