Science Links Japan | Classification of Remote Sensing Data Having Incomplete Supervised data Using a Causality Based on Markov Random Field.

Classification of Remote Sensing Data Having Incomplete Supervised data Using a Causality Based on Markov Random Field.

Accession number;99A0228774

Title;Classification of Remote Sensing Data Having Incomplete Supervised data Using a Causality Based on Markov Random Field.

Author; KAWAMURA MAKOTO (Toyohashi Univ. of Technol., Fac. of Tech.) TSUJIKO YUJI (Toyohashi Univ. of Technol., Fac. of Tech.) FUKABORI MASARU (Toyohashi Univ. of Technol., Grad. Sch.)

Journal Title;Proceedings of JSCE (Japan Society of Civil Engineers)

Journal Code:F0028A

ISSN:0289-7806

VOL.;NO.611;PAGE.1-11(1999)

Figure&Table&Reference;FIG.19, TBL.2, REF.17

Pub. Country;Japan

Language;Japanese

Abstract;A method for classification of remotely sensed data having incomplete supervised data is proposed. To consider the time difference between supervised data and remote sensing data, the causalities based on Markov Random Field are utilized. By including the temporal class dependencies, the reliable classification is undertaken. It expands to a method evaluating the spatial class dependencies between neighboring pixels and the post processing. The performance of the method for land cover classification is investigated using LANDSAT TM covering Aichi Prefecture, Japan, and compared with statistical data. The results show well coincidence with verification data. When post processing is carried out, the accuracy improves to RMSE of about 1%. (author abst.)

Related Articles;