Bezier Curve Based Medicinal Leaf Classification using Capsule Network

In this paper Cubic Bezier curve-based leaf classification system using CapsNet is proposed.This system extracts image features in terms of shape characteristics of query leaf image then the CapsNet is trained to classify the input leaf image to a particular class within a desired range of similarity.The proposed system determines nonlinear relationship between leaf image's features for more accurate similarity comparison between query leaf image and existing class images.Among existing approaches to shape feature analysis, statistical approach to extract shape features is adopted here.It works in form of control points of Bezier curves from both given query leaf image and leaf images of available database.These control points are further used to find out the Fourier Descriptors.control points and Fourier descriptors are further used to train the CapsNet for image classification in proposed system.Nowadays, multiple leaf images are generated and stored online hence, classical classification systems are crucial.The proposed system is compared with other classification systems in terms of performance and results.Comparison reveals proposed system performance is state of the art.In many parameters it outperforms other classification systems.