Identification of Medicinal Plants using Visual Characteristics of Leaves and Flowers

Abstract

The proposed framework helps in ID of plant sickness and gives cures that can be utilized as a safeguard component against the illness. The information base got from the Internet is appropriately isolated and the distinctive plant species are recognized and are renamed to frame a legitimate data set then, at that point get test-data set which comprises of different plant infections that are utilized for checking the exactness and certainty level of the undertaking. Then, at that point utilizing preparing information, the classifier is prepared and afterward yield will be anticipated with ideal exactness. The proposed system comes under Machine learning domain. Machine Learning centers around the improvement of programs that can get to information and use it to find out on their own. Machine Learning has various applications and has been used to tackle real world problems in an efficient manner. It has applications in medicine, communication, entertainment, military and so on.

Convolutional Neural Network (CNN) which comprises of different has been used for classification and prediction. The problem with existing systems is that they are limited to a few numbers of plant species or due to use of inefficient algorithms have not been able to achieve the desired levels of accuracy. With the proposed system and training model an accuracy level of 78% was achieved. The proposed framework gives the name of the plant species with its certainty level and the cure that can be taken as fix.