Robust and Efficient Analysis Approaches of Remote Imagery for Assessing Population and Forest Health in India
Lead Research Organisation:
University of Cambridge
Department Name: Applied Maths and Theoretical Physics
Abstract
India faces tremendous societal and ecological challenges. Cities are growing which is accompanied by an increase in population and consequently traffic. Transport in India's cities plays an important role in air pollution and a large volume of road traffic fatalities. At the same time, while India's forest cover is on average increasing, it is not clear how much of this is due to plantation in contrast to natural forest, a knowledge gap that is possibly endangering biodiversity of India's forests.
Standardly collected remote sensing data of India offers a great opportunity for quantifying the status quo of these factors and turning them into ecological and health models that can inform new government policies to help tackle these challenges.
In this project, we will develop novel mathematical methods that can unlock the wealth of information contained in remote sensing data, with a focus on improving upon two of India's challenges: traffic management and forest conservation.
We will focus on the development of novel image analysis methods for quantifying traffic volume stratified with respect to traffic mode, i.e. car, bus, tuk tuk, lory, bicycle, pedestrian etc. Our analysis will focus on some of the most populated and polluted cities in India such as Dehli, Mumbai and Bengaluru, using image data obtained from satellites combined with more localised traffic camera data. Algorithms developed in the project as well as associated statistics drawn from the data will be made available to the general public as well as communicated to relevant stakeholders in India.
In the context of forest conservation, our project will develop new algorithms for mapping different tree species from India's forests from satellite data.
Supported by an interdisciplinary project team of researchers and stakeholders from academia and industry, and from India and Cambridge, and by tightly combining the development of novel mathematical methods for remote sensing data with knowledge transfer, our project aims to provide a step change towards improved decision making in traffic and forest policies in India
Standardly collected remote sensing data of India offers a great opportunity for quantifying the status quo of these factors and turning them into ecological and health models that can inform new government policies to help tackle these challenges.
In this project, we will develop novel mathematical methods that can unlock the wealth of information contained in remote sensing data, with a focus on improving upon two of India's challenges: traffic management and forest conservation.
We will focus on the development of novel image analysis methods for quantifying traffic volume stratified with respect to traffic mode, i.e. car, bus, tuk tuk, lory, bicycle, pedestrian etc. Our analysis will focus on some of the most populated and polluted cities in India such as Dehli, Mumbai and Bengaluru, using image data obtained from satellites combined with more localised traffic camera data. Algorithms developed in the project as well as associated statistics drawn from the data will be made available to the general public as well as communicated to relevant stakeholders in India.
In the context of forest conservation, our project will develop new algorithms for mapping different tree species from India's forests from satellite data.
Supported by an interdisciplinary project team of researchers and stakeholders from academia and industry, and from India and Cambridge, and by tightly combining the development of novel mathematical methods for remote sensing data with knowledge transfer, our project aims to provide a step change towards improved decision making in traffic and forest policies in India
Planned Impact
Our project aims to benefit policy making in climate change, healthcare and environment conservation in India. We will develop and make available new mathematical algorithms which allow to quantify, in an accurate and highly stratified way, the status quo of traffic in India's cities and the composition of India's forests. Equipped with a picture of the current situation, the Indian Government will be better informed when making policy decisions, ultimately benefiting the whole of society in India.
India has some of the noisiest and most polluted cities in the world, with a growing burden of traffic deaths and an increase in non-communicable disease associated with lack of physical activity. Having accurate knowledge as well as an overall picture of the traffic quantity in cities around India will help the government to better target and assess the impact of intervention policies in transport and health, benefiting the local inhabitants. Directly from this project, the current status will be established in the cities of Delhi, Jaipur, Chennai, Bengaluru, Kolkata and Mumbai. However, the project has inbuilt mechanisms to enable local partners and stakeholders to take up the novel methods developed to continue to monitor change and roll out to the entire country.
India is one of the megadiverse countries, but its large population and development objectives are putting huge pressure on forests. In particular there is not a clear picture of how much of the current forest is plantation as opposed to natural forest. Our species level mapping will be able to alert as to the extent of plantation cover and hence inform future decisions to minimise loss in forest quality and biodiversity, ultimately benefiting local communities who depend on the forests for fuel, food and livelihoods. Naturally, ecological conservation benefits the country as a whole. Via this project information will be obtained on the status of forests in the states of Sikkim, Assam and Madhya Pradesh, but here again the methods will be able to be subsequently extended to other states.
Both of the above actions will also help mitigate climate change, via maintenance of natural forest cover and reductions in carbon emissions from vehicles, and help the Indian economy via transportation and environmental policies that reduce damage to health and the environment.
India has some of the noisiest and most polluted cities in the world, with a growing burden of traffic deaths and an increase in non-communicable disease associated with lack of physical activity. Having accurate knowledge as well as an overall picture of the traffic quantity in cities around India will help the government to better target and assess the impact of intervention policies in transport and health, benefiting the local inhabitants. Directly from this project, the current status will be established in the cities of Delhi, Jaipur, Chennai, Bengaluru, Kolkata and Mumbai. However, the project has inbuilt mechanisms to enable local partners and stakeholders to take up the novel methods developed to continue to monitor change and roll out to the entire country.
India is one of the megadiverse countries, but its large population and development objectives are putting huge pressure on forests. In particular there is not a clear picture of how much of the current forest is plantation as opposed to natural forest. Our species level mapping will be able to alert as to the extent of plantation cover and hence inform future decisions to minimise loss in forest quality and biodiversity, ultimately benefiting local communities who depend on the forests for fuel, food and livelihoods. Naturally, ecological conservation benefits the country as a whole. Via this project information will be obtained on the status of forests in the states of Sikkim, Assam and Madhya Pradesh, but here again the methods will be able to be subsequently extended to other states.
Both of the above actions will also help mitigate climate change, via maintenance of natural forest cover and reductions in carbon emissions from vehicles, and help the Indian economy via transportation and environmental policies that reduce damage to health and the environment.
Organisations
- University of Cambridge, United Kingdom (Lead Research Organisation)
- KritiKal Solutions Private Limited (Project Partner)
- Forest Survey of India (Project Partner)
- German Aerospace Centre DLR, Germany (Project Partner)
- C40 Cities (Project Partner)
- Cognizant Technology Solutions, United Kingdom (Project Partner)
- Iora Ecological Solutions (Project Partner)
Publications

Bungert L
(2020)
Variational regularisation for inverse problems with imperfect forward operators and general noise models
in Inverse Problems


Lunz S
(2021)
On Learned Operator Correction in Inverse Problems
in SIAM Journal on Imaging Sciences

Zhang J
(2020)
Dynamic spectral residual superpixels
in Pattern Recognition