Wahyono

Monitoring Decayed, Missing, and Filled Teeth (DMFT) provides valuable insights into the prevalence and severity of tooth decay. However, existing algorithms often struggle with detecting missing teeth, as they can be overly sensitive to incomplete features. To address this limitation, this study proposes a novel YOLOv8 framework for tooth detection integrated with linear regression, DBSCAN clustering, and Lagrange interpolation for identifying missing teeth. A dataset of 983 frontal intraoral images acquired through smartphone-based imaging was considered. Experimenting on 197 images demonstrated that the proposed method enhances system performance, achieving precision of 64.27%, recall of 75.34%, and mean Average Precision (mAP) of 52.03%. Compared to the standard YOLOv8 model, the proposed approach improves precision by 2.27%, recall by 20.66%, and mAP by 6.83%. These findings highlight the potential of the proposed method as an effective tool for monitoring DMFT, contributing to improved diagnostic accuracy in enhancing missing teeth detection.

Contrast enhancement provides a more precise visualization of anatomical structures, improving diagnostic accuracy in medical images. One of the contrast enhancement methods, Contrast Limited Adaptive Histogram Equalization (CLAHE), often struggles with parameter optimization, leading to suboptimal image quality. Optimal parameter optimization is crucial to balancing contrast enhancement and detail preservation, necessitating robust optimization algorithms. The Cuckoo Search Algorithm (CSA) is well-suited for this task due to its strong global search capabilities and simplicity in handling complex optimization problems. CSA has two parameters, step size and discovery rate, which are often used as constants, resulting in sensitivity to problems, convergence rate, and an optimal solution that cannot be guaranteed simultaneously. To address these limitations, this study proposes an improved CSA, which, unlike conventional CSA with static parameters, introduces dynamic adjustments of the discovery rate ( ) and step size ( ), significantly improving exploration and exploitation capabilities. A hybrid distribution combining normal and uniform distributions is used for cuckoo selection and nest replacement, ensuring a balanced search process. The proposed method, called Dynamic Hybrid CSA (DH-CSA-CLAHE), was tested on MRI images of individuals with autism, showing superiority in MSE, PSNR, AMBE, SSIM, GMSD, and FSIM compared to CSA-CLAHE, PSO-CLAHE, and FA-CLAHE. The experimental results demonstrate the superior performance of the proposed method, achieving average PSNR, SSIM, and FSIM values of 45.54 dB, 0.97, and 0.9995, respectively, indicating excellent structural preservation and image quality. In addition, the method consistently produced the lowest MSE (3.73), AMBE (1.05), and GMSD (0.001) values, confirming its ability to effectively enhance contrast while minimizing distortion. These findings highlight the potential of DH-CSA-CLAHE as an effective tool for medical image preprocessing, contributing to improved diagnostic accuracy in clinical applications.

Artificial intelligence has made significant progress in processing audio, text, and images, but noise remains a major challenge, especially in real-world audio data. This research presents a novel approach to improve audio classification by integrating noise reduction techniques with machine learning models. Focusing on the bonang barung, a traditional Javanese gamelan instrument, the study uses Mel Frequency Cepstral Coefficients (MFCC) and Mel spectrograms to identify the most effective features for classification, and the Multi-Layer Perceptron (MLP) model for the classification task. In addition, the spectral subtraction method is used to reduce noise, which resulted in significant improvements in audio quality, although some artifacts remain. The main contribution of this study is the integration of noise reduction with the MLP model to improve the classification performance. The MLP model successfully classified various bonang barung playing techniques, achieving a classification accuracy of 90% after noise reduction compared to 87.22% with noise, highlighting the importance of preprocessing steps, such as noise reduction. It is also demonstrated that MLP models can be a viable alternative to more complex deep learning models, such as CNN and RNN, for audio classification tasks. Overall, this research provides new insights into the role of noise reduction in audio analysis and offers potential advances in the field of audio classification.

In intelligent transportation systems, vehicle speed estimation plays a vital role in traffic monitoring, speed enforcement, and autonomous vehicles. Therefore, the authors propose a vehicle speed estimation method composed of pipelines: homography transformation using a deep image homography transformation network, vehicle detection by YOLOv8, tracking by ByteTrack, and speed estimation in consecutive frames. Homography transformation is utilized to rectify the monocular video view image into a bird’s-eye view with a constant pixel-per-meter value. In this scheme, the authors propose a method for estimating the speed in consecutive frames using statistical and machine learning approaches by comparing several experimental schemes to determine which method is better for estimating vehicle speed, such as vehicle speed estimation results in Root Mean Square Error (RMSE) and Mean Absolute Error (MAE), traffic conditions, vehicle directions, and state-of-the-art studies. A monocular video DeaSpeedDataset was generated to evaluate the proposed system method by comparing the predicted speed value to the ground truth value obtained by the speed gun. This dataset has 2408 vehicles divided into 20 conditions, four setups, and two locations. Furthermore, the best RMSE was approximately 2.37897 km/h, and the MAE was approximately 1.68977 km/h.

Pulmonary Tuberculosis (TB) is a primary global infectious disease. Diagnosing TB patients involves medical examination and chest X-ray (CXR) imaging. This CXR image creates an opportunity to utilize machine learning to help physicians and radiologists diagnose TB suspects. Due to the inconsistency of image quality, image enhancement is one of the preprocessing steps to overcome the poor quality of the image. This study examines the effects of several image enhancement techniques, i.e., Histogram Equalization (HE), Contrast Limited Adaptive Histogram Equalization (CLAHE), and Fast Fourier Transform (FFT). These enhanced images are input for a Convolutional Neural Network (CNN). InceptionV3 is a transfer learning architecture with ImageNet as the pre-trained model. The image dataset consists of 3,500 normal and 3,500 tuberculosis CXR images. The best performance, in terms of accuracy and processing time, is achieved by the CLAHE enhancement technique, increasing accuracy by 4.57% compared to the original images as input and a processing time of 5.6 ms faster per testing image. A deeper analysis shows despite FFT achieving high performance, the processing time increases by 14.4 ms compared to the original image processing time. This study concluded that each image enhancement needs to consider the characteristics of the images.

Penyakit virus corona 2019 (COVID-19) merupakan keadaan darurat kesehatan masyarakat secara global. Salah satu cara untuk dapat mendeteksi adanya COVID-19 adalah dengan memanfaatkan citra x-ray dada yang mengidentifikasi anomali pada area paru-paru. Namun terkadang citra yang didapatkan pada melalui scan x-ray memiliki kualitas yang buruk sehingga sulit secara langsung untuk bisa dianalisis secara manual atau menggunakan model machine learning. Untuk menghasilkan analisis yang lebih baik, biasanya citra akan ditingkatkan terlebih dahulu kualitasnya dengan teknik image enhancement. Banyak metode image enhancement yang bisa dimanfaatkan untuk keperluan ini yang tentunya akan memberikan pengaruh yang berbeda-beda pada hasil pendeteksian COVID-19. Penelitian ini bertujuan untuk menganalisis pengaruh metode-metode image enhancement pada pendeteksian COVID-19 melalui citra x-ray. Metode image enhancement yang dianalisis adalah log transformation, constrast stretching, histogram equalization (HE), dan contrast limited adaptive histogram equalization (CLAHE). Model pendeteksian COVID-19 akan berdasarkan pada beberapa model machine learning antara lain convolutional neural network (CNN), k-nearest neighbor (KNN) dan support vector machine (SVM). Dari eksperiment didapatkan bahwa pengaplikasian image enhancement HE dan CLAHE memberikan peningkatan akurasi pada pendeteksian COVID-19 khususnya ketika diterapkan pada model CNN yaitu sebesar 1,13% dan 2,25%.

As one of the essential modules in intelligent surveillance systems, loitering detection plays an important role in reducing theft incidents by analyzing human behavior. This paper introduces a novel strategy for detecting the loitering activities of humans in the monitoring area for an intelligent surveillance system based on a vision sensor. The proposed approach combines spatial and temporal information in the feature extraction stage to decide whether the human movement can be regarded as loitering. This movement has been previously tracked using human detectors and particle filter tracking. The proposed method has been evaluated using our dataset consisting of 20 videos. The experimental results show that the proposed method could achieve a relatively good accuracy of 85% when utilizing the random forest classifier in the decision stage. Thus, it could be integrated as one of the modules in an intelligent surveillance system.