Animal Footprint Classification using Machine Learning and Advance Deep Learning Methods
Keywords:
Footprint Classification,, Machine Learning,, Deep Learning, YOLO, Ensemble Model,, MobileNet, Xception, DenseNet,, InceptionResNetV2, ResNet50, NasNetMobile”.Abstract
Categorization of animal footprints is a significant aspect of ecological research, conservation, and
wildlife monitoring. Here in our research, we introduce a novel method to identify and classify animal footprints
through the application of state-of-the-art deep learning and machine learning methods. An ensemble model
combining "Xception and NasNetMobile" is primarily discussed, but "ResNet50, InceptionResNetV2,
DenseNet, MobileNet, Xception, and NasNetMobile" are also used in the classification task. Contemporary
algorithms like "YOLOV5x6, YOLOV5s6, YOLOV8n, and YOLOV9n" are utilized for footprint identification.
The proposed ensemble model performs the highest classification accuracy by combining Xception and
NasNetMobile. In the case of footprint identification, YOLOV5x6 was the superior method since it resulted in
astounding outcomes with excellent precision. The efficacy of the models could be measured through
calculating measures such as "recall, precision, and F1 score" evaluations. Results reveal that such cutting-edge
deep machine learning algorithms find application in environment and wildlife research by simplifying and
enhancing identification and categorization of animal footprints.
