Faculty Publications

Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736

Publications by NITK Faculty

Browse

Author: search.filters.author.Shetty, A.Subject: search.filters.subject.accuracy assessment

Search Results

Now showing 1 - 5 of 5
  • No Thumbnail Available
    Item
    EXhype: A tool for mineral classification using hyperspectral data
    (Elsevier B.V., 2017) Adep, R.N.; Shetty, A.; Ramesh, H.
    Various supervised classification algorithms have been developed to classify earth surface features using hyperspectral data. Each algorithm is modelled based on different human expertises. However, the performance of conventional algorithms is not satisfactory to map especially the minerals in view of their typical spectral responses. This study introduces a new expert system named ‘EXhype (Expert system for hyperspectral data classification)’ to map minerals. The system incorporates human expertise at several stages of it's implementation: (i) to deal with intra-class variation; (ii) to identify absorption features; (iii) to discriminate spectra by considering absorption features, non-absorption features and by full spectra comparison; and (iv) finally takes a decision based on learning and by emphasizing most important features. It is developed using a knowledge base consisting of an Optimal Spectral Library, Segmented Upper Hull method, Spectral Angle Mapper (SAM) and Artificial Neural Network. The performance of the EXhype is compared with a traditional, most commonly used SAM algorithm using Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) data acquired over Cuprite, Nevada, USA. A virtual verification method is used to collect samples information for accuracy assessment. Further, a modified accuracy assessment method is used to get a real users accuracies in cases where only limited or desired classes are considered for classification. With the modified accuracy assessment method, SAM and EXhype yields an overall accuracy of 60.35% and 90.75% and the kappa coefficient of 0.51 and 0.89 respectively. It was also found that the virtual verification method allows to use most desired stratified random sampling method and eliminates all the difficulties associated with it. The experimental results show that EXhype is not only producing better accuracy compared to traditional SAM but, can also rightly classify the minerals. It is proficient in avoiding misclassification between target classes when applied on minerals. © 2016 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS)
  • No Thumbnail Available
    Item
    Prediction accuracy of soil organic carbon from ground based visible near-infrared reflectance spectroscopy
    (Springer, 2018) Minu, S.; Shetty, A.
    The present study was conducted to predict soil organic carbon (SOC) from ground visible near-infrared (Vis-NIR, 400- 2500 nm) spectroradiometer reflectance spectra. The objective was to study the effect of various pre-processing methods and prediction models on the accuracy of SOC estimated. Measured SOC content and reflectance spectra from pasture and cotton fields of Narrabri, Australia were used in the analysis. Reflectance spectra were pretreated with different smoothing methods such as: moving average, median filtering, gaussian smoothing and Savitzky Golay smoothing. A comparison between principal component regression, partial least square regression (PLSR) and artificial neural network models was carried out to get an optimum model for organic carbon prediction. The results indicate that PLSR model performs better with Savitzky Golay as the best pre-processing method for the study area, yielding R2cal = 0:84, RPDcal = 2.55 and RPIQcal = 4.02 in the calibration set and R2val = 0:77, RPDval = 2.17 and RPIQval = 3.19 in the validation set. The study recommends a suitable method in case of limited number of soil data. Based on the study, it can be said that properly pretreated reflectance spectra show tremendous potential in soil organic carbon prediction. © Indian Society of Remote Sensing 2017.
  • No Thumbnail Available
    Item
    A Comparative Analysis of Forest Fire Risk Zone Mapping Methods with Expert Knowledge
    (Springer, 2019) Yathish, H.; Athira, K.V.; Konkathi, K.; Umesh, U.; Shetty, A.
    Despite repeated occurrences of forest fire, very less scientific studies have been reported in the Indian context especially in Kudremukh region to mitigate and suppress the fire. The objective of this article was to pool the expert knowledge on forest fire triggering factors from officials working in wildlife division in the Western Ghats of India through a questionnaire and to validate the risk zones obtained from three popular fire risk zone mapping methods namely logistic regression, multi-criteria decision analysis, and weighted overlay. Based on the earlier studies and expert knowledge, fire ignition parameters considered are elevation, slope, and aspect, proximity to roads, water bodies and area of human activities, normalized difference vegetation index (NDVI), land surface temperature (LST), and vegetation type. The regression model was based on previous fire occurrences and the other two based on expert’s opinion. The three models were validated and compared using past fire occurrence events. The logistic regression model gave 88.89% of accuracy and that of multi-criteria decision analysis with 74.6% accuracy, and that of weighted overlay method with an accuracy of 68.24% for the specific study area. The logistic regression model is useful in the presence of historical data, whereas expert knowledge is helpful for mapping risk zones using multi-criteria decision analysis and weighted overlay analysis when historical data are scarce or not available for mapping risk zones. The obtained risk maps can be used for deciding watchtower locations, installation of sensors, cameras, etc. In every forest division, it is recommended to prepare a standard questionnaire form and document their experiences on forest fire in the region under their supervision before they are getting transferred to another location. © 2019, Indian Society of Remote Sensing.
  • No Thumbnail Available
    Item
    Vertical accuracy assessment of open source digital elevation models under varying elevation and land cover in Western Ghats of India
    (Springer Science and Business Media Deutschland GmbH, 2022) Shetty, S.; Vaishnavi, P.C.; Umesh, P.; Shetty, A.
    The selection of suitable DEM from available open-source DEMs like SRTM, ALOS World 3D, CARTOSAT-1, ASTER-GDEM, TanDEM-X which are acquired through different techniques is difficult without prior guidelines, especially on the rugged mountainous terrain. Therefore, this article aimed to evaluate the role of land cover and altitude on the vertical accuracy of open-source DEMs with near to ground measurements taken by Ice Cloud and Land Elevation (ICESat) Geoscience Laser Altimetry System (GLAS) in and around Western Ghats (WG) of India. The SRTM (30 m) DEM outperformed other DEMs at the scale of WG and in the dense vegetation cover with least performance by ASTER DEM (30 m). The vertical accuracy of DEM is varying with different elevation ranges and land cover conditions and is found to be better than the vertical accuracy specified by the mission. The overestimation of elevation in low terrain relief area, and underestimation on higher elevation with steep terrain is substantive in all the DEMs. The role of land cover and altitude is significant on the elevation and slope more than the aspect and roughness. Good performance by 90-m resolution DEM over 30-m resolution DEMs proves the potential of InSAR in elevation measurement in vegetated areas with low cost and high accuracy. These results help in the selection of pertinent DEM for any geo-climatical applications and in development of merged DEM based on the terrain relief and land cover of the region. © 2021, The Author(s), under exclusive licence to Springer Nature Switzerland AG part of Springer Nature.
  • No Thumbnail Available
    Item
    Enhancing soil organic carbon estimation accuracy: Integrating spatial vegetation dynamics and temporal analysis with Sentinel 2 imagery
    (Elsevier B.V., 2024) Mruthyunjaya, P.; Shetty, A.; Umesh, P.
    This article introduces an improved method for estimating Soil Organic Carbon (SOC) using Sentinel 2 images, with a specific emphasis on the Dakshina Kannada area in India. By examining 364 soil samples, SOC estimation models were constructed using Random forests (RF) and Partial Least Squares Regression (PLSR), focusing on the impact of nearby vegetation pixels. The approach consisted of classifying soil samples by the presence of plant pixels at distances of 0, 10, and 20 m, and evaluating the influence of dry vegetation by the use of the Normalised Burn Ratio 2 (NBR2). The findings demonstrated a significant improvement in the precision of the model (by up to 20 %) when vegetation pixels within a 20-meter radius of the sample locations were omitted. The research also included a temporal analysis utilizing Sentinel-2 images from April 2017 to May 2023. This analysis showed strong relationships between the amount of exposed soil and the accuracy of predicting soil organic carbon (SOC) levels. These results emphasize the need to take into account both the spatial dynamics of vegetation and the temporal variations in bare soil covering to get an accurate estimate of soil organic carbon (SOC). This study improves the accuracy and dependability of SOC evaluations by including geographical and temporal aspects, providing useful insights for agricultural and ecological applications. © 2024 The Author(s)