Claim Missing Document
Check
Articles

Found 9 Documents
Search

Desain Sensor Suhu dan Kelengasan Tanah untuk Sistem Kendali Budidaya Tanaman Cabai (Capsicum annuum L.) Sugeng Triyono; Mareli Telaumbanua; Yessi Mulyani; Titin Yulianti; Muhammad Amin; Agus Haryanto
agriTECH Vol 38, No 4 (2018)
Publisher : Faculty of Agricultural Technology, Universitas Gadjah Mada, Yogyakarta, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (369.33 KB) | DOI: 10.22146/agritech.29095

Abstract

Cultivation crop is influenced by soil, water, climate, and crop properties. Air temperature is one of climate parameters which is considered for plant growing. Soil moisture represents soil and water factors and it generally plays an important role in crop cultivation. A crop requires soil moisture and air temperature for an optimum growth. a control system is proposed to create an optimum air temperature and soil moisture to support plant growth. The aim of this study was to design a precision measurement instrument, a control system that is able to control microclimate (air temperature and soil moisture) for optimal growth of chili (Capsicum annuum L.) crops. A design of environmental control was applied by using sensors for air temperature and soil moisture. Microcontrollers were connected to sensors with the water pump actuator and the irrigation pump through a relay module and a TIP122 transistor. The accuracy of DHT 22 temperature sensors and soil moisture sensors were calculated based on the approximate coefficient of determination, and the total errors of each sensor. The actuator performance in this design included the response rates and the duration of the working time. The performance tests were conducted 3 times without using chili plants. The coefficient of determination (R²) of temperature sensor 1, temperature sensor 2 and temperature sensor 3 were 0.999, 0.999, and 0.999, respectively. The total errors of the three sensors were -0.071 ºC, -0.085 ºC, and 0.014ºC, respectively. The coefficient of determination (R ²) of the soil moisture sensor 1, the soil moisture sensor 2, and the soil moisture sensor were 0.888, 0.8401, and 0.8963, respectively. The mean total errors for these three types of ranging sensors were -0.2204%, -0.0952% and -2.8049%, respectively.
Studi Perbandingan Pengenalan Karakter Aksara Lampung Dengan Metode Deteksi Tepi Roberts dan Sobel Halim Abdillah Sholeh; Yessi Mulyani; Hery Dian Saptama
Inovasi Pembangunan : Jurnal Kelitbangan Vol 6 No 03 (2018): December 2018
Publisher : Balitbangda Provinsi Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (589.974 KB)

Abstract

The culture of Lampung script is now rarely used and will get extinct. In order to preserve the culture, this digital research about Lampung script with image processing method and pattern recognition is conducted. This research is divided into two stages, training and testing. Training stage conducted image processing and training the artificial neural network (ANN) backpropagation. One past of image processing is edge detection. The edge detection method is used to analyze the Lampung script character to find the line pattern of the letters. The image processing methods used in image edge detection are Roberts edge detection method and Sobel edge detection method. The result of image processing will get into testing stage. Sobel method has better performance on recognizing. Lampung script character, in which the result from testing Lampung script character gives 28.5 % of error for Roberts detection method and 14.5 % of error for Sobel detection method.
IDENTIFIKASI KEMATANGAN BUAH KOPI MENGGUNAKAN JARINGAN SYARAF TIRUAN LEARNING VECTOR QUANTIZATION Wahyu Aji Pulungan; Yessi Mulyani; Wahyu Eko Sulistiono
BAROMETER Vol 4 No 2 (2019): Barometer
Publisher : Fakultas Teknik, Universitas Singaperbangsa Karawang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35261/barometer.v4i2.1834

Abstract

Buah kopi yang memiliki kualitas yang baik merupakan buah kopi yang berwarna merah, namun petani secara umum masih menggunakan cara panen yang konvensional sehingga buah yang dipanen masih banyak yang berwarna hijau atau kuning. Penelitian ini menggunakan Jaringan Syaraf Tiruan (JST) metode Learning Vector Quantization (LVQ) untuk mempelajari pola dan mengidentifikasi kematangan buah kopi dengan memanfaatkan fitur warna pada buah kopi. Data input menggunakan akuisisi sekumpulan buah kopi dengan berbagai tingkat kematangan yang kemudian dilakukan ekstraksi ciri warna RGB untuk diambil nilai rata-rata. Data yang digunakan memiliki ukuran 1300 x 1000 piksel  berformat .jpg untuk pelatihan JST. Berdasarkan hal tersebut, dilakukan pengujian jumlah citra pelatihan dengan membandingkan antara 10, 12, 16, dan 20 buah citra pelatihan untuk menentukan jumlah input terbaik. Kemudian dilakukan juga pengujian confusion matrix untuk menguji tingkat keandalan dan error dari sistem yang dibuat, serta pengujian black box terhadap GUI yang dibuat. Pengujian dilakukan menggunakan GUI yang berjalan dengan baik, berdasarkan hasil pengujian jumlah citra terbaik untuk pelatihan adalah 16 buah dengan akurasi sebesar 100% tanpa error pada setiap pengujian yang dilakukan.
Pengembangan Sistem Informasi Pendaftaran Seminar Akademik Di Jurusan Teknik Elektro Fakultas Teknik Universitas Lampung Menggunakan Metode Rapid Application Development (RAD) Hery Dian Septama; Yessi Mulyani; Mahendra Pratama; Nyoman Herman Ardike
BAROMETER Vol 5 No 1 (2020): Barometer
Publisher : Fakultas Teknik, Universitas Singaperbangsa Karawang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35261/barometer.v5i1.2252

Abstract

Jurusan Teknik Elektro Universitas Lampung memiliki sistem pendaftaran seminar akademik dengan proses administratif manual, terdapat kekurangan dari alur pendaftaran yang sudah ada seperti minimnya informasi yang tersaji dengan cepat dengan pengolahan data tidak terpusat di satu database menyulitkan informasi status pendaftaran, sistem seminar akademik dapat memudahkan proses administratif agar dapat diakses oleh pengguna secara bersamaan melalui browser dengan jaringan internet. Tujuan dari penelitian ini adalah untuk mengembangkan sistem informasi pendaftaran seminar akademik di Jurusan Teknik Elektro Universitas Lampung yang dapat memasukkan, mengelola, serta menampilkan data dengan baik agar menghasilkan informasi yang akurat dan tersaji dengan cepat. Metode RAD (Rapid Application Development) digunakan sebagai panduan untuk membangun sistem informasi pendaftaran seminar akademik. Pengujian dilakukan di setiap fungsi use case pada sistem, kemudian menguji sistem menggunakan metode UEQ, kepada 4 admin TU Jurusan Teknik Elektro, 22 mahasiswa Jurusan Teknik Elektro, 4 Staff IT UPT TIK Universitas Lampung, hasil pengujian sistem pendaftaran seminar akademik memenuhi fungsi yang dibutuhkan 4 aktor, dari 6 kategori menggunakan metode UEQ, benchmark yang didapat adalah 5 point mendapat nilai sangat baik dan 1 point mendapatkan nilai baik.
Implementasi Long Short-Term Memory pada Chatbot Informasi Akademik Teknik Informatika Unila Puput Budi Wintoro; Hilmi Hermawan; Mona Arif Muda; Yessi Mulyani
EXPERT: Jurnal Manajemen Sistem Informasi dan Teknologi Vol 12, No 1 (2022): June
Publisher : Universitas Bandar Lampung (UBL)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36448/expert.v12i1.2593

Abstract

The times have made technology develop rapidly and rapidly, one of which is artificial intelligence. According to an Accenture survey, s can help organizations with up to 30% operational costs and users enjoy having 24/7 instant access to the answers they order. However, there are many organizations and companies that have not used s to facilitate business, one of which is the Informatics Engineering Study Program, University of Lampung (PSTI Unila). The purpose of this research is to able to apply the Long Short-Term Memory (LSTM) algorithm to the academic information at PSTI Unila and demonstrate the accuracy of the algorithm. Broadly speaking, this research consists of 4 stages: data collection and exploration, data preparation, building model and training, and testing the model. The model is tested using the user validation method which gets an accuracy of 99%. With these accurate results, this model is quite feasible to be used as a place for students to ask questions about PSTI Unila academic information.
PENERAPAN APLIKASI DIGITAL PADA JUAL BELI SAMPAH PLASTIK DAN PRODUKSI KOMPOS DARI LIMBAH ORGANIK SKALA RUMAH TANGGA DI KELURAHAN RAJABASA, BANDAR LAMPUNG Yessi Mulyani
Jurnal Pengabdian Kepada Masyarakat Sakai Sambayan Vol 6 No 2 (2022)
Publisher : Lembaga Penelitian dan Pengabdian Universitas Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23960/jss.v6i2.332

Abstract

Kegiatan pengabdian ini bertujuan untuk mengedukasi masyarakat tentang pemilahan dan pengelolaan sampah organik menjadi kompos serta membuat aplikasi digital untuk mempermudah jual beli sampah yang dapat didaur ulang. Kegiatan ini dilakukan di Kelurahan Rajabasa, Bandar Lampung dengan mitra adalah Komunitas Sahabat Lingkungan (KSL). Metode yang dilakukan dalam kegiatan pengabdian ini meliputi sosialisasi, pelatihan dan penggunaan aplikasi digital jual beli sampah dengan menggunakan smartphone. Dalam sosialisasi dijelaskan secara singkat tentang pemanfaatan dan penanganan sampah, khususnya sampah organik. Pelatihan pembuatan kompos dan penggunaan aplikasi digital jual beli sampah diberikan kepada mitra KSL. Permasalahan tentang pengelolaan sampah secara umum dan khususnya yang dihadapi mitra adalah belum memiliki pengetahuan dan peralatan untuk proses pembuatan kompos. Dengan kegiatan pengabdian Unggulan ini mitra KSL dapat menggunakan peralatan untuk membuat kompos, dan dapat menggunakan aplikasi digital jual beli sampah untuk mempermudah mereka. Kegiatan pengabdian ini telah memberikan keuntungan bagi mitra. Kata kunci: sampah organik; kompos;aplikasi digital jual beli sampah
Comparison Study of Convolutional Neural Network Architecture in Aglaonema Classification Yessi Mulyani; Dzihan Septiangraini; Meizano Ardhi Muhammad; Gigih Forda Nama
International Journal of Electronics and Communications Systems Vol 2, No 2 (2022): International Journal of Electronics and Communications System
Publisher : Raden Intan State Islamic University of Lampung, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1163.598 KB) | DOI: 10.24042/ijecs.v2i2.13694

Abstract

Convolutional Neural Network (CNN) is very good at classifying images. To measure the best CNN architecture, a study must be done against real-case scenarios. Aglaonema, one of the plants with high similarity, is chosen as a test case to compare CNN architecture. In this study, a classification process was carried out on five classes of Aglaonema imagery by comparing five architectures from the Convolutional Neural Network (CNN) method: LeNet, AlexNet, VGG16, Inception V3, and ResNet50. The total dataset used is 500 image data, with the distribution of training data by 80% and test data by 20%. The segmentation process is performed using the Grabcut algorithm by separating the foreground and background. To build a model for CNN architecture using Google Colab and Google Drive storage. The results of the tests carried out on five classes of Aglaonema images obtained the best accuracy, precision, and recall results on the Inception V3 architecture with values of 92.8%, 93%, and 92.8%. The CNN architecture has the highest level of accuracy in classifying aglaonema plant types based on images. This study seeks to close research gaps, contribute to the field of research, and serve as a platform for primary prevention research.
Comparison Study of Convolutional Neural Network Architecture in Aglaonema Classification Yessi Mulyani; Dzihan Septiangraini; Meizano Ardhi Muhammad; Gigih Forda Nama
International Journal of Electronics and Communications Systems Vol 2, No 2 (2022): International Journal of Electronics and Communications System
Publisher : Raden Intan State Islamic University of Lampung, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24042/ijecs.v2i2.13694

Abstract

Convolutional Neural Network (CNN) is very good at classifying images. To measure the best CNN architecture, a study must be done against real-case scenarios. Aglaonema, one of the plants with high similarity, is chosen as a test case to compare CNN architecture. In this study, a classification process was carried out on five classes of Aglaonema imagery by comparing five architectures from the Convolutional Neural Network (CNN) method: LeNet, AlexNet, VGG16, Inception V3, and ResNet50. The total dataset used is 500 image data, with the distribution of training data by 80% and test data by 20%. The segmentation process is performed using the Grabcut algorithm by separating the foreground and background. To build a model for CNN architecture using Google Colab and Google Drive storage. The results of the tests carried out on five classes of Aglaonema images obtained the best accuracy, precision, and recall results on the Inception V3 architecture with values of 92.8%, 93%, and 92.8%. The CNN architecture has the highest level of accuracy in classifying aglaonema plant types based on images. This study seeks to close research gaps, contribute to the field of research, and serve as a platform for primary prevention research.
Pengembangan Aplikasi Jual Beli Sampah Daur Ulang Menggunakan Framework Multiplatform Ardi Ragil Saputra; Yessi Mulyani
Electrician : Jurnal Rekayasa dan Teknologi Elektro Vol. 17 No. 2 (2023)
Publisher : Department of Electrical Engineering, Faculty of Engineering, Universitas Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23960/elc.v17n2.2480

Abstract

Environmental pollution caused by waste is increasingly worrying if there is no effort to overcome it. The lack of behavior in managing recyclable waste in society is one of the biggest factors in waste pollution in Indonesia. There are many recycling waste management facilities in Indonesia, but the location of waste management facilities is unevenly distributed, and there are price differences at each location, making the community less interested in managing recyclable waste. In this study, a buying and selling system for recyclable waste was developed based on Android devices as a buying and selling tools and a web application for managing user buying and selling data. The focus of this research is to build a buying and selling system for recyclable waste by integrating a multi-platform framework, namely using the Laravel framework for the web admin side and using the Flutter framework for the android application side. Then, to analyze the review of the users for the android application, this study used the System Usability Scale (SUS). The result of this study is an Android-based buying and selling application for recyclable waste and a web admin application for managing the data of the Android application. From the SUS analysis, the application obtained a score of 70, which is acceptable, with the note that the application built is already good, but still needs to be improved for transactions using an electronic wallet (e-wallet).