Use of Bi-Temporal Satellite Data for Feature Extraction
- February 28, 2018
- Posted by: RSIS
- Categories: Electrical and Electronics Engineering, Electronics & Communication Engineering, Engineering, Image Processing
International Journal of Research and Innovation in Applied Science (IJRIAS) | Volume II, Issue V, May 2017 | ISSN 2454-6194
Use of Bi-Temporal Satellite Data for Feature Extraction
K. Devi Prasanna1, K. Bhuvaneswari2, G. Jai Sankar3
1, 3 Andhra University College of Engineering, Visakhapatnam, Andhra Pradesh, India
2DNR College of Engineering & Technology, Andhra Pradesh, India
Abstract— Feature Extraction is an important component of every Image Classification and Object Recognition System. An approach was developed in e-Cognition 9.0 software to perform image segmentation and classification for feature extraction. Classification algorithms formulated on pixel-based analysis often do not give the expected results when applied to high spatial resolution remote-sensing data. To overcome this problem, the concept of Object Based Image Analysis is developed as it gives the desired result when compared to pixel based analysis. The study area is a part of the IWMP watersheds which focuses on Gandarvakottai watershed in Pudukkottai District, Tamil Nadu. The study involves the extraction of watershed developmental activity works like check dams and farm ponds.
Keywords— Feature extraction, Watershed, Segmentation, OBIA
I. INTRODUCTION
A watershed is an area of land that captures rainfall and other precipitation and allows all water flowing it to a lake or stream or wetland. The use of water and its energy are very essential to the needs of modern society. When water is removed or flows are restricted, the function of the watershed is affected.
A. Feature Extraction
Mapping the image pixels into the feature space is known as feature extraction. For automatic identification of the objects from remote sensing data, they are to be associated with certain attributes which characterize them and differentiate them with each other. The various contents of an image such as color, texture, shape etc., are used to classify an image and identify the objects in Feature extraction.
B. Integrated Watershed Management Programme
The Integrated Watershed Management Programme (IWMP) initiated by the Department of Land Resources in 2009, has undertaken development and monitoring of watersheds across the country.
