Wang Shitong

Positive and negative fuzzy rule system, extreme learning machine and image classification

We often use the positive fuzzy rules only for image classification in traditional image classification systems, ignoring the useful negative classification information. Thanh Minh Nguyen and QMJonathan Wu introduced the negative fuzzy rules into the image classification, and proposed combination of positive and negative fuzzy rules to form the positive and negative fuzzy rule system, and then applied it to remote sensing image/natural image classification. Their experiments demonstrated that their proposed method has achieved promising results. However, since their method was realized using the feedforward neural network model which requires adjusting the weights in the gradient descent way, the training speed is very slow. Extreme learning machine (ELM) is a single hidden layer feedforward neural network (SLFNs) learning algorithm, which has distinctive advantages such as quick learning, good generalization performance. In this paper, the equivalence between ELM and the positive and negative fuzzy rule system is revealed, so ELM can be naturally used for training the positive and negative fuzzy rule system quickly for image classification. Our experimental results indicate this claim.

Robust fuzzy clustering-based image segmentation

The fuzzy clustering algorithm fuzzy c-means (FCM) is often used for image segmentation. When noisy image segmentation is required, FCM should be modified such that it can be less sensitive to noise in an image. In this correspondence, a robust fuzzy clustering-based segmentation method for noisy images is developed. The contribution of the study here is twofold: (1) we derive a robust modified FCM in the sense of a novel objective function. The proposed modified FCM here is proved to be equivalent to the modified FCM given by Hoppner and Klawonn [F. Hoppner, F. Klawonn, Improved fuzzy partitions for fuzzy regression models, Int. J. Approx. Reason. 32 (2) (2003) 85-102]. (2) We explore the very applicability of the proposed modified FCM for noisy image segmentation. Our experimental results indicate that the proposed modified FCM here is very suitable for noisy image segmentation.

Applying the improved fuzzy cellular neural network IFCNN to white blood cell detection

Although algorithm NDA based on the fuzzy cellular neural network (FCNN) has indicated the basic superiority in its adaptability and easy hardware-realization for microscopic white blood cell detection [Wang Shitong, Wang Min, A new algorithm NDA based on fuzzy cellular neural networks for white blood cell detection, IEEE Trans. Inf. Technol. Biomed., accepted], it still does not work very well in keeping the boundary integrity of a white blood cell. In this paper, the improved version of FCNN called IFCNN is proposed to tackle this issue. The distinctive characteristic of IFCNN is to incorporate the novel fuzzy status containing the useful information beyond a white blood cell into its state equation, resulting in enhancing the boundary integrity. Our theoretical analysis shows that IFCNN has the global stability and the experimental results demonstrate its obvious advantage over FCNN in keeping the boundary integrity.

Fuzzy taxonomy, quantitative database and mining generalized association rules

Mining association rules from databases is still a hot topic in data mining community in recent years. Due to the existence of multiple levels of abstraction (i.e, taxonomic structures) among the attributes of the databases, several algorithms were proposed to mine generalized Boolean association rules upon all levels of presumed crisp taxonomic structures. However, fuzzy taxonomic structures may be more suitable in many practical applications. In [9], the authors proposed an approach to mine generalized Boolean association rules with such fuzzy taxonomic structures. The main contribution of this paper is to extend their idea to mine generalized association rules from quantitative databases with fuzzy taxonomic structures. A new fuzzy taxonomic quantitative database model is presented, and the experimental results on realistic databases are demonstrated to validate this new model.

Fuzzy kernel hyperball perceptron

Abstract In this paper the novel fuzzy kernel hyperball perceptron is presented. The proposed method first maps the input data into a high-dimensional feature space using some Mercer kernel function. Then, the decision function for each class is derived by the learning rules of the fuzzy kernel hyperball perceptron. The fuzzy membership functions, which resolve unclassifiable zones among classes, are incorporated into its classification algorithm to further enhance the perceptron’s adaptability and classification accuracy effectively. Unlike SVM, the fuzzy kernel hyperball perceptron has no convergence problem and avoids solving so-called quadratic programming problem which often makes SVM ineffective for large data sets. Especially, unlike the classical SVMs, we can directly utilize the fuzzy kernel hyperball perceptrons to solve multiclass problems, without using any pairwise combination. Our experimental results demonstrate its effectiveness.

Cascaded fuzzy system and its robust analysis based on syllogistic fuzzy reasoning

Syllogistic fuzzy reasoning is introduced into fuzzy system, and the new Cascaded Fuzzy System(CFS) is presented. The thoroughly theoretical analysis and experimental results show that syllogistic fuzzy reasoning is more robust than all other implication inferences for noise data and that CFS has better robustness than conventional fuzzy systems, which provide the solid foundation for CFS’s potential application in fuzzy control and modeling and so on.