Review on Poisonous Plants Detection Using Machine Learning

Abstract

Poisonous plants pose a significant threat to human and animal health, leading to various adverse effects ranging from mild discomfort to severe toxicity. Early identification of these harmful plants is crucial for preventing accidental ingestions and minimizing the associated risks. This project focuses on developing an efficient and accurate system for the detection of poisonous plants using machine learning techniques. The proposed solution leverages a comprehensive dataset comprising images of various plant species, categorized into poisonous and non-poisonous classes. Convolutional Neural Networks (CNNs) are employed for image feature extraction, allowing the model to discern subtle visual patterns indicative of poisonous plant characteristics. Transfer learning is applied using pre-trained models, enhancing the system's ability to generalize and adapt to diverse plant species