At Colorado State University, a forward-looking project led by Ph.D. student Cheng Guo is leveraging machine learning to expedite the study of African leopards, an endangered species. Guo is crafting algorithms that enhance image processing, a major boon for ecologists sifting through data from motion-triggered cameras in the wild. The technology identifies individual leopards, analyzing the unique patterns of their spots, and offers researchers relief from the grind of manual image sorting, as reported by CSU SOURCE.
These cameras, vital for monitoring populations across expansive habitats, can capture thousands of images, many triggered by the wrong subjects, resulting in a backlog of blurry or irrelevant shots. Guo is improving upon current models to allow for better recognition, and dealing with an inherent scarcity of high-quality images of these elusive animals. Companies such as Apple, however, enjoy an endless stream of user data, providing a stark contrast in resources. As Guo mentioned, as obtained by CSU SOURCE, "For our project though, we may only have one or maybe two usable images of a specific leopard to then compare to other animals in the set."
Guo's approach to image processing is somewhat akin to the facial recognition technology used in modern smartphones, but catered to the distinctive spot patterns of leopards. While her work could benefit other species with unique markings, such as zebras, Guo and her team face challenges in refining this technology due to limited access to quality data and the sporadic nature of leopard encounters with cameras in their vast habitats, as per CSU SOURCE.
The project has garnered support from Panthera Corp., an organization dedicated to the preservation of wild cats, and has gained academic recognition with a paper in IEEE Transactions on Automation Science and Engineering. Guo's journey from China to CSU is a tale of seizing opportunity; after benefiting from the Intensive English Program, she dove into a master's in computer engineering before embarking on her Ph.D. under the advisement of CSU Professor Tony Maciejewski and Agnieszka Miguel from Seattle University. Her research quest marries her interests in programming, machine learning, and, curiously enough, community events that foster interdisciplinary connection within the academic sphere.
