KAHFI, AHMAD FUAD (2022) PENDETEKSI JENIS TANAH SECARA LANGSUNG MENGGUNAKAN TEKNOLOGI TENSORFLOW BERBASIS ANDROID. Undergraduate thesis, Universitas Stikubank.
![[thumbnail of HALAMAN JUDUL]](/style/images/fileicons/application_pdf.png) |
PDF (HALAMAN JUDUL)
Download (1MB) |
|
PDF (ABSTRAK)
Download (19kB) |
|
PDF (BAB I)
Download (75kB) |
|
PDF (BAB II)
Restricted to Repository staff only Download (485kB) |
|
PDF (BAB III)
Restricted to Repository staff only Download (187kB) |
|
PDF (BAB IV)
Restricted to Repository staff only Download (2MB) |
|
PDF (BAB V)
Restricted to Repository staff only Download (62kB) |
|
PDF (DAFTAR PUSTAKA)
Download (84kB) |
|
PDF (LAMPIRAN)
Restricted to Repository staff only Download (401kB) |
Abstract
Tanah disetiap daerah di indonesia berbeda jenis dan teksturnya. Sehingga tigkat kesuburan di suatu daerah berbeda-beda. Banyak penduduk indonesia yang suka bercocok tanam, seperti halnya menanam tanaman hias di rumah. Masih banyak masyarakat di indonesia yabg terkadang masih bingung untuk memilih atau membedakan jenis tanah yang akan digunakan untuk menanam tanaman. Penelitian ini akan mendeteksi jenis tanah dan mengklasifikasinya berdasarkan presentasi tanah tersebut menggunakan teknologi TensorFlowdengan di dukung jaringan saraf tiruan Convolutional Neural Network (CNN). Dengan total 880 dataset jenis tanah dengan 8 kategori, dan setiap kategori bersisi 90 data training dan 20 data test. Aplikasi yang dihasilkan hanya berjalan di OS android untuk mendeteksi objek tanah. Hasil dari test menggunakan 20% dari data training jenis tanah dengan akurasi yang berbeda-beda setiap jenis tanah dan menghasilkan akurasi rata-rata 90%. Soil in each region in Indonesia is different in type and texture. So the level of fertility in an area is different. Many Indonesians like to cultivate crops, such as planting ornamental plants at home. There are still many people in Indonesia who are sometimes still confused about choosing or differentiating the type of soil that will be used to grow crops. This study will detect soil types and classify them based on the presentation of the soil using TensorFlow technology with the support of the Convolutional Neural Network (CNN) artificial neural network. With a total of 880 soil type datasets with 8 categories, and each category contains 90 training data and 20 test data. The resulting application only runs on the android OS to detect ground objects. The results of the test use 20% of the training data for soil types with different accuracy for each soil type and produce an average accuracy of 90%.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | SKR.I.05.01.2110 NIM.17.01.53.0122 |
| Uncontrolled Keywords: | Sistem Deteksi , DeepLearning,Convolutional Neural Network (CNN), Jenis Tanah, TensorFlow, Detection System, DeepLearning,Convolutional Neural Network (CNN), Type of soil, TensorFlow, |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
| Faculty / Institution: | Fakultas Teknologi Informasi > Program Studi Teknik Informatika |
| Depositing User: | Teteh Hayati |
| Date Deposited: | 20 May 2022 05:18 |
| Last Modified: | 20 May 2022 05:18 |
| URI: | https://eprints.unisbank.ac.id/id/eprint/8428 |
Actions (login required)
 |
View Item |
