Classification of polar satellite data using minimum distance method

• Main Authors: Ken'ichiro Muramoto, Mamoru Kubo, Hideo Saito, Takashi Yamanouchi
• Format: Report
• Language: English
• Published: Faculty of Engineering, Kanazawa University/Toyama National College of Technology/Faculty of Engineering, Kanazawa University/National Institute of Polar Research 1998
• Subjects: Antarctic; Antarc*; Antarctica; Memoirs of National Institute of Polar Research; Polar Research; Sea ice
• Online Access: https://nipr.repo.nii.ac.jp/?action=repository_uri&item_id=2320; http://id.nii.ac.jp/1291/00002320/; https://nipr.repo.nii.ac.jp/?action=repository_action_common_download&item_id=2320&item_no=1&attribute_id=18&file_no=1

Detection of Antarctic clouds is important because of their strong radiation influence on energy balance of the snow and ice surface. In this paper, a method to classify cloud, sea ice and ground is proposed. This study is based upon analysis of the NOAA/AVHRR infrared images in Antarctica. The algorithm consists of two major approaches : extraction of image features and a classification algorithm. A minimum distance classifier was used to classify that region into one of three categories using five image features. To reduce the error rate of the classification, threshold boundaries for minimum distance classifiers have been changed. Both classified and misclassified areas were decreased with increased threshold levels.