“Bird Nest Chronicles: A Novel Method for Avian Nest Classification Using Machine Learning”

Preserving natural habitats for avian species is crucial in addressing the impacts of climate change. The Indian Bird Conservation Foundation encounters challenges in manually distinguishing and categorizing bird nests for behavioural studies and habitat conservation. To address this issue, we propose an automated methodology leveraging machine learning algorithms, focusing on image processing, feature extraction, and image classification. Our image processing workflow includes various steps such as pre-processing and enhancement techniques like local contrast enhancement, sharpening, intensity adjustment, histogram equalization, and color thresholding. For feature extraction and identification, we utilize a powerful Convolutional Neural Network, ResNet152 V2. Following this, Support Vector Machine (SVM) is employed for image classification. Our experimentation demonstrates the superior performance of ResNet152 V2 in analysing bird nest images. This architecture, combined with a kernel SVM, achieves impressive accuracy, precision, recall, and F1-score, showcasing the efficacy of our approach in classifying bird nests in India. By minimizing human intervention in image analysis, our method emerges as a valuable tool for conservationists and researchers dedicated to studying and safeguarding these avian species and their habitats. Future endeavors aim to develop a bird nest detector, contributing significantly to wildlife conservation research in India. This innovative approach holds promise in advancing the field, providing a robust solution for challenges in manual image analysis for the conservation efforts of endangered bird species in the region.