Chih-Chin Lai

Digital Image Watermarking Using Discrete Wavelet Transform and Singular Value Decomposition

The main objective of developing an image-watermarking technique is to satisfy both imperceptibility and robustness requirements. To achieve this objective, a hybrid image-watermarking scheme based on discrete wavelet transform (DWT) and singular value decomposition (SVD) is proposed in this paper. In our approach, the watermark is not embedded directly on the wavelet coefficients but rather than on the elements of singular values of the cover images DWT subbands. Experimental results are provided to illustrate that the proposed approach is able to withstand a variety of image-processing attacks.

A digital watermarking scheme based on singular value decomposition and tiny genetic algorithm

A robust digital image watermarking scheme based on singular value decomposition (SVD) and a tiny genetic algorithm (Tiny-GA) is proposed in this paper. Previous works have shown that both one-way and non-symmetric properties of SVD make it desirable for watermarking techniques. The produced singular values are very stable and vary very little under various image processing operations or attacks. In the proposed scheme, the singular values of a cover image are modified by multiple scale factors to embed the watermark image. Since the values of scale factors determine the watermark strength; therefore, we use the Tiny-GA to search the proper values in order to improve the visual quality of the watermarked image and the robustness of the watermark. Experimental results demonstrate that our scheme is able to withstand a variety of image processing attacks.

A hierarchical evolutionary algorithm for automatic medical image segmentation

Image segmentation denotes a process of partitioning an image into distinct regions. A large variety of different segmentation approaches for images have been developed. Among them, the clustering methods have been extensively investigated and used. In this paper, a clustering based approach using a hierarchical evolutionary algorithm (HEA) is proposed for medical image segmentation. The HEA can be viewed as a variant of conventional genetic algorithms. By means of a hierarchical structure in the chromosome, the proposed approach can automatically classify the image into appropriate classes and avoid the difficulty of searching for the proper number of classes. The experimental results indicate that the proposed approach can produce more continuous and smoother segmentation results in comparison with four existing methods, competitive Hopfield neural networks (CHNN), dynamic thresholding, k-means, and fuzzy c-means methods.

A Novel Image Segmentation Approach Based on Particle Swarm Optimization

Image segmentation denotes a process by which an image is partitioned into non-intersecting regions and each region is homogeneous. Utilizing histogram information to aim at segmenting an image is a commonly used method for many applications. In this paper, we view the image segmentation as an optimization problem. We find a curve which gives the best fit to the given image histogram, and the parameters in the curve are determined by using the particle swarm optimization algorithm. The experimental results to confirm the proposed approach are also included.

Genetic subgradient method for solving location-allocation problems

Facility location-allocation (FLA) problem is a very important subject in todays business. It is an important part of a companys global logistic system. Various FLA problems have been considered in operations research (OR) under somehow stringent conditions. Restrictive conditions are placed to reduce the size of the search space, however, they also make the model inappropriate for the real-business world. In this paper, we consider a class of FLA problems that can assume more realistic conditions in real-life applications. A hybrid method of genetic algorithm and subgradient technique is used to solve the problem efficiently.

A Novel Clustering Approach using Hierarchical Genetic Algorithms

Clustering is a central task for data analysis that partitions heterogeneous data sets into groups of more homogeneous characteristics. However, most of clustering algorithms require the user to provide the number of clusters as input. In this paper, we consider the automatic clustering problem that one has to partition data points without any a priori knowledge about the correct number of clusters. The hierarchical genetic algorithm (HGA) is employed for automatically searching the number of clusters as well as properly locating the centers for clusters. The well-known Davies-Bouldin index is adopted as a measure of the validity of the clusters. Experimental results on artificial and real-life data sets are given to illustrate the effectiveness of the proposed approach.

Image Watermarking Scheme Using Singular Value Decomposition and Micro-genetic Algorithm

In this paper, we introduce a novel image watermarking scheme using singular value decomposition (SVD) and micro-genetic algorithm (micro-GA). In an SVD-based watermarking scheme, the singular values of the cover image are modified by multiple scaling factors to embed the watermark image. The proper values of scaling factors are optimized and obtained efficiently by means of the micro-GA. Experimental results are provided to illustrate the feasibility of the proposed approach.

Particle Swarm Optimization-Aided Feature Selection for Spam Email Classification

Using a finite set of features to help determine an email as spam or non-spam is a very popular way. However, in most cases, the feature selection is empirically verified. This paper investigates how particle swarm optimization algorithm can help select features relevant for spam email classification. The experimental results show that the proposed approach selects the most proper discriminative features while eliminating irrelevant ones.

An empirical study on mining sequential patterns in a grid computing environment

Mining sequential patterns (MSP) is an important task for knowledge discovery and data mining (KDD). Like in most KDD tasks, MSP also invokes a number of iterations for generating, adjusting, and comparing data. This paper presents an empirical study on deploying MSP in a grid computing environment and demonstrates the effectiveness and performance improvements gained in this deployment. GSP, which is a typical MSP method, is used as the mining algorithm to be investigated. A grid computing environment is designed and implemented, where all GSP functions are organized as loosely coupled web-services. MSP is achieved through the cooperation of these web-services using the divide-and-conquer strategy. Several monitoring mechanisms are developed to help manage the MSP process. The experimental results show that the proposed grid computing environment provides a flexible and efficient platform for MSP.

Image Watermarking Scheme Using Genetic Algorithm

With the development of Internet and multimedia technologies, multimedia copyright protection and content authentication have become serious problems that need to be solved urgently. Digital watermarking has been regarded as an effective solution to protect various kinds of digital contents against illegal use. In this paper, a watermarking technique which uses the singular value decomposition is presented. The singular values of a cover image are modified by multiple scaling factors to embed the watermark image. since the values of scaling factors determine the watermark strength, therefore, we use the genetic algorithm to search the proper values in order to satisfy both imperceptibility and robustness requirements. Experimental results are provided to illustrate that the proposed technique is able to withstand some image-processing attacks.