Shu-Chuan Chu

Bio: Shu-Chuan Chu is an academic researcher from Shandong University of Science and Technology. The author has contributed to research in topics: Computer science & Wireless sensor network. The author has an hindex of 28, co-authored 231 publications receiving 3652 citations. Previous affiliations of Shu-Chuan Chu include University of South Australia & Sewanee: The University of the South.

An efficient K-medoids-based algorithm using previous medoid index, triangular inequality elimination criteria, and partial distance search

Abstract: Clustering in data mining is a discovery process that groups similar objects into the same cluster. Various clustering algorithms have been designed to fit various requirements and constraints of application. In this paper, we study several k-medoids-based algorithms including the PAM, CLARA and CLARANS algorithms. A novel and efficient approach is proposed to reduce the computational complexity of such k-medoids-based algorithms by using previous medoid index, triangular inequality elimination criteria and partial distance search. Experimental results based on elliptic, curve and Gauss-Markov databases demonstrate that the proposed algorithm applied to CLARANS may reduce the number of distance calculations by 67% to 92% while retaining the same average distance per object. In terms of the running time, the proposed algorithm may reduce computation time by 38% to 65% compared with the CLARANS algorithm.

Dimensionality reduction based on nonparametric discriminant analysis with kernels for feature extraction and recognition

Abstract: Dimensionality reduction is the most popular method for feature extraction and recognition Recently, Li et al (IEEE PAMI, 2009) proposed Nonparametric Discriminant Analysis (NDA) based dimensionality reduction for face recognition and reported an excellent recognition performance However, NDA has its limitations on extracting the nonlinear features of face images for recognition, and owing to the highly nonlinear and complex distribution of face images under a perceivable variation in viewpoint, illumination or facial expression In order to increase the NDA, we extend the NDA with kernel trick to propose Nonparametric Kernel Discriminant Analysis (NKDA) for feature extraction and recognition Experimental results on ORL, YALE and UMIST face databases show that NKDA outperforms NDA on recognition, which demonstrates that it is feasible to improve NDA with kernel trick for feature extraction

Vision Based Front and Rear Vehicle Collision Warning System

Abstract: In the current driving environment, the top priority is the safety of person. There are two methods proposed to solve safety problems. One is active sensors method and another is passive sensor method. Though with high accuracy, active sensors method has many disadvantages such as high cost, failure to adapt to complex change of environments, and problems relating to laws. Thus there is no way to popularize it. In contrast, passive sensor method is more suitable to current assist systems in virtue of low cost, ability to acquire lots of information. In this paper, the passive sensor method is applied to front and rear vision-based collision warning application. Meanwhile, time-to-contact is used to collision judgment analysis and dedicated short range communications is used to give alert information to near vehicle.

Image Recognition Based on Kernel Self-Optimized Learning

Abstract: Image recognition technologies have been used in many areas, and feature extraction of image is key step for image recognition. A novel feature extraction method using kernel self-optimized learning for image recognition. The scheme of image feature extraction includes textural extraction using Gabor wavelet, textural features reduction based on class-wise locality preserving projection with the nearest neighbor graph and common kernel discriminant vector. The nearest neighbor classifier is applied to image classification. The feasibility and performance of the algorithm are testified in the public image databases.

A comprehensive survey: artificial bee colony (ABC) algorithm and applications

Abstract: Swarm intelligence (SI) is briefly defined as the collective behaviour of decentralized and self-organized swarms. The well known examples for these swarms are bird flocks, fish schools and the colony of social insects such as termites, ants and bees. In 1990s, especially two approaches based on ant colony and on fish schooling/bird flocking introduced have highly attracted the interest of researchers. Although the self-organization features are required by SI are strongly and clearly seen in honey bee colonies, unfortunately the researchers have recently started to be interested in the behaviour of these swarm systems to describe new intelligent approaches, especially from the beginning of 2000s. During a decade, several algorithms have been developed depending on different intelligent behaviours of honey bee swarms. Among those, artificial bee colony (ABC) is the one which has been most widely studied on and applied to solve the real world problems, so far. Day by day the number of researchers being interested in ABC algorithm increases rapidly. This work presents a comprehensive survey of the advances with ABC and its applications. It is hoped that this survey would be very beneficial for the researchers studying on SI, particularly ABC algorithm.

Nature-Inspired Optimization Algorithms

Abstract: Nature-Inspired Optimization Algorithms provides a systematic introduction to all major nature-inspired algorithms for optimization. The book's unified approach, balancing algorithm introduction, theoretical background and practical implementation, complements extensive literature with well-chosen case studies to illustrate how these algorithms work. Topics include particle swarm optimization, ant and bee algorithms, simulated annealing, cuckoo search, firefly algorithm, bat algorithm, flower algorithm, harmony search, algorithm analysis, constraint handling, hybrid methods, parameter tuning and control, as well as multi-objective optimization. This book can serve as an introductory book for graduates, doctoral students and lecturers in computer science, engineering and natural sciences. It can also serve a source of inspiration for new applications. Researchers and engineers as well as experienced experts will also find it a handy reference.Discusses and summarizes the latest developments in nature-inspired algorithms with comprehensive, timely literatureProvides a theoretical understanding as well as practical implementation hintsProvides a step-by-step introduction to each algorithm

Maximum neighborhood margin discriminant projection for classification

Abstract: We develop a novel maximum neighborhood margin discriminant projection (MNMDP) technique for dimensionality reduction of high-dimensional data. It utilizes both the local information and class information to model the intraclass and interclass neighborhood scatters. By maximizing the margin between intraclass and interclass neighborhoods of all points, MNMDP cannot only detect the true intrinsic manifold structure of the data but also strengthen the pattern discrimination among different classes. To verify the classification performance of the proposed MNMDP, it is applied to the PolyU HRF and FKP databases, the AR face database, and the UCI Musk database, in comparison with the competing methods such as PCA and LDA. The experimental results demonstrate the effectiveness of our MNMDP in pattern classification.