College of Forest Resources Publications and Scholarship

Abstract

This technical note determines the feasibility of using an InceptionV4_ResNetV2 convolutional neural network (CNN) to correctly identify hardwood species from macroscopic images. The method is composed of a commodity smartphone fitted with a 14× macro lens for photography. The end-grains of ten different North American hardwood species were photographed to create a dataset of 1709 images. The stratified 5-fold cross-validation machine-learning method was used, in which the number of testing samples varied from 341 to 342. Data augmentation was performed on-the-fly for each training set by rotating, zooming, and flipping images. It was found that the CNN could correctly identify hardwood species based on macroscopic images of its end-grain with an adjusted accuracy of 92.60%. With the current growing of machine-learning field, this model can then be readily deployed in a mobile application for field wood identification.

Publisher

Forests

First Page

298

Publication Date

3-7-2020

College

College of Engineering| College of Forest Resources

Department

Department of Sustainable Bioproducts

Research Center

Center for Advanced Vehicular Systems

Keywords

wood identification, machine-learning, smartphone, macro lens, Inception-ResNet, convolutional neural networks (CNN)

Disciplines

Forest Sciences

README.md (1 kB)
bl.rar (300865 kB)
hl.rar (299947 kB)
hk.rar (363507 kB)
am.rar (412809 kB)
oo.rar (381776 kB)
rm.rar (597585 kB)
ro.rar (416065 kB)
ss.rar (296308 kB)
wo.rar (403563 kB)
wa.rar (446119 kB)

Share

COinS