Scribd Logo
Upload

Welcome to Scribd!

  • 4 views

    For PES Internship

    Uploaded by

    Nisha E

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd

    For PES Internship

    Uploaded by

    Nisha E
    4 views14 pages

    Original Title

    Copy of PPT for PES Internship

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    4 views14 pages

    For PES Internship

    Uploaded by

    Nisha E

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    of 14

    RAPID Internship 2024

    Habitat Suitability and Change Detection for One Horned Rhinos in


    Kaziranga National Park

    Name: Mrunal Kulkarn


    Mullapudi Jahnavi
    Nisha E

    Department: CSE

    Campus: RR Campus
    Background of the research problem highlighting the
    knowledge gap

    Research Problem:

    ● Assessing the impact of LULC changes on rhino habitats.


    ● Predicting future land cover changes and their implications for rhino conservation.
    ● The Indian Rhino (Rhinoceros unicornis) is a vulnerable species with a significant population in
    Assam, India, particularly in Golaghat and Nagaon districts.
    ● Rapid land use and land cover (LULC) changes due to agricultural expansion, urbanization, and
    deforestation pose a threat to rhino habitats.

    Importance:

    ● Ensuring the survival of rhinos by identifying suitable habitats.


    ● Informing conservation strategies and land management policies.
    Knowledge Gap:

    Current Knowledge: Knowledge Gap:


    ● Existing studies have documented the decline ● Insufficient understanding of the impact of future
    in suitable rhino habitats due to human LULC changes on rhino habitats.
    activities. ● Limited integration of predictive modeling
    ● Conservation efforts have focused on techniques in habitat suitability assessments.
    protected areas but often lack data-driven ● Need for comprehensive models that combine
    insights on future habitat suitability. habitat suitability with future land cover
    simulation.

    Research Aim: Significance:


    ● To bridge the gap by developing predictive ● Providing actionable insights for conservation
    models for LULC changes. planning.
    ● To assess their impact on rhino habitats and ● Enhancing the effectiveness of habitat
    population dynamics in Golaghat and Nagaon management and policy-making.
    districts.
    Problem Statement and Objective of research

    PROBLEM STATEMENT
    By utilizing machine learning and computer vision to analyze satellite images, climate data,
    and ecological surveys, we aim to identify and predict habitat condition changes. This will
    provide timely and accurate information to support wildlife conservation efforts. The model
    will be data-driven, incorporating rhino population data and historical satellite imagery to
    enhance predictive analysis.

    OBJECTIVE
    The objective of this research is to develop AI-based methods to detect changes in wildlife
    habitat suitability, specifically for the Indian One-Horned Rhinos in Kaziranga National Park,
    Assam, India.
    Experimental methodology/ Algorithm used

    1. Satellite Imagery 1. Combining Features


    2. Rhino Population Data 2. Preparing Target Variable: 1. Decoder Model:

    Data Data Decoding


    Collection Preparation Features

    Feature Model
    Extraction Training

    1. EfficientNet Model 1. GRU Model


    2. Saving Features 2. RandomForestRegressor:
    Experimental methodology/ Algorithm used

    1. Data Collection: 2. Feature Extraction: 3. Data Preparation: 4. Model Training: 5. Decoding Features:

    ● Satellite Imagery: ● EfficientNet Model: ● Combining ● GRU Model: ● Decoder Model:


    ○ Sources: Google ○ Load from Features: ○ Train on ○ Train on
    Earth Engine, TensorFlow ○ Combine yearly sequences of historical
    other remote Hub. feature vectors. feature vectors. LULC data.
    ○ Extract feature ○ Create lagged ○ Predict future ○ Decode feature
    sensing
    vectors for each features and LULC features. vectors into
    databases.
    image. moving average ● Random Forest land cover
    ○ Years: 2012-2022
    ● Rhino Population features. Regressor: proportions.
    for both districts.
    Data: ● Preparing Target ○ Train on
    ● Rhino Population ○ Save extracted Variable: combined
    Data: feature vectors ○ Rhino features and
    ○ Sources: Wildlife for each year and population data rhino
    conservation district. for population data.
    reports, census corresponding ○ Predict future
    data. years. rhino
    ○ Years: 2012- population.
    2022.
    Working prototype of the
    problem
    Experimental Results (Predicted Land Cover)
    GOLAGHAT:

    NAGAON:
    Experimental Results (Predicted Rhino
    Population)

    R2 score
    Random Forest Regressor: 0.94
    GRU :
    Draft paper
    Publications
    References

    [1] T. Mukherjee, L. K. Sharma, G. K. Saha, M. Thakur, and K. Chandra, "Past, present and future: combining habitat suitability
    and future landcover simulation for long-term conservation management of Indian rhino," *Scientific Reports*, vol. 10, no. 1, p.
    606, 2020.

    [2] N. Nath, D. Sahariah, G. Meraj, J. Debnath, P. Kumar, D. Lahon, K. Chand et al., "Land use and land cover change monitoring
    and prediction of a UNESCO world heritage site: Kaziranga eco-sensitive zone using cellular automata-Markov model," *Land*,
    vol. 12, no. 1, p. 151, 2023.

    [3] K. Sarma, M. K. Saikia, B. Sarania, H. Basumatary, S. S. Baruah, B. P. Saikia, A. Kumar, and P. K. Saikia, "Habitat monitoring
    and conservation prioritisation of Western Hoolock Gibbon in upper Brahmaputra Valley, Assam, India," *Scientific Reports*, vol.
    11, no. 1, p. 15427, 2021.

    [4] N. Rathore, V. J. Lodha, Y. V. Soni, Y. R. Kashyap, and K. S. Srinivas, "Spatio Temporal Analysis for Urban Growth Modelling
    using Cellular Automata and Genetic Algorithm," in *2023 IEEE 6th International Conference on Big Data and Artificial
    Intelligence (BDAI)*, pp. 208-213, 2023.

    [5] E. Imam and S. P. S. Kushwaha, "Habitat suitability modelling for Gaur (Bos gaurus) using multiple logistic regression, remote
    sensing and GIS," *Journal of Applied Animal Research*, vol. 41, no. 2, pp. 189-199, 2013.
    [6] J. Rew, Y. Cho, J. Moon, and E. Hwang, "Habitat suitability estimation using a two-stage ensemble approach," *Remote
    Sensing*, vol. 12, no. 9, p. 1475, 2020.

    [7] F. Wu, Q. Zou, and X. Dai, "Habitat Quality Analysis and Future Simulation Based on Artificial Neural Network-cellular
    Automata Model," *Sensors & Materials*, vol. 36, 2024.

    [8] A. Saleh, M. Sheaves, D. Jerry, and M. R. Azghadi, "Applications of deep learning in fish habitat monitoring: A tutorial and
    survey," *Expert Systems with Applications*, 2023, p. 121841.

    [9] T. Gränzig, A. Clasen, F. E. Fassnacht, A. Cord, and M. Förster, "Combining remote sensing, habitat suitability models and
    cellular automata to model the spread of the invasive shrub Ulex europaeus," *Biological Invasions*, vol. 25, no. 12, pp. 3711-
    3736, 2023.

    [10] N. Rathore, V. J. Lodha, Y. V. Soni, Y. R. Kashyap, and K. S. Srinivas, "Spatio Temporal Analysis for Urban Growth Modelling
    using Cellular Automata and Genetic Algorithm," in *2023 IEEE 6th International Conference on Big Data and Artificial
    Intelligence (BDAI)*, pp. 208-213. IEEE, 2023.

    [11] N. Nath, D. Sahariah, G. Meraj, J. Debnath, P. Kumar, D. Lahon, K. Chand et al., "Land use and land cover change monitoring
    and prediction of a UNESCO world heritage site: Kaziranga eco-sensitive zone using cellular automata-Markov model," *Land*,
    [12] S. P. S. Kushwaha, P. S. Roy, A. Azeem, P. Boruah, and P. Lahan, "Land area change and rhino habitat suitability analysis in
    Kaziranga National Park, Assam," *Tigerpaper*, vol. 27, no. 2, pp. 9-17, 2000.

    [13] F. Ahmad, L. Goparaju, and A. Qayum, "Wildlife habitat suitability and conservation hotspot mapping: remote sensing and
    GIS based decision support system," *AIMS Geosciences*, vol. 4, no. 1, pp. 66-87, 2018.

    [14] Muhammed, Khaleel, Aavudai Anandhi, and Gang Chen. "Comparing Methods for Estimating Habitat Suitability." *Land*, vol.
    11, no. 10, 2022, pp. 1754.

    [15] Georgiades, Pantelis, Yiannis Proestos, Jos Lelieveld, and Kamil Erguler. "Machine Learning Modeling of Aedes albopictus
    Habitat Suitability in the 21st Century." *Insects*, vol. 14, no. 5, 2023, pp. 447.

    [16] Rendana, Muhammad, Wan Mohd Razi Idris, Sahibin Abdul Rahim, Hazem Ghassan Abdo, Hussein Almohamad, and Ahmed
    Abdullah Al Dughairi. "Habitat suitability analysis in a natural peat swamp forest on Sumatran elephants using remote sensing
    and GIS." *Forest Science and Technology*, vol. 19, no. 3, 2023, pp. 221-231.

    [17] Jung, Jihyang, Yo Shimizu, Kenji Omasa, Sungsu Kim, and Sangdon Lee. "Developing and testing a habitat suitability index
    model for Korean water deer (Hydropotes inermis argyropus) and its potential for landscape management decisions in Korea."
    *Animal Cells and Systems*, vol. 20, no. 4, 2016, pp. 218-227.
    THANK YOU

    You might also like