HR: 1340h
AN: IN23A-0948    [Abstracts]
TI: Evaluating Object-Based Image Analysis on Glacial Micromorphology
AU: * Chin, K S
EM: ksc15@sfu.ca
AF: Krystal S. Chin, University of Calgary Department of Geography 2500 University Drive, NW, Calgary, AB T2N 1N4, Canada
AU: Sjogren, D B
EM: sjogren@ucalgary.ca
AB: Micromorphology has recently been applied more in analyzing glacial sediments at a microscopic level. It provides additional information and details that may help to explain glacial processes in areas where macro- scale observations cannot yield sufficient information. However, the process of interpreting thin sections has been very subjective, and reaching general consensus about glacial processes is difficult. Remote sensing technology is increasingly helpful in the development and advancement of many sciences; the concepts that lie behind the technology in object cognition used in other fields, such as landscape ecology, can be applied to micromorphology. Similar to what has been done to landscape ecology in the past, automating the process of interpreting objects in glacial sediments may potentially simplify and decrease the subjectivity of the process. Definiens Professional 5 is an object-based image analysis program that imitates human cognitive methods; it is used in this study to identify objects apart from background matrices in multiple thin section images of glacial sediments. The program's initial results proved that more work was needed to be done for better results, but overall the software produced promising results. The method is repeatable and continues to generate consistent results with no bias or ambiguity, so the application of this method to micromorphology and other areas alike will be valuable.
DE: 5112 Microstructure
DE: 5194 Instruments and techniques
DE: 8030 Microstructures
DE: 8040 Remote sensing
DE: 9820 Techniques applicable in three or more fields
SC: Earth and Space Science Informatics [IN]
MN: 2007 Fall Meeting