DETEKSI HASIL TANAMAN MELALUI CITRA THERMAL MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORK (CNN)

JULIANSYAH, AGUS and Siswanti, Sri Desy and Ubaya, Huda (2020) DETEKSI HASIL TANAMAN MELALUI CITRA THERMAL MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORK (CNN). Undergraduate thesis, Sriwijaya University.

[img] Text
RAMA_56201_09011181320034.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (1MB) | Request a copy
[img] Text
RAMA_56201_09011181320034_TURNITIN.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (9MB) | Request a copy
[img]
Preview
Text
RAMA_56201_09011181320034_0007127410_0216068101_01_front_ref.pdf - Accepted Version
Available under License Creative Commons Public Domain Dedication.

Download (401kB) | Preview
[img] Text
RAMA_56201_09011181320034_0007127410_0216068101_02.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (664kB) | Request a copy
[img] Text
RAMA_56201_09011181320034_0007127410_0216068101_03.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (791kB) | Request a copy
[img] Text
RAMA_56201_09011181320034_0007127410_0216068101_04.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (717kB) | Request a copy
[img] Text
RAMA_56201_09011181320034_0007127410_0216068101_05.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (192kB) | Request a copy
[img] Text
RAMA_56201_09011181320034_0007127410_0216068101_06_ref.pdf - Bibliography
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (196kB) | Request a copy
[img] Text
RAMA_56201_09011181320034_0007127410_0216068101_07_lamp.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (185kB) | Request a copy

Abstract

Turmeric (Curcuma longa) is one of the most important spices which is believed to have originated from India. In computer vision, there are several problems including object detection and image classification with thermal imaging techniques, which are invasive methods that are increasingly being studied for quality evaluation of agricultural and food products. In this study, discussing a plant yield by looking at the color of the plant using a Thermal camera to get an image of turmeric, by heating the plants at the same time. In this study also used the ROI (Region Of Interest) and CNN (Convolutional Neural Network) methods with the final results obtained on the detection of Turmeric Image using a Thermal Camera divided into 3 times. 1 day turmeric image 34.5c, 4 day turmeric image 32.5c and Image of turmeric 7 days 31.8c by getting the value of training results using the CNN (Convolutional Neural Network) method on Image Turmeric 1 with a value of 75.72% Through the training process by getting the results of the model and label, then the model is tested or predicted, the results of turmeric 1 : 75.72%, turmeric 4: 23.92% and turmeric 7: 0.37% So the biggest value is turmeric 1, so the image comes out and the value of turmeric1: 75.72%.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Object Detection, Thermal Image, ROI (Region Of Interest), HSV (Hue Saturation Velue), CNN (Convolutional Neural Network)
Subjects: T Technology > T Technology (General) > T1-995 Technology (General)
T Technology > TR Photography > TR250-265 Cameras
T Technology > TR Photography > TR510-545 Color photography
Divisions: 09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User: Mr Agus Juliansyah
Date Deposited: 21 Jan 2021 06:50
Last Modified: 21 Jan 2021 06:50
URI: http://repository.unsri.ac.id/id/eprint/40532

Actions (login required)

View Item View Item