Visão biodiversa| Scanning for Lifeforms
Scanning for Lifeforms
Visão Biodiversa: a world beyond our eyes
Summary
The application developed with the idea of monitoring biodiversity through a photo is based on the premise of improving human knowledge about fauna. We will make use of the technology known as machine learning which consists of training the computer so that it recognizes the fauna.The project consists of using an infrared camera to scan the temperature emanating from the body recognized in the photo.With the support of NASA and its monitoring satellites such as TERRA which we will use to build a biome database and their relationship with the species belonging to it.
How We Addressed This Challenge
We developed an app to monitor the local biodiversity through infrared readings that gives us some important information, such as: the species mapping, how dangerous the animal is, if the observed species is a bioindicator, its diet, risk of going extinct and also a little database about specific biomes.
It works by capturing the heat emitted by mammals and birds with the help of a photo taken by the user with a thermal camera. This quantity of IR radiation is measured and then related to certain species. From this reading, we can identify things such as the animal’s popular name, it’s scientific name, the species distribution, it’s diet, if the species is at risk of going extinct and some characteristics of the species common habitat. After that, we combine these data with data from NASA’s satellite (Terra) to create a database that can connect the locations where the species was seen with the weather characteristics of the exact location where the photo was taken.
Our main goal is to create a biodiversity map on a global scale as well as provide relevant information about certain species.
We aim to open the eyes of society to preserve the things we don’t see and revolutionize today's industry with thermographic technology.
How We Developed This Project
The main idea came when when navigating between challenges, we came across something that directly impacts our current society, which is the control of biodiversity indexes and how we can see it all over the planet.
Our approach to the challenge was to create an app so that we could collect information about species on a global scale so that we could cross it with NASA's database to trace profiles about animals habitats.
At first, we chose to create only one layout of our application's user interface, using the Adobe XD platform. The XD, offers us the ability to simulate the operation of the application through test programs, such as the version that was made available for observation of the project. Through this and with the creation of a formed layout, we are faced with the difficulty of making available the gadget used to take the photos. With that, we went in search of a potential user of this application, and we asked for the help to remove the photos, in such a way that the understanding about the use of it had a small dissemination.
After making the photos available, from which we used a “Flir One Pro” camera, which can be found at the link https://www.flir.com.br/products/flir-one-pro/, we obtained a greater contact with technology to be presented as a differential of our project.
Finally, we continue with the study of the platforms that will be used for the development of a Database and our AI, focusing on two main factors, which are the storage of this data, and feeding the AI with this data.
For the creation of the database, we can use Firebase, which is a platform for storing these data in the cloud, facilitating storage and access. This database will be continuously fed with information provided by both users and NASA data regarding the climate and conditions of the locations where the photos were found.
For AI, on the other hand, we first needed to define what type of artificial intelligence would be used. We opted for supervised learning, as we need human assistance so that the development of our AI is able to increasingly improve its methods of association. Therefore, we will use the Python language to build the code of this AI, together with O scikit-learn, a tool for the development of machine learning, that is, what AI will teach.
How We Used Space Agency Data in This Project
Our app uses data from NASA's satellite "Terra" together with our readings to trace a profile for each species and it's common habitat so that we can predict how certain species will behave when facing climate changes.
Project Demo
Project Code
Data & Resources
1- http://sea-entomologia.org/IDE@/revista_57.pdf artcile about megalopteras (acess: 10/03/2020)
2- http://www.invivo.fiocruz.br/cgi/cgilua.exe/sys/start.htm?infoid=958&sid=2 data about the amazonic biome (acess: 10/03/2020)
3- http://www.invivo.fiocruz.br/cgi/cgilua.exe/sys/start.htm?infoid=963&sid=2 data about the pantanal biome (acess: 10/03/2020)
4- https://www.flir.com/globalassets/imported-assets/document/flir-one-pro-series-datasheet.pdf specifications about the IR camera (acess: 10/04/2020)
5- https://worldview.earthdata.nasa.gov/?t=2019-10-04-T22%3A50%3A14Z&l=MODIS_Combined_L4_FPAR_8Day,Reference_Labels(hidden),Reference_Features(hidden),Coastlines,VIIRS_NOAA20_CorrectedReflectance_TrueColor(hidden),VIIRS_SNPP_CorrectedReflectance_TrueColor,MODIS_Aqua_CorrectedReflectance_TrueColor(hidden),MODIS_Terra_CorrectedReflectance_TrueColor(hidden) data about forest density (acess: 10/04/2020)
6- https://worldview.earthdata.nasa.gov/?v=-315.75263527640533,-116.30061236506099,168.29328308008132,105.16988468677222&t=2019-10-04-T22%3A50%3A14Z&l=MODIS_Combined_L4_FPAR_8Day,Reference_Labels(hidden),Reference_Features(hidden),Coastlines,VIIRS_NOAA20_CorrectedReflectance_TrueColor(hidden),VIIRS_SNPP_CorrectedReflectance_TrueColor,MODIS_Aqua_CorrectedReflectance_TrueColor(hidden),MODIS_Terra_CorrectedReflectance_TrueColor(hidden) data about chlorophyll levels (acess: 10/04/2020)
7- http://www.dgi.inpe.br/documentacao/satelites/terra data about "Terra" satellite (acess: 10/04/2020)
8- Al-doski, J., Mansor, S.M., Shafri, H.Z.B.M. (2016) THERMAL IMAGING FOR PESTS DETECTING—A REVIEW. International Journal of Agriculture, Forestry and Plantation, Vol. 2 (February).10-30.