Varnish-IBM-Call-For-Code-Global-Challenge
Challenge Overview
Call for Code aims to tackle the imminent and existential threat to Planet Earth: climate change. As the United Nations describes, “The impacts of climate change are global in scope and unprecedented in scale. Without drastic action today, adapting to these impacts in the future will be more difficult and costly.” The level of urgency surrounding the threats of climate change require immediate action, and Call for Code is arming its growing community of 400,000 developers and problem solvers across 179 nations with the tools to build tech solutions that can fight back.
Link: https://developer.ibm.com/blogs/2021-call-for-code-launch/
My Approach
Problem:
Research suggests that over the past 75 years, flowers have also adapted to rising temperatures and declining ozone by altering ultraviolet (UV) pigments in their petals.
Flowers’ UV pigments are invisible to the human eye, but they attract pollinators and serve as a kind of sunscreen for plants. Just as UV radiation can be harmful to humans, it can also damage a flower’s pollen. The more UV-absorbing pigment the petals contain, the less harmful radiation reaches sensitive cells.
Matthew Koski, a plant ecologist at Clemson University, examined plant collections from North America, Europe, and Australia dating back to 1941. In all, they examined 1238 flowers from 42 different species. They photographed flower petals from the same species collected at different times throughout their natural range using a UV-sensitive camera, which captured changes in UV pigment. They then matched these changes to data on the local ozone level and temperature.
On average, pigment in flowers at all locations increased over time—an average of 2% per year from 1941 to 2017,
Cause:
Pollen hidden within petals is naturally shielded from UV exposure, but this extra shielding can also act like a greenhouse, trapping heat. When these flowers are exposed to higher temperatures, their pollen is in danger of being cooked, he says. Reducing UV pigments in the petals causes them to absorb less solar radiation, bringing down temperatures.
Solution:
- Convolutional neural networks (CNNs) are achieving great performance in several computer vision tasks, Here I propose a robust
- and generalizable CV-based system for - automatically detecting depigmentation in various species in satellite and aerial images
- based on open data and tools. In particular, we designed a three-step approach, where the first CV finds the input images with
- change in color presence, the second CV provides a comparison to its original color and the third CV measures the intensity of UV in pollen. How Is This Better:
Prior to this solution scientists use samples to examine under compound microscope which would need optimal storage condition and would have much lesser accuracy of measurement. Technology & Concept Used: