IEEE Region 10 Conference 

About: IEEE Region 10 Conference is an academic conference. The conference publishes majorly in the area(s): Electric power system & Control theory. Over the lifetime, 11025 publications have been published by the conference receiving 57993 citations.

Microelectronics reliability

Abstract: The author points out the importance of the involvement of the design and manufacturing team in achieving reliability of microelectronic devices. A method of verifying reliability goals through calculation of failure rates based on life test parameters is described. An example illustrating the method is shown. >

A Bayesian approach to skin color classification in YCbCr color space

Abstract: This paper addresses an image classification technique that uses the Bayes decision rule for minimum cost to classify pixels into skin color and non-skin color. Color statistics are collected from YCbCr color space. The Bayesian approach to skin color classification is discussed along with an overview of YCbCr color space. Experimental results demonstrate that this approach can achieve good classification outcomes, and it is robust against different skin colors.

A wavelet based statistical approach for speckle reduction in medical ultrasound images

Abstract: The paper introduces a novel speckle reduction method based on soft thresholding the wavelet coefficients of the logarithmically transformed medical ultrasound image. The method is based on the generalized Gaussian distributed (GGD) modeling of subband coefficients. The proposed method is a variant of the recently published BayesShrink method (Chang, G et al., IEEE Trans. Image Processing, vol.9, no.9, p.1522-31, 2000) derived in the Bayesian framework for denoising natural images. It is scale adaptive because the parameters required for estimating the threshold depend on scale and subband data. The threshold is computed by K/spl sigma//sup 2///spl sigma//sub x/ where /spl sigma/ and /spl sigma//sub x/ are the standard deviation of the noise and the subband data of the noise-free image, respectively, and K is a scale parameter. Experimental results show that the proposed method performs better than the median filter as well as the homomorphic Wiener filter, especially in terms of feature preservation for better diagnosis as desired in medical image processing.

Performance analysis of software reliability growth models with testing-effort and change-point

Abstract: In this paper, a scheme for constructing software reliability growth model based on Non-Homogeneous Poisson Process is proposed. The main focus is to provide a method for software reliability modeling, which considers both testing-effort and change-point. In the vast literature, most researchers assume a constant detection rate per fault in deriving their software reliability models. They suppose that all faults have equal probability of being detected during the software testing process, and the rate remains constant over the intervals between fault occurrences. In reality, the fault detection rate strongly depends on the skill of test teams, program size, and software testability. Therefore, it may not be smooth and can be changed. On the other hand, sometimes we have to detect more additional faults in order to reach the desired reliability objective during testing. It is advisable for project managers to purchase new automated test tool, technology or additional manpower. These approaches can provide a conspicuous improvement in software testing and productivity. In this case, the fault detection rate will be changed during the software development process. Therefore, here we incorporate both generalized logistic testing-effort function and change-point parameter into software reliability modeling. New theorems are proposed and software testing data collected from real application are utilized to illustrate the proposed model. Experimental results show that the proposed framework to incorporate both testing-effort and change-point for SRGM has a fairly accurate prediction capability.

Classification of ECG arrhythmias using multi-resolution analysis and neural networks

Abstract: Automatic detection and classification of cardiac arrhythmias is important for diagnosis of cardiac abnormalities. We propose a method to accurately classify ECG arrhythmias through a combination of wavelets and artificial neural networks (ANN). The ability of the wavelet transform to decompose signal at various resolutions allows accurate extraction/detection of features from non-stationary signals like ECG. A set of discrete wavelet transform (DWT) coefficients, which contain the maximum information about the arrhythmia, is selected from the wavelet decomposition. These coefficients are fed to the back-propagation neural network which classifies the arrhythmias. The proposed method is capable of distinguishing the normal sinus rhythm and 12 different arrhythmias and is robust against noise. The overall accuracy of classification of the proposed approach is 96.77%.