Gujarat Forest Department introduced Software with Intelligent Marking Based Identification of Asiatic Lions (SIMBA) which is an improved AI-based photo-identification software designed specifically to understand the population dynamics, dispersal and social system of the Asiatic lions. SIMBA is now housed in the recently established “Gir Hi-Tech Monitoring Unit” at Sasan-Gir.
SIMBA functions using the Machine Learning technique that enables a pairwise comparison that automates the individual identifications. The identification process is done by analysing the presence of scars on the face, variability in a whisker spot pattern, notches on ears and other metadata of the photograph. Using this software unique features of the lion can be extracted and similar patterns or marks can be clustered within the embedding space of the Machine Learning.
During the training process, with the sufficient data installed in the model, it becomes more efficient and makes it possible to separate the biological aspects of the lion according to the needs of the researcher. Separated areas can be used for further identification. After the identification, SIMBA also generates a Unique Identification Number, which is stored in the software, for future reference. Through this ID number, we can understand if a particular lion’s records already exist in the data or if it's new.
Other criteria such as name, gender, life status, microchip number, lactating, can also be understood using this database. The software also allows Wildlife veterinary records to be added to the system.
Source:
Comments