OEM service precision parts of bronze knobs
Just as crime scene tracks can be vital evidence, wildlife tracks can be just as valuable to wildlife conservationists. OEM service precision parts of bronze knobs
Every species — and every animal within that species — has its own unique footprint. Each footprint and the substrate beneath it tells a story about animal activity and the environment. Information gleaned from the tracks helps conservationists track wildlife behavior and monitor endangered species. This information also helps protect biodiversity on Earth.
go all out
Conservationists can learn a lot from trails, but manually processing all the information is a tedious task. Wildlife conservation group WildTrack created Footprint Identification Technology (FIT) as a non-invasive way to monitor wildlife. Digital images of animal footprints are measured to create geographic profiles, which are then analyzed to determine the animal’s species, age class, sex, and individual identity. OEM service precision parts of bronze knobs
FIT is very accurate, but time consuming. Each image requires manual data entry and annotation by field experts. Even with all the data entered, analysis is slow due to human and geographic constraints. To expedite this process and support the collection of data from citizen scientists around the world, WildTrack sought an AI-based image analysis solution.
Conservationists teamed up with a team of researchers at UC Berkeley to automate the FIT process. The resulting tool integrates JMP data visualization software and SAS data analytics, leveraging artificial intelligence, computer vision, and machine learning to identify and analyze animal footprints faster and more accurately. SAS and Intel are long-term partners in developing global analytics solutions. WildTrack is currently using an end-to-end solution in the field. OEM service precision parts of bronze knobs
Image analysis a success
To train the algorithm, the researchers lured tigers and other wild animals on a prepared field. They took multiple images of each foot—four feet per animal equals four times as much data as can be collected. OEM service precision parts of bronze knobs
Citizen scientists around the world can also upload photos. Researchers, trackers and citizen scientists can download a smartphone app to upload images from anywhere in the world. Images and related data, such as location, date and time, are fed into an AI-based platform that can identify dozens of species with more than 90 percent accuracy. Additionally, the eBeeX drone from senseFly is equipped with an interface that supports aerial data collection.
The WildTrack database contains rhino, tapir and bear tracks. The expanded data collection further strengthens the machine learning algorithms and encourages discussions among researchers. Additionally, the person uploading the image can track the animal’s movements online. OEM service precision parts of bronze knobs
The AI-based solution enables the WildTrack team to expand its research. The original footprint recognition now includes other biometrics, including coat color, coat type, and call. WildTrack can monitor migration patterns, mating activity, foraging and hunting behavior, and social groups. Differences in the substrate beneath the footprints indicate changes in humidity levels, weather conditions and gait, the latter of which can indicate walking speed or signs of injury.
Adjustments to machine learning algorithms will also support research in other areas. AI-based solutions can be used to count and monitor endangered animals and prevent poaching. Future uses could include tracking turtles based on their unique shell types, identifying plant and animal types and health, or detecting and monitoring traces of animal or human activity. OEM service precision parts of bronze knobs
