1.

NRSC Bhuvan

 Website: https://bhuvan.nrsc.gov.in/

 Description: Bhuvan is the Indian geospatial platform developed by the National Remote
Sensing Centre (NRSC), part of ISRO. It provides high-resolution satellite imagery, thematic
maps, and various geospatial tools for analyzing land use, urban planning, disaster
management, and environmental monitoring.

 Key Data: Satellite imagery from Cartosat, Resourcesat, Landsat, and Sentinel; thematic
maps for agriculture, urban planning, water resources, etc.

2. Survey of India (SOI)

 Website: https://www.surveyofindia.gov.in/

 Description: The Survey of India is the national mapping agency responsible for producing
topographic maps and geospatial data. It offers maps, boundary data, and geodetic
information for various applications.

 Key Data: Topographic maps, thematic maps, digital elevation models (DEMs), boundary
maps, cadastral maps.

3. Indian National Centre for Ocean Information Services (INCOIS)

 Website: https://incois.gov.in/

 Description: INCOIS provides oceanographic and coastal geospatial data for marine and
coastal applications, including tsunami early warnings, ocean state forecasts, and marine
biodiversity mapping.

 Key Data: Coastal and oceanographic data, bathymetric maps, marine weather data,
tsunami and cyclone monitoring.

4. National Remote Sensing Centre (NRSC)

 Website: https://www.nrsc.gov.in/

 Description: NRSC is part of ISRO and focuses on remote sensing data and satellite imagery.
It provides access to high-resolution satellite images, thematic datasets, and analytical tools
for land use, urban planning, and disaster management.

 Key Data: Satellite imagery from Cartosat, Resourcesat, and other ISRO satellites, including
land cover, urban growth, and hydrological data.

5. National Spatial Data Infrastructure (NSDI)

 Website: https://nsdiindia.gov.in/

 Description: NSDI aims to create a national framework for sharing and integrating geospatial
data across various sectors. It provides standardized datasets for easy access and use in
development planning and decision-making.

 Key Data: Cadastral maps, administrative boundaries, topographic data, and thematic
datasets.

6. India-WRIS (Water Resources Information System)

  Website: https://indiawris.gov.in/

 Description: Developed by the Central Water Commission and NRSC, India-WRIS provides
geospatial data related to water resources, including river basins, reservoirs, and water
bodies.

 Key Data: River basin maps, water resources data, hydrological information, and flood
monitoring tools.

7. Forest Survey of India (FSI)

 Website: http://www.fsi.nic.in/

 Description: FSI provides geospatial data on forest cover, deforestation, and biodiversity in
India. It conducts biennial assessments of India's forest cover and publishes reports on forest
health.

 Key Data: Forest cover maps, forest type maps, vegetation data, and forest inventory
reports.

8. Open Government Data (OGD) Platform India

 Website: https://data.gov.in/

 Description: The OGD platform provides a wide range of datasets across various sectors,
including geospatial data. It offers open access to data from different government ministries
and departments.

 Key Data: Administrative boundaries, demographic data, land use data, environmental data.

9. National Atlas and Thematic Mapping Organisation (NATMO)

 Website: http://www.natmo.gov.in/

 Description: NATMO produces thematic maps and atlases for various sectors, including
agriculture, infrastructure, environment, and demographics. It provides detailed geospatial
data and mapping services.

 Key Data: Thematic maps, atlases, and geospatial data for various sectors like environment,
population, resources, and tourism.

10. Bhaskaracharya Institute for Space Applications and Geoinformatics (BISAG-N)

 Website: https://bisag-n.gov.in/

 Description: BISAG-N provides geospatial services for various sectors like agriculture,
disaster management, and urban planning. It offers GIS-based decision support systems and
satellite data for state and national projects.

 Key Data: Geospatial data for agriculture, disaster management, and infrastructure
development, including GIS-based decision-making tools.

11. Ministry of Environment, Forest and Climate Change (MoEFCC)

 Website: http://www.moef.nic.in/
  Description: MoEFCC provides data related to environmental monitoring, biodiversity, and
forest resources. The ministry focuses on sustainability, conservation, and the management
of natural resources.

 Key Data: Biodiversity data, environmental impact assessments, forest resources, climate
change-related data.

12. Geospatial Delhi Limited (GSDL)

 Website: http://www.gsdl.org.in/

 Description: GSDL provides geospatial data for the National Capital Territory of Delhi,
offering datasets related to land use, infrastructure, and utilities.

 Key Data: Land use/land cover maps, cadastral data, utility mapping, infrastructure data for
Delhi.

13. Regional Remote Sensing Centre (RRSC)

 Website: https://www.isro.gov.in/rrsc

 Description: RRSCs, under NRSC, are regional centers that provide satellite data and
geospatial services for various states in India. They focus on regional planning, agriculture,
and resource management.

 Key Data: Region-specific satellite imagery, land use maps, environmental monitoring data.

14. Vikram Sarabhai Space Centre (VSSC)

 Website: https://www.vssc.gov.in/

 Description: VSSC, part of ISRO, provides geospatial and satellite data services related to
space research and satellite technology, contributing to national geospatial capabilities.

 Key Data: Satellite mission data, space research outputs, environmental monitoring data.

15. Digital India Land Records Modernization Programme (DILRMP)

 Website: https://dilrmp.gov.in/

 Description: DILRMP focuses on digitizing land records and cadastral data across India,
aiming to improve transparency in land ownership and land use management.

 Key Data: Cadastral maps, land ownership data, land use information.
Supervised and unsupervised classification are two common methods used in remote
sensing and geographic information systems (GIS) for analyzing and classifying image data,
such as satellite or aerial imagery. Both techniques involve categorizing pixels in an image
into different classes or themes (e.g., water, vegetation, urban areas), but they differ in how
this process is carried out.

1. Supervised Classification

In supervised classification, the user provides the algorithm with predefined, labeled
examples of different land cover types or features. This process involves "training" the
algorithm to recognize patterns based on the input data.

Steps in Supervised Classification:

 Training Data Collection: The user selects representative sample areas for each land
cover class (e.g., water, forest, urban). These areas are called training sites.
 Algorithm Training: The classification algorithm analyzes the spectral
characteristics (e.g., reflectance values) of the pixels in the training sites and uses this
information to define the spectral signature for each class.
 Classification: Once trained, the algorithm applies these spectral signatures to the rest
of the image, assigning each pixel to one of the predefined classes.
 Accuracy Assessment: After classification, the results are typically validated by
comparing them with ground truth data or other reference information to assess
accuracy.

Advantages of Supervised Classification:

 Accuracy: Because the user defines the classes, the classification is generally more
accurate, assuming the training data is representative.
 Customization: The user can choose specific classes that are of interest for the
analysis.

Disadvantages of Supervised Classification:

 Time-Consuming: Selecting good training samples requires significant effort and

expertise.
 Subjectivity: The accuracy of the classification heavily depends on the quality and
representativeness of the training data.

Examples of Algorithms for Supervised Classification:

 Maximum Likelihood Classification (MLC)

 Support Vector Machine (SVM)
 Random Forest (RF)
2. Unsupervised Classification

In unsupervised classification, the algorithm automatically groups pixels into different

classes based on their spectral properties, without the need for user-defined training data.
This method is useful when there is limited prior knowledge about the land cover types
present in the image.

Steps in Unsupervised Classification:

 Clustering: The algorithm groups pixels with similar spectral characteristics into
clusters. The number of clusters (or classes) can be predefined or decided by the
algorithm.
 Labeling: After the clustering is complete, the user interprets the clusters and assigns
each one to a specific land cover class (e.g., water, forest, urban) based on their
knowledge of the study area or by comparing the results with reference data.

Advantages of Unsupervised Classification:

 No Training Data Required: This makes it faster and easier to use, especially in
areas where there is little known information.
 Data-Driven: It relies purely on the natural spectral patterns in the data, which can
sometimes reveal unexpected classes.

Disadvantages of Unsupervised Classification:

 Less Accurate: Because the algorithm is not trained, the resulting classes may not
always correspond well to actual land cover types.
 Interpretation Required: The user must manually interpret the resulting clusters,
which can introduce subjectivity.

Examples of Algorithms for Unsupervised Classification:

 K-means Clustering
 ISODATA (Iterative Self-Organizing Data Analysis Technique)

Key Differences:

Aspect Supervised Classification Unsupervised Classification

No training data required; data-driven
Training Data Requires user-defined training data.
clustering.
User has control over the classes Limited user control; classes are
Control
being classified. defined by the algorithm.
Generally more accurate with good Typically less accurate unless clusters
Accuracy
training data. are well-defined.
 Aspect Supervised Classification Unsupervised Classification
More time-consuming due to training
Speed Faster since no training data is needed.
data collection.
Suitable when prior knowledge of Useful when little is known about the
Application
the area is available. area.
Classes are predefined and named by User must interpret and label the
Interpretation
the user. resulting clusters.

Example Applications:

 Supervised Classification:
o Urban planning: Defining specific classes like residential, industrial, forest,
and water for urban land-use analysis.
o Agriculture: Classifying different crop types based on their known spectral
signatures.
 Unsupervised Classification:
o Exploration in new or unknown areas: Discovering different land cover types
without prior knowledge of the landscape.
o Environmental monitoring: Identifying changes in land cover over time
without the need for specific class labels.