Abstract

Abstract:- In the present era of living where flower is used for cosmetics, medicines, artificial food colors and so on, the mainaim of researchers and biologists is to first identify the variety of the flower, its species and its origin to understand its propertiesand make an appropriate use of each. This research approach to provide an automated flower recognition which is helpful inareas such as botany research, Ayurveda treatment, farming, horticulture etc. Nature has accomplished our surroundings withdifferent kinds of flower species and, traditional recognition of the flower is done by botanist, being a challenging task. Theabove problem can be resolved with the help of machine learning algorithms where each flower species is differentiated basedon shape, geometry, texture, color and also various plant parts such as leaves and stems. While modern search engines providemethods to visually variation among millions of flower species around the world. Hence in this proposed research work, a Deeplearning approach using Convolutional neural networks (CNN) is used to recognize flower species with high accuracy. In thisproject training of the system is provided with the data collection I.e., images of 102 species and ranging from around 1000images of each. The output can be visualized by checking with an input image and classifying its variety with the system andoutput is shown on the website. In the website drag and drop of the image can be done which further predicts the species of theflower in the image and also the accuracy of prediction is also given.Keywords: Floral recognition, CNN, Model training, Floral identification Website