“A NOVEL SOFT CLASSIFICATION APPROACH TO DETECT OBJECT FOR MULTI SPECTRAL SATELLITE IMAGERY”
Dr. Ranjana Sharma 1, Dr. Gesu Thakur 2, Sarthika Dutt 3, Dr. Shambhu Bharadwaj 4,
Er. Madhulika Mittal 5, Priyanka Suyal 6
1Associate Professor, College of Engineering Roorkee (COER) Roorkee, Uttarakhand, India.
2Professor, University of Engineering and Technology, Roorkee, Uttarakhand, India.
3Assistant Professor, College of Engineering Roorkee (COER) Roorkee, Uttarakhand, India
4Associate Professor, CCSIT, Teerthanker Mahaveer University Moradabad, India.
5 Assistant Professor, Department of Computer Science and Engineering, Quantum University, Roorkee, Uttarakhand, India.
6Assistant Professor, College of Engineering Roorkee (COER) Roorkee, Uttarakhand, India.
firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com
Nowadays, modern earth observation programs produce massive volumes of satellite data time series (SITS) that can be beneficial to display geographical regions thru time. A way to effectively examine such kind of facts continues to be an open query with in the far flung sensing subject. To differing size, direction and background of target object, object detection very challenging in research area. Due Some of the leading GIS software which have well defined image processing module are ERDAS Imagine, IDRISI, ENVI, and ER Mapper but the assessment of accuracy is not support by these software for the evaluation of soft classified output. So, in this research article I am proposing two new algorithms which capable detection of area with high accuracy in comparison of two built algorithm.
Our contributions are 1) the assessment of accuracy percentage is equal to referential value be 2.385 mean less than 3 which indicate maximum accuracy.
2) The classification value is less than 1 in soft classifiers like FCM nd PCM and other hybridize classifiers (PCME and FCME).
3) I have trained model by different algorithm and testing of model using independent indicator “Entropy”
Abbreviation used in paper
Fuzzy c-mean Entropy
Possibilistic c-Means Entropy
Keywords: Pure pixel, Mixed pixel, Assessment of Accuracy, Entropy, Fuzzy c-mean Entropy, Possibilistic c-Means Entropy.