Yau-Hwang Kuo

Weighted fuzzy mean filters for image processing

Copyright (c) 1997 Elsevier Science B.V. All rights reserved. A new fuzzy filter for the removal of heavy additive impulse noise, called the weighted fuzzy mean (WFM) filter, is proposed and analyzed in this paper. The WFM-filtered output signal is the mean value of the corrupted signals in a sample matrix, and these signals are weighted by a membership grade of an associated fuzzy set stored in a knowledge base. The knowledge base contains a number of fuzzy sets decided by experts or derived from the histogram of a reference image. When noise probability exceeds 0.3, WFM gives very superior performance compared with conventional filters when evaluated by mean absolute error (MAE), mean square error (MSE), peak signal-to-noise-rate (PSNR) and subjective evaluation criteria. For dedicated hardware implementation, WFM is also much simpler than the conventional median filter.

Automated ontology construction for unstructured text documents

Ontology is playing an increasingly important role in knowledge management and the Semantic Web. This study presents a novel episode-based ontology construction mechanism to extract domain ontology from unstructured text documents. Additionally, fuzzy numbers for conceptual similarity computing are presented for concept clustering and taxonomic relation definitions. Moreover, concept attributes and operations can be extracted from episodes to construct a domain ontology, while non-taxonomic relations can be generated from episodes. The fuzzy inference mechanism is also applied to obtain new instances for ontology learning. Experimental results show that the proposed approach can effectively construct a Chinese domain ontology from unstructured text documents.

The new intrusion prevention and detection approaches for clustering-based sensor networks [wireless sensor networks]

In this paper, we propose two approaches to improve the security of clustering-based sensor networks: authentication-based intrusion prevention and energy-saving intrusion detection. In the first approach, different authentication mechanisms are adopted for two common packet categories in generic sensor networks to save the energy of each node. In the second approach, different monitoring mechanisms are also needed to monitor cluster-heads (CHs) and member nodes according to their importance. When monitoring CHs, member nodes of a CH take turns to monitor this CH. This mechanism reduces the monitor time, and therefore saves the energy of the member nodes. When monitoring member nodes, CHs have the authority to detect and revoke the malicious member nodes. This also saves the node energy because of using CHs to monitor member nodes instead of using all the member nodes to monitor each other. Finally, simulations are performed and compared with LEACH, based on an ns2 LEACH CAD tool. The simulation result shows that the proposed approaches obviously extend the network lifetime when the clustering-based sensor network is under attack.

eHIP: An energy-efficient hybrid intrusion prohibition system for cluster-based wireless sensor networks

Verifying authenticity and integrity of delivered data is indispensable for security-sensitive wireless sensor networks (WSN). Unfortunately, conventional security approaches are unsuitable for WSN because energy efficiency is really not an important issue. However, energy conservation is truly a critical issue in WSN. In this paper, a proposed hybrid security system, called energy-efficient hybrid intrusion prohibition (eHIP) system, combines intrusion prevention with intrusion detection to provide an energy-efficient and secure cluster-based WSN (CWSN). The eHIP system consists of authentication-based intrusion prevention (AIP) subsystem and collaboration-based intrusion detection (CID) subsystem. Both subsystems provide heterogeneous mechanisms for different demands of security levels in CWSN to improve energy efficiency. In AIP, two distinct authentication mechanisms are introduced to verify control messages and sensed data to prevent external attacks. These two authentication mechanisms are customized according to the relative importance of information contained in control messages and sensed data. However, because the security threat from compromised sensor nodes cannot be fully avoided by AIP, CID is therefore proposed. In CID, the concept of collaborative monitoring is proposed to balance the tradeoff between network security and energy efficiency. In order to evaluate the performance of eHIP, theoretical analyses and simulations of AIP and CID are also presented in this paper. Simulation results fully support the theoretical analysis of eHIP.

A new fuzzy edge detection method for image enhancement

A new method for detecting object edges, called the fuzzy Sobel method, is proposed. The fuzzy logic methodology is applied to extract the feature value for an image and a Sugeno-type fuzzy reasoning strategy is adopted for edge enhancement. Such a method can improve the drawbacks of conventional approaches, for instance, the Prewitt and Sobel methods. In the two traditional methods, they use fixed parameters for all kinds of images so that they might have success in one image but fail in another one. The fuzzy Sobel method, on the other hand, can find four threshold values automatically and then use them to construct membership functions for each detected image. Therefore, the fuzzy Sobel method can determine the parameters adaptively for different images. The edges detected in the Sobel method are often vague, but the fuzzy Sobel method can enhance edges and makes the resultant image clear.

Adaptive Fuzzy Filter and Its Application to Image Enhancement

This chapter describes the design and evaluation of a novel adaptive fuzzy filter, and discusses its application to image enhancement. Most traditional edge detectors can perform well for uncorrupted images but are highly sensitive to impulse noise, so they can not work efficiently for blurred images. The proposed adaptive fuzzy filter consists of two major mechanisms: Adaptive Weighted Fuzzy Mean (AWFM) filter and Fuzzy Normed Inference System (FNIS) to realize the function of edge detection for smeared images. The membership functions of all fuzzy sets used in this filter can be adaptively determined for different images. Moreover, the adaptive fuzzy filter is capable of converting blurred edges to clear ones and suppressing noise at the same time. According to the experimental results, it works well in full range of random impulse noise probability and performs efficiently in the environment of mixed Gaussian impulse noise. This chapter also analytically evaluates the important properties of the filter to show its high performance in general cases.

Fairness and QoS guarantees of WiMAX OFDMA scheduling with fuzzy controls

A fairness and QoS guaranteed scheduling approach with fuzzy controls (FQFCs) is proposed for WiMAX OFDMA systems. The controllers, respectively, adjust priority and transmission opportunity (TXOP) for each WiMAX connection according to QoS requirements and service classes. The FQFC provides intra- and interclass fairness guarantees by making connections within the same class achieve equal degree of QoS while at the same time making those without QoS requirements equally share the remaining resources. Even in dynamic environments such as mobile WiMAX networks with time-variant traffic specifications, the FQFC fairly guarantees delay, throughput, and jitter, which are seldom achieved at the same time by state-of-the-art solutions.

Hardware implementation of CMAC neural network with reduced storage requirement

The cerebellar model articulation controller (CMAC) neural network has the advantages of fast convergence speed and low computation complexity. However, it suffers from a low storage space utilization rate on weight memory. In this paper, we propose a direct weight address mapping approach, which can reduce the required weight memory size with a utilization rate near 100%. Based on such an address mapping approach, we developed a pipeline architecture to efficiently perform the addressing operations. The proposed direct weight address mapping approach also speeds up the computation for the generation of weight addresses. Besides, a CMAC hardware prototype used for color calibration has been implemented to confirm the proposed approach and architecture.

A fuzzy CMAC model for color reproduction

Abstract Color reproduction is a complex nonlinear mapping problem due to gamut mismatch, resolution conversion, quantization, nonlinear color relationship between scanner and printer. To solve such a complex problem in color reproduction, this paper proposes a fuzzy CMAC model, which adopts a special parallel fuzzy inference-like process to realize the function similar to higher-order CMAC. In this model, recursive B-spline receptive field functions are replaced by fuzzy sets with bell-shaped membership function, and the weights to evaluate output values are also not crisp values but fuzzy sets. The learning algorithm is based on the maximum gradient method. For the situations of insufficiently or irregularly distributed training patterns, this paper develops a sampling method to generate uniformly distributed training patterns. According to experimental results, the proposed fuzzy CMAC model has shown its effectiveness on color reproduction and general function approximations. Besides, it has advantages of fast learning speed, simple computation, and high stability on model parameters.

Cross-Lingual Document Representation and Semantic Similarity Measure: A Fuzzy Set and Rough Set Based Approach

As cross-lingual information retrieval is attracting increasing attention, tools that measure cross-lingual semantic similarity between documents are becoming desirable. In this paper, two aspects of cross-lingual semantic document similarity measures are investigated: One is document representation, and the other is the formulation of similarity measures. Fuzzy set and rough set theories are applied to capture the inherently fuzzy relationships among concepts expressed by natural languages. Our approach first develops a language-independent sense-level document representation based on the fuzzy set model to reduce the barrier between different languages and further explores the fuzzy-rough hybrid approach to obtain a more robust macrosense-level document representation through the partitioning of the integrated sense association network of the document collection into macrosenses. Then, Tverskys notion of similarity and the F1 measure on information retrieval are adopted to formulate, respectively, two document similarity measures with fuzzy set operations on the two proposed document representations. The effectiveness of our approach is demonstrated by its success rate in identifying the English translations to their corresponding Chinese documents in a collection of Chinese-English parallel documents. Moreover, the proposed approach can be easily extended to process documents in other languages. It is believed that the proposed representations, along with the similarity measures, will enable more effective text mining processes.