Effectiveness of morphological reconstruction operators in change detection for remote sensing images. (2017)

• Record Type: Journal Article
• Authors: Khurana, Madhu; Saxena, Vikas
• Is Part Of: International journal of spatial, temporal and multimedia information systems. Volume 1:Number 2(2017)
• Page Start: 151
• Page End: 166
• Publication Date: 2017
• Subjects: change detection -- morphological reconstruction -- extreme learning machine -- ELM -- remote sensing -- texture features -- structuring element; Geographic information systems -- Periodicals; Temporal databases -- Periodicals; Multimedia systems -- Periodicals; 005.753
• Languages: English
• ISSNs: 2052-3556

Abstract

Change detection (CD) in multi-temporal images aims at quantifying the temporal effects or changes in remote sensing images taken at different times of the same area of Earth. With huge constellations of satellites setup by various countries, monitoring the Earth's surface for changes helps us understand and respond to various natural phenomenon affecting the Earth's atmosphere. In this paper, we have used texture features along with the spectral features for CD. The texture features have been extracted after applying three different morphological reconstruction operators, namely, opening by reconstruction, closing by reconstruction and opening of closing by reconstruction. Also flat and non-flat structuring elements (SE) have been used for applying morphological reconstruction. The paper compares the impact of different types of morphological operators on the accuracy of the CD. It also presents the comparison between the effectiveness of flat and non-flat SEs on the accuracy.