Wu Jian Kang
Email Address
Organizational Units
18 results
Publication Search Results
Now showing 1 - 10 of 18
Publication Recognition by recall(1996) Wu, J.K.; INSTITUTE OF SYSTEMS SCIENCERecognition is a major capability of human being and animals. In the discipline of pattern recognition, research has been carried out extensively on recognition by classification. As a matter of fact, most of the time, human perform recognition tasks by recall. This paper describes the framework of recognition by recall. The definitions are given from the perspective of content-based retrieval. The approach and preliminary results are also presented. © 1996 IEEE.Publication Identifying faces using multiple retrievals(1994-06) Wu, Jian Kang; Narasimhalu, Arcot Desai; INSTITUTE OF SYSTEMS SCIENCETo facilitate crime investigation, a team at the Institute of Systems science designed Computer-Aided Facial Image Inference and Retrieval System (CAFIIR) which was a database management system was incorporated to mug shot identification. The system was based on the expert feedbacks of crime investigators. It encompassed physical traits as well as subjective descriptions that require fuzzy searching. For fuzzy subjects, a mapping function was employed. The CAFIIR system also provided for any inaccuracy or shortcomings of the descriptions given by the witness and investigators, rendering a practically foolproof program.Publication Analysis of singapore marine sediments by PIXE(1998-03) Tang, S.M.; Orlic, I.; Wu, X.K.; INSTITUTE OF SYSTEMS SCIENCE; PHYSICSThirty eight core samples of sediment were collected from seven coastal zones around the Island of Singapore for the purpose of surveying the concentrations of metallic pollutants in the marine sediments. The samples were analyzed by means of the Proton-Induced X-ray Emission (PIXE) technique to obtain the concentrations of the trace metals Cr, Mn, Ni, Cu, Zn, As and Pb. Some of them were also analyzed with the X-ray Fluorescence (XRF) technique to determine the concentration of Sn. The survey provided valuable information about the levels of marine pollution in the various coastal zones and shed light on the major source of some of the pollutants. Interesting concentration depth profiles of the metallic pollutants in sediments were also obtained from a number of the core samples collected. The concentration depth profiles of Cu, Zn, Sn and Pb obtained from one of the core samples exhibited a steep decrease at a depth of 10 cm, implying that a sudden increase of these pollutants occurred some 20-33 years ago. © 1998 Published by Elsevier Science B.V.Publication Towards a semantic image database system(1997-04) Yang, L.; Wu, J.; NATIONAL SUPERCOMPUTING RESEARCH CENTRE; INSTITUTE OF SYSTEMS SCIENCEA semantic image database system (OISDBS) has been designed for structural management of image data. A semantic image data model has been presented to describe the inner structure and contents of images. A diagrammatic query language with a QBE (Query-by-example) flavor is proposed together with a discussion of its foundation on object rewriting. This data model describes the structure and contents of images by incorporating type constructors, functions on types, and inheritance. It supports composite modeling for entities which consist of pictorial data as well as alphanumeric attributes. The interactive query language allows users to do direct manipulation on the pictorial database because such diagrammatic query formulations represent the users' view of the database schema. Various aspects of OISDBS such as the schema design, storage management, and mapping from semantic schema to relational schema, are also discussed.Publication Recognition by recall - a new paradigm for object recognition(1997) Wu, Jian Kang; INSTITUTE OF SYSTEMS SCIENCERecognition is a major capability of human being and animals. In the discipline of pattern recognition, research has been carried out extensively on recognition by classification. As a matter of fact, most of the time, human performs recognition tasks by recall. This paper describes the framework of recognition by recall. The definitions and approaches are described, preliminary results presented.Publication Content-based retrieval for trademark registration(1996) Wu, J.K.; INSTITUTE OF SYSTEMS SCIENCEWith ever increasing number of registered trademarks, the task of trademark office is becoming increasingly difficult to ensure the uniqueness of all trademarks registered. Trademarks are complex patterns consisting of various image and text patterns, called device-mark and word-in-mark respectively. Due to the diversity and complexity of image patterns occurring in trademarks, due to multi-lingual word-in-mark, there is no very successful computerized operating trademark registration system. We have tackled key technical issues: multiple feature extraction methods to capture the shape, similarity of multi-lingual word-in-mark, matching device mark interpretation using fuzzy thesaurus, and fusion of multiple feature measures for conflict trademark retrieval. A prototype System for Trademark Archival and Registration (STAR) has been developed. The initial test run has been conducted using 3000 trademarks, and the results have shown satisfaction to trademark officers and specialists. © 1996 Kluwer Academic Publishers.Publication Inference and retrieval of facial images(1994-06) Wu, J.K.; Ang, Y.H.; Lam, P.; Loh, H.H.; Narasimhalu, A.D.; INSTITUTE OF SYSTEMS SCIENCEAttempts have been made to extend SQL to work with multimedia databases. We are reserved on the representation ability of extended SQL to cope with the richness in content of multimedia data. In this paper we present an example of a multimedia database system, Computer Aided Facial Image Inference and Retrieval system (CAFIIR). The system stores and manages facial images and criminal records, providing necessary functions for crime identification. We would like to demonstrate some core techniques for multimedia database with CAFIIR system. Firstly, CAFIIR is a integrated system. Besides database management, there are image analysis, image composition, image aging, and report generation subsystems, providing means for problem solving. Secondly, the richness of multimedia data urges feature-based database for their management. CAFIIR is feature-based. A indexing mechanism, iconic index, has been proposed for indexing facial images using hierarchical self-organization neural network. The indexing method operates on complex feature measures and provides means for visual navigation. Thirdly, special retrieval methods for facial images have been developed, including visual browsing, similarity retrieval, free text retrieval and fuzzy retrieval. © 1994 Springer-Verlag.Publication Fuzzy content-based retrieval in image databases(1998-09) Wu, J.K.; Narasimhalu, A.D.; INSTITUTE OF SYSTEMS SCIENCEImage data are inherently visual. The description of visual characteristics of images are imprecise. Fuzzy retrieval of images stored in a feature-based image database is a natural means to access the data. Unfortunately, to the authors knowledge, little work has been done on fuzzy image database models and fuzzy retrieval of feature-based image databases. In this paper, a fuzzy image database model and a concept of fuzzy space are proposed and fuzzy query processing in fuzzy space and fuzzy indexing on complex fuzzy vectors are described. An example image database, the computer-aided facial image inference and retrieval system (CAFIIR), is used for explanation throughout the paper. © 1998 Published by Elsevier Science Ltd. All rights reserved.Publication Benchmarking Multimedia Databases(1997) Narasimhalu, A.D.; Kankanhalli, M.S.; Wu, J.; INSTITUTE OF SYSTEMS SCIENCEMultimedia technologies are being adopted both in the professional and commercial world with great enthusiasm. This has led to a significant interest in the research and development of multimedia databases. However, none of these efforts have really addressed the issues related to the benchmarking of multimedia databases. We analyze the problem of benchmarking multimedia databases in this paper and suggest a methodology.Publication Remote sensed image classification using multi-perspective neural networks(1993) Wu, Jian Kang; Takagi, M.; INSTITUTE OF SYSTEMS SCIENCERemotely sensed imagery classification is widely used for earth resource inventory. Due to variations of imaging condition the signature on images and the objects on the land have no unique correspondence. This results in great difficulties to the computer processing of remotely sensed imagery. Here in this paper, we describe a novel neural network model LEP (Learning based on Experiences and Perspectives), and its application to remote sensed image classification. Because the network properly makes use of multi-perspective data and its learning is finely tuned by experiences, the classification results have been much improved.