IMPLEMENTASI CONVOLUTIONAL NEURAL NETWORK MENGGUNAKAN TENSORFLOW UNTUK MENDETEKSI PENYAKIT PADA DAUN TANAMAN MELON

Winarto, Winarto (2022) IMPLEMENTASI CONVOLUTIONAL NEURAL NETWORK MENGGUNAKAN TENSORFLOW UNTUK MENDETEKSI PENYAKIT PADA DAUN TANAMAN MELON. Skripsi thesis, institut informatika dan bisnis darmajaya.

[img] Text
00_halaman_judul.pdf
Download (128kB)
[img] Text
01_halaman_pernyataan.pdf
Download (614kB)
[img] Text
02_halaman_persetujuan.pdf
Download (1MB)
[img] Text
03_hal_pengesahan.pdf
Download (969kB)
[img] Text
04_Abstrak.pdf
Download (81kB)
[img] Text
05_daftar_isi.pdf
Download (87kB)
[img] Text
06_bab_1.pdf
Download (150kB)
[img] Text
07_bab_2.pdf
Download (1MB)
[img] Text
08_bab_3.pdf
Download (1MB)
[img] Text
09_bab_4.pdf
Download (8MB)
[img] Text
10_bab_5.pdf
Download (65kB)
[img] Text
11_daftar_pustaka.pdf
Download (97kB)
[img] Text
12_lampiran.pdf
Download (1MB)

Abstract

The melon plant is a plant that belongs to the gourd category. According to data taken from the BPS site, the growth of melons experienced a significant decrease in yields. This decline was due to climatic conditions, pests, and diseases. This study considered three types of melon diseases originating from leaves, specifically: fusarium wilt, caterpillars, and aphids. To classify them, we used a convolutional neural network technique to identify objects of colour and contour of an image. The goal of this research is to create a CNN model that can classify leaf diseases of melon. Also, we used the machine learning life cycle method as a stage of developing the model. It started by collecting data with a total of 250 images, cleaning them, giving the labelling, and dividing the data into test data and training data with the proportion of 80% and 20% sequentially. Additionally, we made CNN architecture, trained it, evaluated the models, and embedded this feature into the Tani Cerdas application. The results of this study found that the accuracy of the model was 78%. Then, the Tani Cerdas application showed that the disease detection feature using the CNN model has succeeded in detecting leaf diseases of melon plants.

Item Type: Thesis (Skripsi)
Subjects: Ilmu Komputer
eSkripsi
Divisions: Skripsi/TA & PKPM/KP - Fakultas Ilmu Komputer > Teknik Komputer
Depositing User: Winarto _ _
Date Deposited: 01 Jan 2023 23:23
Last Modified: 01 Jan 2023 23:23
URI: http://repo.darmajaya.ac.id/id/eprint/10203

Actions (login required)

View Item View Item