Detect Poisonous Plants

Abstract

"Poisonous plants are dangerous for both people and animals since they can cause serious health issues or even death. The absence of a reliable method for accurately recognising these poisonous plants raises the possibility of accidental exposure to risks. This study aims to address the need for an accurate and efficient method of identifying potentially harmful plants. Deep learning techniques are used as part of the method for classification of image task. An extensive collection of pre-processed images of both poisonous and non-poisonous plant species is gathered in order to train the CNN model. To improve model performance and speed up the training process, transfer learning is used with pre-trained models like MobileNetV2. The prototype's findings reveal good accuracy rates. To ensure that the poisonous plant detection system is successfully implemented, it is also necessary to address issues with computing resources and model interpretability. The proposed model detects poisonous plant successfully and it was developed efficiently by utilizing available resources without using more computing resources and time."