Sugondo Hadiyoso

This research develops a virtual mouse control system that uses real-time hand gesture recognition implemented on an ESP32-CAM–based Internet of Things (IoT) platform. By leveraging OpenCV for image processing, the system translates hand gestures into corresponding mouse actions, including cursor movement, clicking, and scrolling. The study evaluates system performance under different lighting conditions and Wi-Fi speeds. Results show that higher Wi-Fi speeds significantly reduce latency, enabling smoother real time gesture recognition and high definition video output, while lower speeds lead to noticeable delays and reduced accuracy. The system successfully enables remote cursor control through camera captured hand gestures, supporting functions such as left click, right click, scrolling, and dragging. In latency tests performed with an internet speed of approximately 60 Mbps, the system achieved an average delay of about 50 milliseconds. Under optimal lighting conditions with minimal background interference, it accurately tracked hand movements and recognized gestures such as pointing, clicking, dragging, and scrolling in real time, achieving an accuracy rate of 95%. Despite its lower resolution compared to conventional webcams, the ESP32-CAM proves to be an effective solution for virtual mouse control, particularly in scenarios where high-resolution imaging is unnecessary. Its IoT capabilities support remote operation, allowing users to control the virtual mouse from a distance as long as both the ESP32-CAM and the computer remain connected. Overall, the findings highlight the ESP32-CAM based IoT platform as a viable alternative for gesture based interaction in real applications, although further enhancements are needed to improve performance in challenging environments.

Sleep apnea is a sleep disorder that occurs when breathing is repeatedly interrupted during sleep. This condition can lead to various serious health problems if left untreated, such as high blood pressure, poor sleep quality, and difficulty concentrating. Sufferers often do not realize they have sleep apnea because it occurs during sleep. Generally, diagnosis is made through interviews with the patient and their family to identify common symptoms such as snoring, and then confirmed through physical examination and Polysomnography (PSG). Since sleep apnea is related to respiratory activity that correlates with changes in cardiac activity, electrocardiogram (ECG) examination during sleep serves as an alternative diagnostic method. Therefore, this study presents a comparative analysis of deep learning models for detecting sleep apnea from spectrogram-based ECG representations. The raw ECG signals were transformed into spectrograms and then saved as images for classification into normal and abnormal categories. Deep learning (DL) methods were applied for the classification of normal and sleep apnea ECGs. EfficientNet, MobileNetV2, DenseNet, AlexNet, and VGG16 were used to evaluate the performance of the proposed method and identify the best-performing model. The evaluation results show that EfficientNet achieved the highest performance with an accuracy of 91.01%, precision of 90.70%, recall of 95.76%, and an F1-score of 92.61%. EfficientNet outperformed the other evaluated models in this study. By utilizing a spectrogram-based approach combined with a scalable architecture, the method demonstrates competitive accuracy for sleep apnea detection. Exploring other approaches to further improve accuracy remains an interesting direction for future research

Analyzing the impact of sports activity intensity on muscle capacity through integrated biosensor technology

Online games have become a popular form of entertainment, particularly for relieving stress, and the rise in online gaming has led to an increase in problematic gaming behaviors. Excessive use of the internet for gaming has raised concerns about its neurophysiological impact, particularly on cognitive and emotional functions. Electroencephalogram signal and Event-Related Potential analysis are valuable tools for monitoring these effects. Given the vast amount of features that can be extracted from EEG signals, it is crucial to apply efficient feature selection methods to identify the most informative ones. This study utilizes the Go/No-Go Association Task combined with the recording of 16-channel EEG signals, chosen as the data-recording method to observe the response of individuals who are problematic online gamers to several stimulus themes. In this context, metaheuristic algorithms like Genetic Algorithm, Ant Colony Optimization, and Particle Swarm Optimization are employed to enhance feature selection. A hybrid approach, combining one of these methods with Binary Stochastic Fractal Search is proposed to improve classification accuracy and optimize feature selection. The results demonstrate that the hybridization of the best algorithm with B-SFS successfully selects the optimal features, achieving perfect classification performance, with an accuracy, sensitivity, and specificity of 1.00 for all respondents. This emphasizes the effectiveness of B-SFS, particularly its diffusion process, where Gaussian distribution facilitates the search for the best solution, thereby improving the reliability of feature selection for detecting problematic gaming behavior.

Compressive Sensing (CS) technique in image compression represents efficient signal which offering solutions in image classification where the resources are constrained especially on a large image processing, storage resource, and computing performance. Compressive learning (CL) is a framework that integrates signal acquisition via compressed sensing (CS) and machine/deep learning for inference tasks directly on a small number of measurements, On the other hand, in real-world high-resolution (HR) data, where the image dataset is very limited CL, has the drawback of reduced accuracy under conditions of aggressive compression ratio. Here, a reconstruction method is necessary to maintain high levels of accuracy. To address this, we proposed a framework Deep Learning (DL) and Compressive Sensing that processing a small dataset of 92 images maintaining high accuracy. The framework developed in this paper employs processing sensing matrix A in compressive sensing with two transformation methods: DCT CL with Multi Neural Networks and the SVD method with GoogleNet framework. To maintain the same computation efficiency as DCT Compressive learning, SVD with GoogleNet framework provides a solution for object recognition, achieving accuracy values ranging from 89.47% to 63.15% for compression ratios of 3.97 to 31.75. This performance shows a linear tendency concerning the PSNR level, an index of signal reconstruction quality, and demonstrates an efficient process in the S matrix.

Obstructive sleep apnea (OSA) is a respiratory obstruction that occurs during sleep and is often known as snoring. OSA is often ignored even though it can cause cardiovascular problems. Early diagnosis is needed for prevention towards worse complications. OSA clinical diagnosis can use polysomnography (PSG) while the patient is sleeping. The PSG examination includes calculating total apnea plus hypopnea every hour during sleep. However, PSG examination tends to be high cost, takes a long time, and is impractical. Since OSA is related to breathing and heart activity, the electrocardiogram (ECG) examination is an alternative tool in OSA analysis. Therefore, this study proposes OSA detection on single lead ECG using one dimensional (1D)-convolutional neural network (CNN). The proposed CNN architecture consists of 4 convolutional layers, 4 pooling layers, 1 dropout layer, 1 flatten layers, 2 dropout layers, 1 dense layer with rectified linear unit (ReLU) activation function, and 1 dense layer with SoftMax activation function. The proposed method was then tested on the ECG sleep apnea dataset from PhysioNet. The proposed model produces an accuracy of 92.69% with the average pooling scenario. The proposed method is expected to help clinicians in diagnosing OSA based on ECG signals.

Statistical calculations on signals commonly used in feature extraction. In software processing, statistical computation is an easy task. However, providing a computer requires high costs for simple statistical processing. Another consideration is the need for implementation with real-time and portable processing. Therefore, an alternative device is needed, one of which is the field programmable gate array (FPGA). FPGA is a logic circuit board that can be reconfigured according to computing needs. FPGA can also be used as a prototyping of electronic chips. However, implementing statistical formulas in FPGA is interesting in developing its architecture. Therefore, this research proposes a logic circuit design that can be used for first-order statistical calculations. Statistical parameters include the mean, variance, standard deviation, skewness, and kurtosis. The validation test was performed on the electrocardiogram (ECG) signal series and compared with manual calculations. Validation shows that the mean and variance has very high accuracy with an average error of less than 0.06%.

The electroencephalogram (EEG) examination provides information on the brain’s electricity, especially in cases of epilepsy. Since the characteristics of EEG signals are nonlinear and nonstationary, visual inspection becomes very difficult. To overcome this problem, digital EEG signal processing was developed. Automatic epileptic EEG recognition is an area of interest on which much research focuses. The complexity approach to EEG signal analysis is interesting to be used as feature extraction, referring to the nonlinear characteristics of the signal. This study proposed an automatic epileptic EEG classification method based on the multiscale Hjorth descriptor measurement. EEG signals consisting of normal, interictal, and seizure (ictal) were simulated. The signal is scaled into new signals using the coarse-grained procedure on a scale of 1–20. Then, the Hjorth parameter which consists of activity, mobility, and complexity is calculated on the new signal. This process produces a feature vector that is used in the classification stage. Support vector machine (SVM) is used to evaluate the proposed feature extraction method. Simulation results showed that the Hjorth parameter on a scale of 1–15 yields 99.5% accuracy. The proposed method is expected to be applied to digital EEG for seizure detection and prediction.

Oyster mushroom (Pleurotus Ostreatus) is a fungus-like plant that is often cultivated in Indonesian agriculture. Oyster mushroom is raised by manipulating environmental parameters, so that it can grow in the provided Kumbung. Oyster mushroom requires a temperature that is used, which ranges from 23°-28°C, for humidity used between 70% -90% and for light intensity it requires light of ±300 lux. In this study, a system was designed to carry out automatic monitoring and control in real time based on the Internet of Things (IoT) which integrates DHT22 humidity and temperature sensors, BH1750 light intensity sensors and NodeMCU as a microcontroller with measurement results sent to the Firebase database. In addition, a water pump connected to the sprayer nozzle is installed in this system to maintain the humidity of the oyster mushroom curd. From the test results, the system can work automatically to stabilize temperature, humidity, and light intensity according to the ideal parameters.

ABSTRAK Saat ini sektor industri di Indonesia tumbuh semakin pesat yang menggeser lahan pertanian menjadi sempit. Hal tersebut mengakibatkan para petani perlu mencari lahan lain untuk tetap dapat memproduksi bahan pangannya. Hidroponik merupakan teknik bertanam menggunakan media air yang memanfaatkan lahan sempit. Salah satu tanaman yang sering diterapkan ialah tanaman selada. Namun, dengan penerapan teknik hidroponik ini masih terdapat kualitas tanaman selada yang kurang baik karena kurang memperhatikan pemeliharaannya sehingga mengakibatkan kurangnya nutrisi pada tanaman selada. Maka dari itu, pada penelitian ini akan membuat sistem klasifikasi defisiensi nutrisi pada tanaman hidroponik selada melalui citra daun dengan menggunakan Convolutional Neural Network (CNN) berbasis arsitektur MobileNetV2. Hasil dalam skenario uji penelitian ini memperoleh akurasi sebesar 88%. Dengan begitu, diharapkan dapat membantu para petani untuk mengetahui defisiensi nutrisi pada tanaman selada agar tetap dapat menjaga kualitas produksi tanaman selada. Kata kunci: CNN, hidroponik, MobileNetV2, nutrisi, selada ABSTRACT Currently the industrial sector in Indonesia is growing rapidly which shifts agricultural land to narrow. This resulted in farmers needing to look for other land to continue to be able to produce their food. Hydroponics is a farming technique using water media that utilizes narrow land. One of the plants that is often used is lettuce. However, with the application of this hydroponic technique, the quality of lettuce plants is still not good due to lack of attention to maintenance, resulting in a lack of nutrition in lettuce plants. Therefore, this research will create a nutritional deficiency classification system in hydroponic lettuce through leaf images using a Convolutional Neural Network (CNN) based on the MobileNetV2 architecture. The results in this research test scenario obtained an accuracy of 88%. That way, it is hoped that it can help farmers to find out nutritional deficiencies in  lettuce plants so that they can maintain the quality of lettuce production. Keywords: CNN, hydroponic, lettuce, MobileNetV2, nutrition