Chien-Hsing Chou

A new cluster validity measure and its application to image compression

Many validity measures have been proposed for evaluating clustering results. Most of these popular validity measures do not work well for clusters with different densities and/or sizes. They usually have a tendency of ignoring clusters with low densities. In this paper, we propose a new validity measure that can deal with this situation. In addition, we also propose a modified K-means algorithm that can assign more cluster centres to areas with low densities of data than the conventional K-means algorithm does. First, several artificial data sets are used to test the performance of the proposed measure. Then the proposed measure and the modified K-means algorithm are applied to reduce the edge degradation in vector quantisation of image compression.

The Generalized Condensed Nearest Neighbor Rule as A Data Reduction Method

In this paper, we propose a new data reduction algorithm that iteratively selects some samples and ignores others that can be absorbed, or represented, by those selected. This algorithm differs from the condensed nearest neighbor (CNN) rule in its employment of a strong absorption criterion, in contrast to the weak criterion employed by CNN; hence, it is called the generalized CNN (GCNN) algorithm. The new criterion allows GCNN to incorporate CNN as a special case, and can achieve consistency, or asymptotic Bayes-risk efficiency, under certain conditions. GCNN, moreover, can yield significantly better accuracy than other instance-based data reduction methods. We demonstrate the last claim through experiments on five datasets, some of which contain a very large number of samples

A binarization method with learning-built rules for document images produced by cameras

In this paper, we propose a novel binarization method for document images produced by cameras. Such images often have varying degrees of brightness and require more careful treatment than merely applying a statistical method to obtain a threshold value. To resolve the problem, the proposed method divides an image into several regions and decides how to binarize each region. The decision rules are derived from a learning process that takes training images as input. Tests on images produced under normal and inadequate illumination conditions show that our method yields better visual quality and better OCR performance than three global binarization methods and four locally adaptive binarization methods.

Estimation of skew angles for scanned documents based on piecewise covering by parallelograms

We propose a fast and robust skew estimation method for scanned documents that estimates skew angles based on piecewise covering of objects, such as textlines, figures, forms, or tables. The method first divides a document image into a number of non-overlapping slabs in which each object is covered by parallelograms. It then estimates the skew angle based on these parallelograms or, equivalently, their complementary regions. Putting our method to a systematic test and comparing it with some alternatives, we find that it yields favorable results in terms of accuracy, sensitivity to non-textual objects, effectiveness in dealing with documents of unspecified reading order, and computational efficiency. Some work is also conducted to find an effective way to further shorten its computation time at the expense of an extremely small loss of accuracy.

A prototype classification method and its use in a hybrid solution for multiclass pattern recognition

In this paper, we propose a prototype classification method that employs a learning process to determine both the number and the location of prototypes. This learning process decides whether to stop adding prototypes according to a certain termination condition, and also adjusts the location of prototypes using either the K-means (KM) or the fuzzy c-means (FCM) clustering algorithms. When the prototype classification method is applied, the support vector machine (SVM) method can be used to post-process the top-rank candidates obtained during the prototype learning or matching process. We apply this hybrid solution to handwriting recognition and address the convergence behavior and runtime consumption of the prototype construction process, and discuss how to combine our prototype classifier with SVM classifiers to form an effective hybrid classifier.

A prototype classification method and its application to handwritten character recognition

We propose a new prototype classification method that can be combined with support vector machines (SVM) (Cortes, C and Vapnik, V, 1995) for recognizing handwritten numerals and Chinese characters. This method employs a learning process for determining both the number and location of prototypes. The possible techniques used in this process for adjusting the location of prototypes include the K-means (KM) algorithm and the fuzzy c-means (FCM) algorithm (Bezdek, J. C., 1981). When the prototype classification method is applied, the SVM method can be used to process top rank candidates obtained in the prototype learning or matching process. We apply this hybrid method to the recognition of handwritten numerals and Chinese characters. Experiment results show that this hybrid method saves great amount of training and testing time when the number of character types is large, and achieves comparable accuracy rates to those achieved by using SVM solely. Our results also show that the proposed method performs better than the nearest neighbor (NN) classification method. These outcomes suggest that the proposed method can serve as an effective solution for large-scale multiclass classification

A Novel Measure for Quantifying the Topology Preservation of Self-Organizing Feature Maps

Recently, feature maps have been applied to various problem domains. The success of some of these applications critically depends on whether feature maps are topologically ordered. In this paper, we propose a novel measure for quantifying the neighborhood preserving property of feature maps. Two data sets were tested to illustrate the performance of the proposed method.

A new approach to fuzzy classifier systems and its application in self-generating neuro-fuzzy systems

A classifier system is a machine learning system that learns syntactically simple string rules (called classifiers) through a genetic algorithm to guide its performance in an arbitrary environment. In a classifier system, the bucket brigade algorithm is used to solve the problem of credit assignment, which is a critical problem in the field of reinforcement learning. In this paper, we propose a new approach to fuzzy classifier systems and a neuro-fuzzy system referred to as ACSNFIS to implement the proposed fuzzy classifier system. The proposed system is tested by the balancing problem of a cart pole and the back-driving problem of a truck to demonstrate its performance.

Application of associative memory in human face detection

In this paper we present an associative-memory-based face detection system. First, the symmetry of human faces is used to quickly locate all the candidates of human faces with all possible sizes and locations. Then two associative memories are used to decide whether or not a human face exists at the locations. Some experimental results are given.

A reinforcement-learning approach to robot navigation

This paper presents a reinforcement-learning approach to a navigation system which allows a goal-directed mobile robot to incrementally adapt to an unknown environment. Fuzzy rules which map current sensory inputs to appropriate actions are built through the reinforcement learning. Simulation results illustrate the performance of the proposed navigation system. In this paper, ACSNFIS is used as the main network architecture to implement the reinforcement-learning based navigation system.