Micro plastic detection in water samples

Microplastic pollution in water bodies poses a significant threat to marine ecosystems and human health. Detecting the presence of microplastics in water samples is crucial for monitoring and mitigating environmental damage. This project aims to develop a deep learning-based system for the detection of microplastics in water samples using image classification techniques. The system utilizes four state-of-the-art deep learning algorithms: MobileNetV2, ResNet50, Vision Transformer, and Swin Transformer, to analyze images of water samples and classify them as either containing microplastics or not. The models are trained on a large dataset of labeled water sample images, where each image is annotated to indicate the presence or absence of microplastics. The goal is to automate the detection process, enabling rapid and accurate identification of microplastics in water, which can aid in environmental monitoring, policy-making, and the development of effective waste management strategies. The system allows for easy input of water sample images, processes them through the trained models, and provides a prediction regarding the presence of microplastics. This research demonstrates the potential of deep learning techniques in addressing environmental challenges, particularly in the context of microplastic pollution. 

  Keywords: Microplastic detection, MobileNetV2, ResNet50, Vision Transformer, Swin Transformer, water samples, image classification, environmental monitoring, deep learning, pollution detection.