Novel Algorithm for RS Image Classification
- PDF / 96,325 Bytes
- 3 Pages / 595.276 x 790.866 pts Page_size
- 30 Downloads / 209 Views
SHORT NOTE
Novel Algorithm for RS Image Classification Pravada S. Bharatkar & Rahila Patel & Dilip G. Durbude
Received: 3 February 2013 / Accepted: 14 October 2013 # Indian Society of Remote Sensing 2013
Abstract In the present era of modern technology, the efficacy and accuracy of output is demanding. Based on rigorous survey, Bharatkar and Patel (International Journal of Advanced Research in Computer Science 3(7):218–223, 2012) concluded that the incorporation of Block Truncation Coding (BTC) approach in the existing image classification algorithm can be used to improve the classification accuracy. Therefore, in the present study, an effort has been made to explore the Content Based Remote Sensing Image (CBRSI) classification algorithm to enhance classification accuracy with BTC approach. It is revealed from the study that BTC based maximum likelihood classifier gives better overall accuracy and kappa statistics. Keywords Remote sensing . Classification algorithm . Satellite image . BTC . etc. Classifying remotely sensed image into a thematic map remains a challenge because many factors, such as the complexity of the landscape in a study area, selected
P. S. Bharatkar (*) : R. Patel Department of Computer Science, RCERT (Nagpur University), Chandrapur, India e-mail: [email protected] R. Patel e-mail: [email protected] D. G. Durbude Environmental Hydrology Division, National Institute of Hydrology, Roorkee, India e-mail: [email protected]
remotely sensed data, and image-processing and classification approaches. Content of image such as color, texture and shape and size plays an important role in semantic image classification. Many researchers apply various classification approaches/method for image classification. Each of these has some specific problems, which generate the scope for further improvement in the existing algorithm (Lu and Weng 2007). Hence, developing computationally efficient algorithms for image classification without compromising the classification accuracy is of primary importance. Several researchers have made efforts to develop the classification algorithms by improving the classification accuracy. Among all, Kekre et al. (2012) made an effort on Block Truncation Coding (BTC) approach for classifying the photographic image classification of human being, animals, and natural scenery and found a better classification performance. Silvia et al. (2011) also used BTC approach and found that the performance of BTC approach as superior. Of late, Bharatkar and Patel (2012) pointed out the possibility of this BTC approach for RS image classification. As such, no study has been conducted so far to classify RS image based on BTC approach. Hence, there is scope to develop the novel Content Based Remote Sensing Image (CBRSI) classification algorithm by incorporating the BTC approach in the existing RS image classification algorithm. Basically, BTC is a simple image coding technique developed in the early years of digital imaging more than 30 years ago. Although, it is simple technique, it h
Data Loading...
