Thesis study: DEVELOPMENT OF AN ABACA (Musa textilis) FIBER GRADE CLASSIFIER USING CONVOLUTIONAL NEURAL NETWORK
Author: Neptali E. Montañez
Adviser: Jomari Joseph A. Barrera

Abstract:
This study was developed to automate the grade classification of Abaca fiber using Convolutional Neural Network (CNN) paired with an objective instrument. The abaca fiber samples were taken from Ching Bee Trading Corporation and classified by a certified Philippine Fiber Industry Development Authority (PhilFIDA) inspector. These fibers with different grades are inserted in the objective instrument and several sample images of abaca fibers were captured. In this study, 140 sample images of abaca fibers were used, which were divided into two sets: 70 images; 10 per grade, each for training and testing. The input images were then scaled to 112x112 pixels. Next, using a customized version of VGGNet-16 CNN architecture, the training set images were used for training. Then, the performance of the classifier was evaluated by computing the overall accuracy of the system with its Cohen kappa value. The classifier achieved 83% accuracy in correctly classifying the Abaca fiber grade of a sample images and attained a Cohen kappa value of 0.52 — Weak, Level of Agreement. The implementation of this study would greatly help Abaca producers and traders ensure that their Abaca fiber would be graded fairly.

Published journal article:
(ASTESJ) Automated Abaca Fiber Grade Classification Using Convolution Neural Network (CNN) https://dx.doi.org/10.25046/aj050327