Prototype Designing of Computer-aided Classification System for Leaf Images of Medicinal Plants
Journal Title: Journal of Biomedical Engineering and Medical Imaging - Year 2017, Vol 4, Issue 2
Abstract
The objective of the present work is to design computer-aided classification (CAC) system to discriminate among various medicinal leaves. This objective is achieved with the designing of a classification system using shape features to obtain high accuracy of end results. In order to evaluate the performance of the presented work six leaf classes were adopted and its shape features were extracted. The role of each shape feature in differentiating the leaves from one another is studied using region of convergence (ROC) curves. The combined effect of both the shape features is studied using neural network classifiers. Two neural network (NN) Classifiers were designed to classify the leaves and their performance is compared. First classifier is designed using radial bases function neural network (RBFNN) and second using feedforward backpropagation neural network (FFBNN) with a single hidden layer. A comparative study for this dataset reveals that RBFNN shows 92% classification accuracy which is 2.7% higher than that of FFBPNN.
Authors and Affiliations
Govardhan Jain, Deepti Mittal
Keywords
Related Articles
- EP ID EP296547
- DOI 10.14738/jbemi. 42.3053
- Views 56
- Downloads 0
Calculation of Talocrural Joint Axis Motion by Approximating Trochlea Tali with Conical Side Surface
The motion of the talus, the most complex and important bone in ankle motion, is determined by the geometric characteristics of the articular surface of the talocrural joint, known as the trochlea tali. Therefore, modeli...
Real and Complex Valued Ripplet-I Transform for Medical Image Denoising and Analysis of Thresholding Constants and Scales Effects
Medical image processing is an important diagnostic tool in the field of medical. Medical images might be affected by the noises that manipulate the resolution negatively during screening or transmission. These images ne...
Evolution Via EPR-Entanglement Algorithm
By quantifying EPR-generated accumulations of entangled proton qubits populating duplex microsatellite base pairs, observed as G-C → G'-C', G-C → *G-C* and A-T → *A-*T, the potential to exhibit expansion or contraction o...
Automated Pulmonary Lung Nodule Detection Using an Optimal Manifold Statistical Based Feature Descriptor and SVM Classifier
The pulmonary lung nodule is the most common indicator of lung cancer. An efficient automated pulmonary nodule detection system aids the radiologists to detect the lung abnormalities at an early stage. In this paper, an...
A Novel Two-Stage Thresholding Method for Segmentation of Malaria Parasites in Microscopic Blood Images
Developing computerized diagnostic tool for the detection of malaria infected cells in microscopic blood images can help to reduce malaria-induced mortality. Segmentation of malaria infected cells is a key step in the au...