Nor Ashidi Mat Isa

Noise Adaptive Fuzzy Switching Median Filter for Salt-and-Pepper Noise Reduction

This letter presents a novel two-stage noise adaptive fuzzy switching median (NAFSM) filter for salt-and-pepper noise detection and removal. Initially, the detection stage will utilize the histogram of the corrupted image to identify noise pixels. These detected ¿noise pixels¿ will then be subjected to the second stage of the filtering action, while ¿noise-free pixels¿ are retained and left unprocessed. Then, the NAFSM filtering mechanism employs fuzzy reasoning to handle uncertainty present in the extracted local information as introduced by noise. Simulation results indicate that the NAFSM is able to outperform some of the salt-and-pepper noise filters existing in literature.

Color image segmentation using histogram thresholding - Fuzzy C-means hybrid approach

This paper presents a novel histogram thresholding - fuzzy C-means hybrid (HTFCM) approach that could find different application in pattern recognition as well as in computer vision, particularly in color image segmentation. The proposed approach applies the histogram thresholding technique to obtain all possible uniform regions in the color image. Then, the Fuzzy C-means (FCM) algorithm is utilized to improve the compactness of the clusters forming these uniform regions. Experimental results have demonstrated that the low complexity of the proposed HTFCM approach could obtain better cluster quality and segmentation results than other segmentation approaches that employing ant colony algorithm.

Adaptive fuzzy-K-means clustering algorithm for image segmentation

Clustering algorithms have successfully been applied as a digital image segmentation technique in various fields and applications. However, those clustering algorithms are only applicable for specific images such as medical images, microscopic images etc. In this paper, we present a new clustering algorithm called Adaptive Fuzzy-K-means (AFKM) clustering for image segmentation which could be applied on general images and/or specific images (i.e., medical and microscopic images), captured using different consumer electronic products namely, for example, the common digital cameras and CCD cameras. The algorithm employs the concepts of fuzziness and belongingness to provide a better and more adaptive clustering process as compared to several conventional clustering algorithms. Both qualitative and quantitative analyses favour the proposed AFKM algorithm in terms of providing a better segmentation performance for various types of images and various number of segmented regions. Based on the results obtained, the proposed algorithm gives better visual quality as compared to several other clustering methods.

Adaptive contrast enhancement methods with brightness preserving

Brightness preserving methods are very high demand to the consumer electronic products. Numerous histogram equalization (HE)-based brightness preserving methods tend to produce unwanted artifacts. Thus, we propose two methods to overcome the drawbacks. The first proposed method divides the histogram based on the median, and iteratively divides into the lower and upper sub-histograms, to produce a totally four sub-histograms. The separating points in the lower and upper sub-histograms are assigned to a new dynamic range and clipping process is implemented to each sub-histogram. Finally, the conventional HE is implemented. The second method is the extension of the bi-histogram equalization plateau limit (BHEPL). This method segments the histogram of input image based on its mean value. Then, clipping process is implemented to each sub-histogram based on their median value. The proposed methods are compared with several conventional methods. The experiment results show that, both of the proposed methods outperform those conventional methods by producing clearer enhanced image with brightness and details preserving ability.

Adaptive fuzzy moving K-means clustering algorithm for image segmentation

Image segmentation remains one of the major challenges in image analysis. Many segmentation algorithms have been developed for various applications. Unsatisfactory results have been encountered in some cases, for many existing segmentation algorithms. In this paper, we introduce three modified versions of the conventional moving k-means clustering algorithm called the fuzzy moving k-means, adaptive moving k-means and adaptive fuzzy moving k-means algorithms for image segmentation application. Based on analysis done using standard images (i.e. original bridge and noisy bridge) and hard evidence on microscopic digital image (i.e. segmentation of Sprague Dawley rat sperm), our final segmentation results compare favorably with the results obtained by the conventional k-means, fuzzy c-means and moving k-means algorithms. The qualitative and quantitative analysis done proved that the proposed algorithms are less sensitive with respect to noise. As such, the occurrence of dead centers, center redundancy and trapped center at local minima problems can be avoided. The proposed clustering algorithms are also less sensitive to initialization process of clustering value. The final center values obtained are located within their respective groups of data. This enabled the size and shape of the object in question to be maintained and preserved. Based on the simplicity and capabilities of the proposed algorithms, these algorithms are suitable to be implemented in consumer electronics products such as digital microscope, or digital camera as post processing tool for digital images.

Quadrants dynamic histogram equalization for contrast enhancement

In this paper, we introduce a histogram equalization (HE)-based technique, called quadrant dynamic histogram equalization (QDHE), for digital images captured from consumer electronic devices. Initially, the proposed QDHE algorithm separates the histogram into four (quadrant) sub-histograms based on the median of the input image. Then, the resultant sub-histograms are clipped according to the mean of intensity occurrence of input image before new dynamic range is assigned to each sub-histogram. Finally, each sub-histogram is equalized. Based on extensive simulation results, the QDHE method outperforms some methods existing in literature, which can be considered as state-of-the-arts, by producing clearer enhanced images without any intensity saturation, noise amplification, and over-enhancement. Furthermore, image details of the processed image are well preserved and highlighted. For this reason, the proposed QDHE algorithm is suitable for images captured in low-light environments - an unavoidable situation by many consumer electronics products such as camera devices in cell phone.

Particle swarm optimization with increasing topology connectivity

In this paper, we propose a new variant of particle swarm optimization (PSO), namely PSO with increasing topology connectivity (PSO-ITC), to solve unconstrained single-objective optimization problems with continuous search space. Specifically, an ITC module is developed to achieve better control of exploration/exploitation searches by linearly increasing the particles topology connectivity with time as well as performing the shuffling mechanism. Furthermore, we introduce a new learning framework that consists of a new velocity update mechanism and a new neighborhood search operator that aims to enhance the algorithms searching performance. The proposed PSO-ITC is extensively evaluated across 20 benchmark functions with various features as well as two engineering design problems. Simulation results reveal that the performance of the PSO-ITC is superior to nine other PSO variants and six cutting-edge metaheuristic search algorithms. The graphical illustration of the proposed particle swarm optimization with increasing topology connectivity (PSO-ITC), consisting of the ITC module and the proposed learning framework.Display Omitted A PSO variant, abbreviated as PSO-ITC, is developed.An ITC module is developed to achieve better balance of global/local searches.A new learning framework is proposed to improve algorithms searching performance.PSO-ITC has prominent searching accuracy and convergence speed in optimization.Results show that PSO-ITC outperforms other PSO variants and MS algorithms.

Teaching and peer-learning particle swarm optimization

The graphical illustration of the proposed teaching and peer-learning PSO (TPLPSO), consisting of the teaching phase, the peer-learning phase, and the stagnation prevention strategy (SPS). A PSO algorithms variant, abbreviated as TPLPSO, is proposed.Teaching and peer-learning framework is proposed to improve PSOs performance.Stagnation prevention strategy is proposed to mitigate the premature convergence.TPLPSO has higher searching accuracy and convergence speed during the optimization.Results show that TPLPSO outperforms other state-of-the-art PSO variants. Most of the recent proposed particle swarm optimization (PSO) algorithms do not offer the alternative learning strategies when the particles fail to improve their fitness during the searching process. Motivated by this fact, we improve the cutting edge teaching-learning-based optimization (TLBO) algorithm and adapt the enhanced framework into the PSO, thereby develop a teaching and peer-learning PSO (TPLPSO) algorithm. To be specific, the TPLPSO adopts two learning phases, namely the teaching and peer-learning phases. The particle firstly enters into the teaching phase and updates its velocity based on its historical best and the global best information. Particle that fails to improve its fitness in the teaching phase then enters into the peer-learning phase, where an exemplar is selected as the guidance particle. Additionally, a stagnation prevention strategy (SPS) is employed to alleviate the premature convergence issue. The proposed TPLPSO is extensively evaluated on 20 benchmark problems with different features, as well as one real-world problem. Experimental results reveal that the TPLPSO exhibits competitive performances when compared with ten other PSO variants and seven state-of-the-art metaheuristic search algorithms.

G2Way A Backtracking Strategy for Pairwise Test Data Generation

Our continuous dependencies on software (i.e. to assist as well as facilitate our daily chores) often raise dependability issue particularly when software is being employed harsh and life threatening or (safety) critical applications. Here, rigorous software testing becomes immensely important. Many combinations of possible input parameters, hardware/software environments, and system conditions need to be tested and verified against for conformance. Due to resource constraints as well as time and costing factors, considering all exhaustive test possibilities would be impossible (i.e. due to combinatorial explosion problem). Earlier work suggests that pairwise sampling strategy (i.e. based on two-way parameter interaction) can be effective. Building and complementing earlier work, this paper discusses an efficient pairwise test data generation strategy, called G2Way. In doing so, this paper demonstrates the correctness of G2Way as well as compares its effectiveness against existing strategies including AETG and its variations, IPO, SA, GA, ACA, and All Pairs. Empirical evidences demonstrate that G2Way, in some cases, outperformed other strategies in terms of the number of generated test data within reasonable execution time.

Clustered-Hybrid Multilayer Perceptron network for pattern recognition application

This paper introduces a modified version of the Hybrid Multilayer Perceptron (HMLP) network to improve the performance of the conventional HMLP network. We adopted the Clustering Algorithm from the Radial Basis Function (RBF) network architecture and incorporated it into the conventional HMLP network architecture. The modified model is called Clustered-Hybrid Multilayer Perceptron (Clustered-HMLP) network. The proposed Clustered-HMLP network architecture is trained using modified training algorithm called Clustered-Modified Recursive Prediction Error (Clustered-MRPE). The capability of the Clustered-HMLP network with Clustered-MRPE training algorithm is demonstrated using seven benchmark datasets from the University of California at Irvine (UCI) machine learning repository (i.e. Iris, Ionosphere, Pima Indian Diabetes, Wine, Lung Cancer, Hayes-Roth and Glass) and compared with the performance of other twelve classifiers reported in literature. Further, the new network is implemented to model a Transformer Fault Diagnosis System and Aggregate Shape Identification System. The results indicate that the proposed Clustered-HMLP network outperforms other eleven classifiers and provides a significant improvement to the conventional HMLP network for pattern recognition application.