Design and development of aroid plant defect classifier using convolutional neural network / by Mariel G. Bustamante and Jacob D. Mojares.

By:

Bustamante, Mariel G [author]

Contributor(s):

Material type: Text

TextLanguage: English Publication details: Indang, Cavite : Cavite State University- Main Campus, 2022.Description: xv, 72 pages : illustrations ; 28 cmContent type:

text

Media type:

unmediated

Carrier type:

volume

Subject(s):

DDC classification:

621.367 B96 2022

Online resources:

Click here to view the Abstract and Table of Contents

Production credits:

College of Engineering and Information Technology (CEIT)

Abstract: BUSTAMANTE, MARIEL G., and MOJARES, JACOB D. Design and Development of Aroid Plant Defect Classifier using Convolutional Neural Network. Undergraduate Design Project. Bachelor of Science in Electronics Engineering. Cavite State University, Indang, Cavite. August 2022. Adviser: Dr. Edwin R. Arboleda. The general objective of the study was to design and develop an Aroid Plant Defect Classifier using Convolutional Neural Network. It was intended to be portable and rechargeable. It was targeted to be capable of classifying defects of a Black Velvet "Alocasia reginula" plant based on its leaves to avoid severe damage due to improper watering. The design project was composed of Raspberry Pi 3B+ which handled all the processes the device had to execute. This microcomputer was used to interface other components of the device such as the Camera, Battery, and Touchscreen Display. The back-end system of this device was a trained Convolutional Neural Network Model with an Accuracy of 89.50% while the front-end system was a Graphical User Interface displayed on the Touchscreen display when it turned ON. The project was tested by the researchers at a farm in Dasmariñas, Cavite, where the device had accuracy values of 90.48%, 87.62%, and 91.43%, and precision values of 87.88%, 78.75%, and 88.24% in classifying Normal, Underwatered, and Overwatered leaves, respectively, using a confusion matrix. The evaluation was conducted at the Engineering Science Building, Cavite State University, Indang, Cavite with the Electronics Engineer Professors. Based on the results of the evaluation, the design project succeeded in achieving its objectives. The overall performance was considered satisfactory. The cost of the construction of the system amounted to P11,176.00.

Average rating: 0.0 (0 votes)

Holdings
Item type	Current library	Collection	Call number	Materials specified	URL	Status	Notes	Date due	Barcode
Theses / Manuscripts	Ladislao N. Diwa Memorial Library Theses Section	Non-fiction	621.367 B96 2022 (Browse shelf(Opens below))		Link to resource	Room use only	DP-753		00081854

Design Project (Bachelor of Science in Electronics and Communications Engineering) Cavite State University.

Includes bibliographical references.

College of Engineering and Information Technology (CEIT)

BUSTAMANTE, MARIEL G., and MOJARES, JACOB D. Design and Development of
Aroid Plant Defect Classifier using Convolutional Neural Network. Undergraduate Design
Project. Bachelor of Science in Electronics Engineering. Cavite State University, Indang,
Cavite. August 2022. Adviser: Dr. Edwin R. Arboleda.
The general objective of the study was to design and develop an Aroid Plant Defect
Classifier using Convolutional Neural Network. It was intended to be portable and rechargeable.
It was targeted to be capable of classifying defects of a Black Velvet "Alocasia reginula" plant
based on its leaves to avoid severe damage due to improper watering. The design project was
composed of Raspberry Pi 3B+ which handled all the processes the device had to execute. This
microcomputer was used to interface other components of the device such as the Camera,
Battery, and Touchscreen Display. The back-end system of this device was a trained
Convolutional Neural Network Model with an Accuracy of 89.50% while the front-end system
was a Graphical User Interface displayed on the Touchscreen display when it turned ON. The
project was tested by the researchers at a farm in Dasmariñas, Cavite, where the device had
accuracy values of 90.48%, 87.62%, and 91.43%, and precision values of 87.88%, 78.75%, and
88.24% in classifying Normal, Underwatered, and Overwatered leaves, respectively, using a
confusion matrix. The evaluation was conducted at the Engineering Science Building, Cavite
State University, Indang, Cavite with the Electronics Engineer Professors. Based on the results
of the evaluation, the design project succeeded in achieving its objectives. The overall
performance was considered satisfactory. The cost of the construction of the system amounted
to P11,176.00.

Submitted to the University Library 09/01/2022 DP-753

Click on an image to view it in the image viewer