Knowledge Discovery in an Object-Oriented Oceanographic Database System

Abstract : The rate at which scientific data is collected today has overwhelmed the ability of scientists to store and analyze the data. This report describes the results of a three year effort in the development of a knowledge discovery system for use by oceanographers at the Naval Oceanographic Office (NAVOCEANO) at the Stennis Space Center in the identification of provinces of interest in the ocean floor from acoustic imagery. The system is composed of a knowledge discovery component built to interact with a database system currently in use at Stennis Space Center. The knowledge discovery system applies machine learning techniques to features extracted from sonar images to identify provinces of the ocean floor based on visual texture. This requires that the images be segmented into regions of homogeneous texture using a region-growing technique, that features describing the texture of these regions be extracted, that machine learning techniques be applied to classify the regions, that classified images be constructed for visualizing the results, and that the classified images be combined and geo-referenced using a mosaic procedure. NAVOCEANO is currently supporting efforts to integrate the software developed from this project with their image analysis system.

[1]  Jorma Laaksonen,et al.  SOM_PAK: The Self-Organizing Map Program Package , 1996 .

[2]  David Haussler,et al.  KDD for Science Data Analysis: Issues and Examples , 1996, KDD.

[3]  Jianhua Xu.  Texture analysis approaches to extract ocean wavelengths, directions, and heights from ocean wave images , 1992 .

[4]  Salvatore J. Stolfo,et al.  Learning Arbiter and Combiner Trees from Partitioned Data for Scaling Machine Learning , 1995, KDD.

[5]  Anthony Skjellum,et al.  SCALABLE KNOWLEDGE DISCOVERY FROM OCEANOGRAPHIC DATA , 1998 .

[6]  Ronald J. Brachman,et al.  The Process of Knowledge Discovery in Databases , 1996, Advances in Knowledge Discovery and Data Mining.

[7]  Larry A. Rendell,et al.  The Feature Selection Problem: Traditional Methods and a New Algorithm , 1992, AAAI.

[8]  Jiri Matas,et al.  On Combining Classifiers , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  Salvatore J. Stolfo,et al.  Sharing Learned Models among Remote Database Partitions by Local Meta-Learning , 1996, KDD.

[10]  James Kelly,et al.  AutoClass: A Bayesian Classification System , 1993, ML.

[11]  R.M. Haralick,et al.  Statistical and structural approaches to texture , 1979, Proceedings of the IEEE.

[12]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[13]  Donald Karpovich.  Choosing the optimal features and texel sizes in image , 1998, ACM-SE 36.

[14]  Matthew Self,et al.  Bayesian Classification , 1988, AAAI.

[15]  Bruce Wooley,et al.  Region-growing techniques based on texture for provincing the ocean floor , 1998, ACM-SE 36.

[16]  Padhraic Smyth,et al.  From Data Mining to Knowledge Discovery: An Overview , 1996, Advances in Knowledge Discovery and Data Mining.

[17]  Steven J. Fenves,et al.  Applying AI clustering to engineering tasks , 1993, IEEE Expert.

[18]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[19]  Richard R. Muntz,et al.  Scalable Exploratory Data Mining of Distributed Geoscientific Data , 1996, KDD.

[20]  Philip S. Yu,et al.  Data Mining: An Overview from a Database Perspective , 1996, IEEE Trans. Knowl. Data Eng..

[21]  Peter C. Cheeseman,et al.  Bayesian Classification (AutoClass): Theory and Results , 1996, Advances in Knowledge Discovery and Data Mining.

[22]  T. Reed,et al.  Digital image processing techniques for enhancement and classification of SeaMARC II side scan sonar imagery , 1989 .

[23]  David Wai-Lok Cheung,et al.  Efficient Mining of Association Rules in Distributed Databases , 1996, IEEE Trans. Knowl. Data Eng..