Cheng-Hsiung Hsieh

The integration of sensor technology and artificial intelligence is revolutionizing athletic training. This paper presents a novel cost-effective smart table tennis racket embedded with a nine-axis inertial measurement unit (IMU) for real-time stroke classification directly on the device. Unlike systems that are reliant on external computation, our approach leverages Tiny Machine Learning (TinyML) to deploy a customized Convolutional Neural Network (CNN) model onto a microcontroller unit (STM32F7), enabling real-time inference at the edge. The system captures accelerometer and gyroscope data, which is automatically segmented via a recursive algorithm and classified into six fundamental strokes (e.g., forehand/backhand stroke, pull, and chop) or a non-swing state. The classified results are wirelessly transmitted to a computer application for real-time feedback. Experimental results with actual players demonstrate that the optimized CNN model achieves an average classification accuracy of 98.3% in controlled tests and over 94% in mixed-stroke scenarios, validating the system’s high accuracy and robustness. This work exemplifies the practical implementation of an end-to-end intelligent sensor system, highlighting the potential of TinyML to enable advanced, low-power motion analysis in sports.

This study presents a laboratory-scale prototype that integrates a zero-trust-based permission-control mechanism with real-time DDoS traffic detection in a 5G NSA IoT testbed. The proposed system was evaluated under three controlled traffic conditions: normal traffic, TCP SYN flood traffic, and UDP flood traffic. Packet-level data were collected from the experimental testbed and used to train and compare LSTM and SVM classifiers. Under the evaluated conditions, the LSTM model achieved the highest accuracy of 99.56%, outperforming the best SVM result of 93.20%. The selected LSTM detector was further deployed in the edge-computing pipeline and correctly identified the three tested traffic conditions during real-time operation. After malicious traffic was identified, the permission-control mechanism updated the corresponding authorization status, generated an alert, and restricted suspicious communication within the testbed. These results demonstrate the feasibility of linking traffic detection with authorization adjustment in a controlled 5G NSA IoT prototype. The findings should not be interpreted as a general validation against all DDoS variants or large-scale commercial 5G IoT deployments.

While pre-trained deep learning models have significantly advanced image dehazing, their restoration performance often fluctuates substantially across varying haze densities, leading to inconsistent performance across diverse atmospheric conditions. To address this limitation, this study introduces a performance analysis approach based on Haze Image Clustering (HIC) to systematically evaluate the specialized strengths of various state-of-the-art models within specific haze-level intervals. Building upon these evaluations, we propose a heterogeneous modular framework equipped with a dynamic switching mechanism that adaptively activates the optimal pre-trained module for each detected haze level. Extensive experiments conducted on the OTS and ODF benchmark datasets demonstrate that while individual models exhibit regional performance drops, the proposed framework consistently maintains superior performance across all haze intensities. Quantitative results indicate that the proposed modular network achieves a significant PSNR improvement of up to 6.946 dB compared to DehazeFlow. Furthermore, regarding the no-reference Dehazing Quality Index (DHQI), our framework attains a top score of 68.448, surpassing the best individual baseline. These findings validate that the proposed strategy effectively enhances both restoration fidelity and visual naturalness without the need for additional training or fine-tuning, offering a robust and computationally efficient solution for real-world image dehazing.

Recently, image haze removal has gained increasing attention in the field of image restoration. Data-driven and model-based methods are popular among researchers. The dark channel prior prevails in model-based methods, where the model parameters, atmospheric light, and transmittance are generally estimated through a block-based dark channel. This paper proposes a model-based approach with integrated pixel- and block-based dark channels for initial transmittance estimation. Additionally, we developed a two-stage guided image filtering process to refine the initial transmittance while utilizing the pixel-based dark channel to estimate atmospheric light. Our approach introduces two scaling factors for atmospheric light and initial transmittance, which are optimized using the Whale Optimization Algorithm. A CNN image regression model is employed to learn the mapping between hazy images and their corresponding optimized scaling factors, thus eliminating the need for ground-truth images. This makes our approach applicable in real-world scenarios. The proposed approach was validated using two datasets: an artificially generated image dataset, RESIDE, and a natural image dataset, KeDeMa. The results show that our approach outperforms four other dehazing methods, i.e., GCAN, RRO, RFDN, and Ka-Net. With the RESIDE dataset, our approach outperforms GCAN, RRO, RFD, and Ka-Net by 2.009 dB, 6.042 dB, 3.488 dB, and 8.975 dB, respectively, in terms of PSNR. With the KeDeMa dataset, our approach generally demonstrates superior visual quality to the four comparison methods. The results suggest that the proposed model-based approach may outperform data-driven methods.

Vehicle exhaust is a major source of air pollution that contributes to environmental degradation and poses risks to public health. This paper presents an image-based method to estimate opacity (OP) and particulate matter (PM) from vehicle exhaust. In the proposed method, YOLOv7 was used to identify vehicles and, thus, the region of interest (ROI). Then, a support vector regression was trained, with four image features extracted from the ROI as the input vectors, while OP or PM was used as the output. The proposed method was verified by experiments where moving and static scenarios with three ROIs were considered. The ROIs used in the experiments were exhaust pipe area (EPA), vehicle bounding box (VBB), and white background (WBG). In the moving scenario, the EPA and VBB ROIs were considered. For the VBB ROI, the average R² values for OP and PM in the given examples were 0.834 and 0.894, respectively. For the EPA ROI, the average R² values for OP and PM were 0.838 and 0.910, respectively. In the static scenario, the EPA and WBG ROIs were considered. For the EPA ROI, the average R² values for OP and PM were 0.619 and 0.612, respectively. For the WBG ROI, the average R² values for OP and PM were 0.748 and 0.732, respectively. The results suggest that the EPA ROI is preferable in the moving scenario and the WBG ROI in the static scenario to estimate OP and PM from vehicle exhaust. The proposed method is promising in the estimation of OP and PM from vehicle exhaust because satisfactory R² values were achieved.

Recently, supervised deep learning methods have been widely used for image haze removal. These methods rely on training data that are assumed to be appropriate. However, this assumption may not always be true. We observe that some data may contain hazy ground truth (GT) images. This can lead to supervised deep image dehazing (SDID) models learning inappropriate mapping between hazy images and GT images, which negatively affects the dehazing performance. To address this problem, two difficulties must be solved. One is to estimate the haze level in an image, and the other is to develop a haze level indicator to discriminate clear and hazy images. To this end, we proposed a haze level estimation (HLE) scheme based on dark channel prior and a haze level indicator accordingly for training data cleaning, i.e., to exclude image pairs with hazy GT images in the data set. With the data cleaning by the HLE, we introduced an SDID framework to avoid inappropriate learning and thus improve the dehazing performance. To verify the framework, using the RESIDE data set, experiments were conducted with three types of SDID models, i.e., GCAN, REFN and cGAN. The results show that our method can significantly improve the dehazing performance of the three SDID models. Subjectively, the proposed method generally provides better visual quality. Objectively, our method, using fewer training image pairs, was capable of improving PSNR in the GCAN, REFN, and cGAN models by 3.10 dB, 5.74 dB, and 6.44 dB, respectively. Furthermore, our method was evaluated using a real-world data set, KeDeMa. The results indicate that the better visual quality of the dehazed images is generally for models with the proposed data cleaning scheme. The results demonstrate that the proposed method effectively and efficiently enhances the dehazing performance in the given examples. The practical significance of this research is to provide an easy but effective way, that is, the proposed data cleaning scheme, to improve the performance of SDID models.