Mobile Leaf Classification Application Utilizing a Convolutional Neural Network

ASU Author/Contributor (non-ASU co-authors, if there are any, appear on document)
Timothy Jassmann (Creator)
Institution
Appalachian State University (ASU )
Web Site: https://library.appstate.edu/
Advisor
Rahman Tashakkori

Abstract: Plant classification is an important task in biological research. However, plant classification is a complex task that very few biologists are qualified experts to conduct. Therefore, an application to assist in this task would be extremely useful for biology students, researchers, and enthusiasts.A significant amount of research has been done for the task of classifying plants based upon images of their leaves; however, all of that research has utilized images of single leaves on a white background for classification to allow easy extraction of shape features. This is not realistic for field work since a natural picture of a leaf will have a complex background.This thesis applies a convolutional neural network to the problem in order to allow classification of images with natural backgrounds. A mobile application is built that can run this neural network on images taken by the device’s camera. This tool can be used to assist in complex plant classification tasks anywhere as long as they have a mobile device with them.

Additional Information

Publication
Thesis
Jassmann, T.J. (2015). Mobile Leaf Classification Application Utilizing a Convolutional Neural Network. Unpublished master's thesis. Appalachian State University, Boone, NC.
Language: English
Date: 2015
Keywords
Computer Vision, Image Processing, Convolutional Neural Network, Mobile Application, Artificial Intelligence ,

Email this document to