Indian Institute of Remote Sensing

About: Indian Institute of Remote Sensing is a government organization based out in Dehra Dūn, India. It is known for research contribution in the topics: Land cover & Normalized Difference Vegetation Index. The organization has 756 authors who have published 1355 publications receiving 16915 citations. The organization is also known as: Indian Photo-interpretation Institute.

Discrimination of maize crop in a mixed Kharif crop scenario with synergism of multiparametric SAR and optical data

Abstract: Accurate mapping of coarse cereals envisaged maize to be discriminated from co-existing Kharif crops using multiplatform, multiparametric SAR and Optical data in various polarization combination fr...

A Spectro-radiometric Analysis of Ocean Colour Sensors and Proposal for a Miniature Hyper Spectral Imager for Future

Abstract: Ocean colour sensors traditionally are of fixed spectral channel systems with specified bandwidth of about 20 nm in the visible region and about 40 nm in Near Infrared region. In these systems, it is known that a radiometric error of 1% in the measurement of top of the atmosphere signal may lead to an error of 10% in the retrieved ocean upwelling radiance. In this paper we investigated the range of wavelengths participating in signal collection (effective spectral pass band, ESPB) using relative spectral response data of various sensors flown earlier. ESPB values were computed for each spectral channel for various percentages of signal and the results showed that they are quite high compared to bandwidths specified. These values were found to vary with sensor and channel. ESPB shall be small for accurate computation of spectral radiance. As the knowledge of spectral profile of the signal in the range of ESPB helps in better estimation of spectral radiance at the intended wavelengths, a miniature high performance linear variable filter based hyperspectral sensor is proposed as an alternative. We present here the design concept and report the estimated performance of such sensor that can be realized even with commercial off the shelf components for operational implementation.

Earthquake damage identification: a case study using soft classification approach

Abstract: The October 8, 2005, Kashmir earthquake (M w 7.6) affected the rough mountainous regions of India and Pakistan with poor accessibility, and thus, no proper comprehensive ground survey was possible. However, due to the ability of remote sensing technology to acquire spectral measurements of damaged areas at various spatial and temporal scales, extraction of damaged areas can be carried out quickly and with great reliability. The fuzzy-based classifiers [Possibilistic c-Means (PCM), noise cluster (NC), and NC with entropy (NCE)] were applied to identify 2005 Kashmir earthquake, induced landslides, as well as built-up damage (BD) areas, as soft computing approaches using supervised classification. Results indicate that for the identification of landslides and BD areas, NCE classifier generated the best outputs, while for the identification of built-up undamaged areas, NC classifier generated the best output. Further, it was found that the proposed Class Based Sensor Independent (CBSI) technique can improve spectral information of specific class for better identification.

Utilizing Geo-Social Media as a Proxy Data for Enhanced Flood Monitoring

Abstract: Social media plays an important role in disseminating spontaneous information during natural disasters/emergencies. It is a crowdsourcing platform, capable of complementing and supplementing remote sensing data in disaster mapping. Continuous monitoring of disasters such as floods from pre-stage to post stage is essential and geo-social media can attain it. The present research proposes a robust architecture to develop a social media-based near-real time flood monitoring system. In addition, our research article also emphasizes the efficient methods to process, analyse and explore multiple data dimensions of social media. A prototype model was prepared and tested on the tweets of Chennai floods 2015 to demonstrate social media potential in disaster monitoring. We implemented Natural Language Processing and Supervised Machine Learning in data processing and analysis segments of the framework by assembling various open source python libraries to develop the prototype. Initially, we built the required tweet corpus and performed pre-processing steps on it. Later, the collected tweets were geocoded with the place names available in the tweets and classified them into various flood topic related classes using Naive Bayes classifier. Subsequently, the tweets showcasing the flood condition were determined to generate a point map and a point density map to identify the flood hotspots. We verified our results with openly available 2015 Chennai flood map that is generated using remote sensing images and found positive outcome. A prototype web portal was developed to publish the results from above model as web maps. Furthermore, the portal would prove to be useful as a source for disseminating information to the public. The results prove that the proposed framework is evidently supportive in establishing a near real time monitoring system during emergencies.

Remote-sensing the urban area: Automatic building extraction based on multiresolution segmentation and classification

Abstract: Multiresolution segmentation is a new procedure for image object extraction. It allows the segmentation of an image into a network of homogeneous image regions at any chosen resolution. These image objects represent image information in an abstracted form and serve as building blocks for subsequent classifications. An exercise was undertaken to perform and study an object oriented segmentation and classification using high resolution satellite data (Cartosat-1 fused with IRS-1C, LISS IV data) for automatic building extraction in India. The study area covered the administrative area of BHEL (Bharat Heavy Electrical Limited) colony of Haridwar, Uttrakhand ( 29°56'55.51"N to 29°56'11.49"N latitude and 78°05'42.45"E to 78°07'00.9"E longitude). Two approaches were used for feature extraction, namely, applying different spatial filers, and object- oriented fuzzy classification. The merged image was filtered using the different high pass filters like Kirsch, Laplace, Prewitt, Sobel, Canny filtered images. The results showed that the overall accuracy of classified image was 0.93 and the Kappa accuracy was 0.89. The producer accuracy for building, vegetation and shadow were 0.9545, 1.0 and 0.8888 respectively whereas user accuracy for building, vegetation and shadow are 1.0, 0.9375 and 1.0 respectively. The overall classification accuracy based on TTA mask (training and test area mask) was 0.97 while the Kappa accuracy was 0.95. The producer accuracy for building, forest and shadow were 1.0, 1.0 and 0.7144 respectively and the user accuracy for building, vegetation and shadow were 1.0, 0.9375 and 1.0 respectively.