Zebra and whale vacation photos posted by tourists on social media can offer the unexpected benefit of helping researchers track and collect information about endangered species.
Scientists use artificial intelligence (AI) to analyze photographs of zebras, sharks, and other animals to identify and track individuals and provide new insights into their movements and population trends. ..
Tanya Burger Wolf, director of the Translational Data Analysis Institute at Ohio State University, said:
“These images contain a wealth of data that can be extracted and analyzed to protect animals and combat extinction.”
In a new field called image omics, we are taking wildlife images one step further by using AI to extract animal biological information directly from photographs. , Ohio State University Ecology and Biological Biology.
With recent advances in analyzing wildlife images using AI, she presentation February 20 American Association for the Advancement of Science Annual Meeting..She spoke in a science session “Crowdsourced science: volunteering and machine learning protect the wild for everyone.”
One of the biggest challenges facing environmentalists is the many endangered species and Endangered species..
“We are losing biodiversity at an unprecedented rate. We don’t even know how much and what we are losing,” Burger Wolf said.
“If you want to save African elephants from extinction, you need to know how many are in the world, where they are, and how quickly they are declining,” Burger Wolf said.
“We don’t have enough GPS collars and satellite tags to monitor all elephants and answer those questions. However, we can use AI technologies such as: Machine learning It analyzes images of elephants and provides much of the information you need. “
Berger-Wolf and her colleagues Wildbook It uses computer vision algorithms to analyze photographs taken by tourists and field researchers during vacations to identify individuals as well as animal species.
“Our AI algorithms can identify individuals using the shape of whale flukes and dolphin dorsal fins, such as stripes, spots, wrinkles, and notches,” she says.
For example, the Wildbook contains more than 2 million photos of approximately 60,000 whales and dolphins uniquely identified from around the world.
“This is one of the main sources that scientists currently have about killer whales. They are no longer lacking data,” she said.
In addition to sharks and whales, there are wild books on zebras, turtles, giraffes, African carnivores and other species.
Berger-Wolf and her colleagues have developed an AI agent for publicly shared searches. Social media Posts of related species. That is, the vacation photos of many sharks seen in the Caribbean will be used in wildbooks, for example, for science and protection, she said.
These photographs provide information about when and where the images were taken, as well as populations, birth and death dynamics, species range, social interactions, and interactions with other species, including humans. , Helps protect.
This was very useful, but Berger-Wolf said researchers are aiming to move this field forward with image omics.
“The ability to extract biological information from images is the basis of imaging,” she explained. “We teach machines to see in images what humans may miss or not see.”
For example, does the zebra stripe pattern resemble the mother’s pattern in some meaningful way? If so, can it give information about their genetic similarities? How do bat skulls change with environmental conditions, and what evolutionary adaptations drive that change? These and many other questions may be answered by machine learning analysis of photographs.
The National Science Foundation awarded Ohio $ 15 million in September to lead the creation of the Imageomics Institute to help guide scientists around the world in this new field. Burger Wolf is a Principal Investigator at the Institute.
As the use of AI in analyzing wildlife images continues to grow, one key is to ensure that AI is used fairly and ethically, Berger-Wolf said.
For one thing, researchers must make sure that it does no harm. For example, data should be protected from being used by poachers to target endangered species.
But that’s not all.
“We need to make sure that it’s a human-machine partnership that humans trust AI. AI is by design participatory and needs to connect people, data, and geographic locations. There is, “she said.
Ohio State University
Quote: See https: //phys.org/news/2022-02-vacation-photos-zebras-whales.html to see how zebra and whale vacation photos can help with conservation (February 20, 2022). Obtained February 20, 2022
This document is subject to copyright. No part may be reproduced without written permission, except for fair transactions for personal investigation or research purposes. Content is provided for informational purposes only.
How zebra and whale vacation photos can help protect
Source link How zebra and whale vacation photos can help protect