In this paper, a new similarity measure for fractal image compression (FIC) is introduced. In the proposed Huber fractal image compression (HFIC), the linear Huber regression technique from robust statistics is embedded into the encoding procedure of the fractal image compression. When the original image is corrupted by noises, we argue that the fractal image compression scheme should be insensitive to those noises presented in the corrupted image. This leads to a new concept of robust fractal image compression. The proposed HFIC is one of our attempts toward the design of robust fractal image compression. The main disadvantage of HFIC is the high computational cost. To overcome this drawback, particle swarm optimization (PSO) technique is utilized to reduce the searching time. Simulation results show that the proposed HFIC is robust against outliers in the image. Also, the PSO method can effectively reduce the encoding time while retaining the quality of the retrieved image.

As is well known in statistics, the resulting linear regressors by using the rank-based Wilcoxon approach to linear regression problems are usually robust against (or insensitive to) outliers. This motivates us to introduce in this paper the Wilcoxon approach to the area of machine learning. Specifically, we investigate four new learning machines, namely Wilcoxon neural network (WNN), Wilcoxon generalized radial basis function network (WGRBFN), Wilcoxon fuzzy neural network (WFNN), and kernel-based Wilcoxon regressor (KWR). These provide alternative learning machines when faced with general nonlinear learning problems. Simple weights updating rules based on gradient descent will be derived. Some numerical examples will be provided to compare the robustness against outliers for various learning machines. Simulation results show that the Wilcoxon learning machines proposed in this paper have good robustness against outliers. We firmly believe that the Wilcoxon approach will provide a promising methodology for many machine learning problems.

The well-known (two-parameter) sequential minimal optimization (2PSMO) algorithm for support vector classification is generalized to three-parameter sequential minimal optimization (3PSMO) algorithm in this paper. This new algorithm retains all the good properties of the former one. The main difference between these two algorithms is that the optimization is performed in each iteration of the 2PSMO algorithm on a line segment, whilst that of the 3PSMO algorithm on a region consisting of infinitely many line segments. Four public data sets are used to show the performance of both algorithms.

In this paper, the Particle Swarm Optimization (PSO) is utilized to enhance the concavity searching for the control points of Active Contour Model (ACM). In the traditional methods for ACM, the best candidates are searched in a small window. Consequently, the boundary concavities cannot be searched accurately. Some improvements have been made in the past to enlarge the searching space, yet they are still time-consuming. To overcome these drawbacks, a multi-population particle swarm optimization technique is adopted in this paper to reduce the search time but in a larger searching window. This optimizer inherits the spirit of the original PSO as well as shares information among the swarms surrounding control points. Experimental results demonstrate that the proposed method can improve the search of object concavities without spending extra time.

In this paper, we propose five simple algorithms for solving the least square support vector machine (LS-SVM) learning problems. For linear regression, we first present a Widrow-Hoff-like algorithm for the primal optimization problem. The dual form of this algorithm is then provided. For kernel-based nonlinear LS-SVM, we first present a Widrow-Hoff-like algorithm. The elegant and powerful two-parameter sequential minimization optimization (2P-SMO) algorithm is then provided. Finally, we give a detailed derivation of the three-parameter sequential minimization optimization (3P-SMO) algorithm. A numerical example is provided.

An adaptive controller composed of several independent self tuning neurons for multivariable dynamic systems is developed. Unlike traditional fully connected neural network, there are no synaptic connections among the independent neurons in our proposed controller. Each self tuning neuron can be regarded as an independent adaptive controller. It eases the hardware implementation due to the simplicity of our proposed neuron controller. Two case studies are provided to demonstrate the use of the proposed controller where the system dynamics is fully known or only partially known.