IJEIS (Indonesian Journal of Electronics and Instrumentation Systems)
Not a member yet
300 research outputs found
Sort by
Vehicle Detection System Using Ultrasonic And Magnetic Fields Sensors Based On LoRa Communication
No full textNowadays, privately owned vehicles still holds the position as the staple of transportation used by the majority of people, especially in Indonesia. Along with the increasing number of vehicles, its necessary to develop infrastructure that can support its usage. One of them is the development of a parking management system. The current conventional parking system in Indonesia requires users to surround the parking area to find an empty parking space. These activity takes time for the user so it is inefficient. To resolve that problem, it needs a smart parking system that can provide information about empty parking space location directly to the users.This study discusses the design of a vehicle detection system in the implementation of outdoor smart parking. The device that used in this research is the ultrasonic sensor HC-SR04 and the HMC5883L magnetic field sensor for the detection process and uses LoRa RFM95W to communicate with the gateway device. The smart parking system designed is able to detect the presence of car in the parking area and differentiate the car from other objects with a 100% success rate percentage
Sistem Pemantauan Pertumbuhan Anggrek Berdasarkan Pengolahan Citra Digital
Get PDFGrowth monitoring and plant protectionis the major aspect of horticulture because its productivity depends on the health of the plants. Manual direct measurement methods tend to be destructive towards observed plants. In this research, a smart non-contact growth monitoring system was implemented on a chamber with orchid plants asthe objectsobserved. The images of the orchids were taken and became the input of the system to be processed to estimate the height of the plants.The contour of the orchid plant as the object was obtained and the height was calculated based on the highest and the lowest contour.The result shows that the developed system is proven to be capable of measuring orchid’s height in real-time with accuracy more than 95,7%. Thus, this system will effectively help farmers to improve the quality and the quantity of the plant’s productivity
Penggunaan Pre-trained Model untuk Klasifikasi Kualitas Sekrup
Get PDFInspeksi kualitas produk berbasis citra merupakan hal yang penting bagi industri manufaktur. Tugas tersebut sebagian besar masih dilakukan oleh manusia yang memiliki unit per hour rendah. Metode konvensional untuk inspeksi citra masih mengandalkan metode berbasis fitur, yang memiliki masalah sulitnya generalisasi dan ekstraksi fitur. Masalah tersebut diatasi dengan metode CNN, tetapi CNN membutuhkan data yang besar dan waktu training yang lama. Penggunaan pre-trained model dan augmentasi citra dapat menyelesaikan permasalahan pada metode-metode sebelumnya. Namun, belum ada penelitian yang secara lengkap meneliti dan membandingkan performa berbagai pre-trained model dan variasi augmentasi citra untuk tugas inspeksi citra kualitas produk manufaktur.Proses penelitian menggunakan dataset sekrup berjenis multi class dan binary class pada 33 jenis pre-trained model dan 8 jenis augmentasi citra. Pengujian pre-trained model menggunakan dataset gabungan seluruh jenis augmentasi citra. Model dengan akurasi tertinggi adalah EfficientNetV2-L untuk dataset multi class (97.8%) dan VGG-19 untuk dataset binary class (96.5%). Augmentasi citra dengan signifikansi tertinggi terhadap performa model adalah blur, dengan akurasi 81.1% pada multi class dan 92% pada binary class. Keseluruhan proses pengujian pre-trained model dan augmentasi citra berjalan dengan baik. Kata kunci—Inspeksi kualitas produk, Pre-trained model, Augmentasi citr
Analisis Photoplethysmography Jarak Jauh dalam berbagai Kondisi Pencahayaan
Get PDFOne of the limitations of photoplethysmography (PPG) using a contact sensor to estimate the heart rate is that the sensor must be attached directly to the patient's body. rPPG (remote-Photoplethysmography) can remotely monitor a patient's heart ratebased on an image. However, rPPG has limitations in instances where this technology is directly affected by the lighting conditions and direction of the observed subject. This study used rPPG based on the Green Channel and HSV (Hue, Saturation, and Value) color model to estimate heart rate under different lighting conditions. Analysis, computational methods,and image transformation functions are used for data selection, denoising, colormodel conversion, spectral analysis, and visualization to extract biomedical signals from inputs. The estimatedheart rate was then derivedusing spectral analysis on videostaken from an area of interest on the forehead. Compared to the ground truth, theaverage percentage error from the facial lighting tests conducted at 260 lux, 19 lux, and 11 lux for the Green Channel color modelis 0.038, 0.118, and 0.229, which is less than the HSV's error of 0.095, 0.212, and 0.247
Pembelajaran Mesin untuk Sistem Keamanan - Literatur Review
No full textSecurity systems are one of the crucial topics in the era of digital transformation. In the use of digital technology, security systems are used to ensure the confidentiality, integrity, and availability of data. Machine learning techniques can be applied to support the system's adaptability to the environment, so that prevention, detection and recovery can be carried out. Given the importance of these things, it is necessary to review the literature to find out how machine learning is applied to security systems. This paper presents a summary of 31 research papers to determine what machine learning techniques or methods are the most promising for prevention, detection and recovery. The research stages in this paper consist of 6 stages, namely: formulating research questions, searching for articles, documenting search strategies, selecting studies, assessing article quality, and extracting data obtained from articles. Based on the results of the study, it was found that the K-means method was the most promising for prevention, while for detection, SVM could be used, and for security recovery, machine learning could be implemented using NLP-based features
Kendali Stabilisasi Pesawat Tanpa Awak Sayap Tetap untuk Pendaratan Otomatis Menggunakan Fuzzy
Get PDFPesawat tanpa awak sayap tetap atau fixed wing memiliki 3 fase terbang, yaitu lepas landas, terbang jelajah, dan pendaratan. Dalam fase pendaratan dibagi menjadi 2 tahapan yaitu glideslope dan flare. Selama fase pendaratan, kestabilan wahana merupakan hal yang krusial untuk dapat melakukan pendaratan dengan selamat. Sehingga kendali yang digunakan harus mampu menstabilkan pesawat saat melakukan pendaratan.Kendali penstabil pendaratan otomatis yang digunakan pada penelitian ini adalah full-state feedback gain K yang nilainya diperoleh menggunakan metode Linear Quadratic Regulator (LQR) dan logika fuzzy. Nilai gain K yang didapat akan dikonversikan terlebih dahulu menjadi Pulse Width Modulation (PWM) dan akan digunakan sebagai nilai masukkan pada sistem. Sinyal PWM tersebut akan mengendalikan kecepatan putar brushless motor dan sudut defleksi Servo. Ketika memasuki mode auto, wahana akan menjaga kestabilan pada sudut roll dan sudut serang wahana selama memasuki pendaratan otomatis. Nilai sudut acuan roll merupakan 0˚ agar wahana tetap stabil. Berdasarkan penelitian ini, wahana mampu mempertahankan kestabilan pada sumbu roll selama melakukan pendaratan otomatis. Selain itu didapatkan nilai risetime sebesar 0,3 detik serta steady state error sebesar 2,71 derajat dan tidak ditemukannya overshoo
Klasifikasi Suara Paru-Paru Berdasarkan Ciri MFCC
Get PDFThe lungs are an important organ in the human respiratory system, which functions to exchange carbon dioxide from the blood with oxygen in the air. Detection of respiratory disorders and lung disorders can be done in various ways; view medical records, physical examination, detection by x-ray and also auscultation of breathing. Digital signal processing can be used as a method to detect lung disorders based on the sound produced. In this study, lung sounds were classified into normal, crackle, wheeze, and crackle-wheeze classes using the Mel Frequency Cepstral Coefficient (MFCC) and Convolutional Neural Network (CNN) methods.Observations were made by varying the MFCC feature extraction using MFCC 8 and 13 coefficients, the number of frames are 50 and 60, and the width of the frames used was 0,1, 0,15 and 0,2 seconds. The result of feature extraction is then applied to the CNN classification system, and the confusion matrix is used to get the accuracy and precision values. The highest accuracy and precision values were obtained at 71,85% and 65,70% on the MFCC 13 coefficient with an average of 71,18%. Based on these results, the system that has been created can classify normal lung sounds, crackle, wheeze and crackle-wheeze quite well
Prediksi Diabetes Berdasarkan Pengukuran Mean Amplitude Glycemic Excursion (MAGE) Menggunakan Naïve Bayes
Get PDF The mean amplitude of glycemic excursions (MAGE) is an important indicator in the assessment of glycemic variability (GV) which is used as a reference for continuous blood glucose control. In this case, quantitative considerations in monitoring blood sugar in diabetes are very important for diagnosis and then proceed with clinical treatment. This study focuses more on strengthening the training and testing data processing system and reducing the independent variables that occur during the classification process. To support this purpose, this study uses Cross Validation as a training and testing data processing with the number of K-Fold is 10 and Naïve Bayes as a classification method. The resulting accuracy is 93% which is an increase from previous studies with an RMSE value (error value) of 0.267. It was concluded that patients in the pre-diabetic and diabetic groups tend to have more varied blood glucose values than patients from the normal class
Music Genre Identification Using SVM and MFCC Feature Extraction
No full text Indonesia is a very diverse country because it has a vast territory and is occupied by millions of people from various tribe. Therefore, traditional music in Indonesia is also diverse because each region has its own culture and art. In this study, the author used the Support Vector Machine(SVM) pattern recognition to identify the Indonesian traditional music genre. This genre identification system is able to produce an accuracy of 83% using MFCC.Keywords : traditional music identification, Mel Frequency Cepstral Coefficient, Support Vector Machine
Penempatan Posisi Transduser Ultrasonik Pada Penampang Pipa untuk Pengukuran Laju Aliran Fluida
Get PDFFluid flow rate measurement is important in industries, especially determining fluid flow rate. This process requires a good level of precision and accuracy because it refers to each volumetric's price or custody transfer processor. Many devices are used to measure flow rates, but from some devices, ultrasonic flowmeters are considered, which have more advantages than others. Ultrasonic flowmeters also have some problems, especially in installation, so this research aims to simulate the position of path configuration. The method refers to the weighting process of multi-path configuration and the simulation of track performance, which includes three-factor, hydrodynamic (H), orientation sensitivity (S) and orientation range (T). Each trajectory pattern is rotated 1ᴼ at each angle. In addition, there are also parameter functions that are used to image the profile. The test uses 7 path configurations, so an ideal form is obtained to be implemented. After multiplying weighting factors, the obtained value of hydrodynamic (H) for Area weighting method (1.002), the best value 1. Orientation sensitivity (S), with Area weighting method (0.019), the best result is 0. Meanwhile, with orientation range (T) 1%, with Area weighting method (163,2), the best value is 180