Gerbera Flower Detection by Machine Learning Approach

Abstract

Gerbera species from South Africa were formally recognized for the first time in 1889. It is now protected in its native habitats across the tropics. The flower has become a common ornament for backyard gardens in many parts of the world. Gerberas are originally cultivated in Magura. Many young farmers are engaged in farming in Faridpur, Jashore, and other locations. Godadhordangi village, Aliabad union, Faridpur Sadar upazila has a Gerbera flower garden. This "gerbera flower Detection by Machine Learning Approach" improves our life while also expanding our floral knowledge. This paper presents a strategy for finding and identifying indigenous gerbera blooms in Bangladesh using image processing and neural networking techniques. The project effort aims to use computer vision and AI techniques to teach the next generation how to distinguish Bangladeshi flowers, since most young people in the city have no idea how to differentiate between traditional and desi gerbera blooms. I assessed the experiment method's credibility using my own Dataset of 3140 sample images. To detect and analyze the color of a gerbera flower, the proposed model employs a sequential grassfire algorithm in conjunction with pre-processing approaches such as noise cancellation, gray scalability, the flood-fill approach, and binarization. A convolutional neural network (CNN) and the Visual Geometry Group (VGG-16) technique were then utilized to identify and recognize the observed gerbera flower.