KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 10 of 41

Some of the metrics are blocked by your
consent settings
Frame-Level Accident Recognition via Detection Confidence Aggregation: A Cross-Domain Validation Framework for Thai Roadway Surveillance
(2026-07-01)
Gabbualoy, Somprasonk
;
Phasukkit, Pattarapong
;
Houngkamhang, Nongluck
Real-time roadway surveillance now leans hard on automated detection. How a model trained in one geographic context actually behaves on another, though, is still underexplored for Southeast Asian deployments. We answer that question for Thai roadway closed-circuit television with a cross-domain validation framework. A YOLOv11n (Ultralytics v8.2.0; Ultralytics, Los Angeles, CA, USA) detector trained with focal loss feeds a confidence-aggregation step that turns per-detection scores into a per-frame accident score, and we put four aggregation operators head-to-head. Reliability comes from DeLong variance estimation paired with non-parametric bootstrap on 1245 Thai frames that carry 23 positive accident events. Under maximum-class aggregation the proposed configuration reaches a frame-level AUROC of 0.959 ± 0.020 across three random seeds. Under top-K aggregation it reaches 0.965 ± 0.018. Per-seed DeLong 95 percent intervals exclude chance performance throughout. We also evaluate three baseline configurations: YOLOv5su comes in at 0.738, YOLOv8n at 0.868, and a Chiang Mai-tuned YOLOv11n variant at 0.918. The architectural progression seen on standard benchmarks therefore carries cleanly into the cross-domain setting. The same Chiang Mai-tuned variant reached an in-domain mAP50 of 0.952 yet only 0.918 cross-region AUROC on a separate Thai region, which is a quiet but clear signal that geographic proximity within a country does not on its own remove distributional shift. Bounding-box localisation appears as a secondary diagnostic because the operational target here is frame-level alerting rather than pixel-precise annotation. Edge deployment optimisation falls outside the present scope. What the work leaves behind is a reproducible baseline and a statistical protocol that follow-up Southeast Asian roadway-safety research can build on.
Some of the metrics are blocked by your
consent settings
A Contactless Edge-AI Prototype for Simulated Apnea-like Respiratory Suppression and Motion Artifact Detection Using 60 GHz FMCW Radar
(2026-07-01)
Pairoch, Sathit
;
Phasukkit, Pattarapong
;
Houngkamhang, Nongluck
Sleep-related respiratory disturbances are difficult to monitor continuously outside specialized laboratories because conventional polysomnography is resource-intensive and intrusive. This study presents a contactless edge-AI engineering prototype for detecting controlled voluntary respiratory-motion suppression and motion artifacts using a 60 GHz frequency-modulated continuous-wave radar. The system integrates a 60 GHz radar front end, lightweight local preprocessing, an INT8 one-dimensional convolutional neural network deployed on the Analog Devices MAX78000 CNN accelerator (Analog Devices Thailand, Chon Buri, Thailand), and an event-driven Raspberry Pi Zero 2W gateway for alert transmission. Evaluation was performed using a controlled healthy-volunteer dataset consisting of normal breathing, voluntary breath-holding-induced respiratory suppression, and deliberate motion artifact. The final valid test set contained 270 technically valid 30 s windows balanced across the three classes. The INT8 model achieved an overall accuracy of 92.6% (95% confidence interval: 88.8–95.2%), with a macro-averaged precision, recall, and F1-score of 92.6%, 92.6%, and 92.5%, respectively. Active CNN inference on the MAX78000 consumed 0.152 ± 0.011 mJ and was completed in 5.20 ± 0.11 ms, corresponding to approximately 280-fold lower active inference energy than Python 3.14.6/TensorFlow Lite 2.21.0-based execution on the Raspberry Pi Zero 2W. These results demonstrate the feasibility of privacy-aware, low-power respiratory-pattern classification at the edge. However, the study should be interpreted strictly as an engineering proof-of-concept based on controlled voluntary breathing and movement tasks in healthy volunteers. It is not a clinically validated apnea or obstructive sleep apnea detection system and did not include polysomnography, oxygen saturation measurement, airflow sensing, sleep staging, or diagnosed patient cohorts.
Some of the metrics are blocked by your
consent settings
A Predictive Dual-Stage Neural Framework for Phase-Coherent Auditory Synthesis on Edge Devices
(2026-06-01)
Pairoch, Sathit
;
Phasukkit, Pattarapong
;
Suteewong, Teeraporn
Real-time binaural beat synthesis in dynamic acoustic environments is challenged by carrier non-stationarity, interaural phase discontinuities, and processing delay in conventional digital signal processing pipelines. This study proposes a predictive dual-stage neural framework for phase-coherent auditory synthesis under non-stationary acoustic conditions. The framework decouples real-time carrier estimation from phase-coherent signal generation through two specialized modules. An intelligent acoustic sensing module (AI-1) estimates time-varying carrier information across harmonic, fluctuating, and broadband acoustic profiles using a causal neural front-end with an adaptive confidence-driven strategy. A predictive phase-coherent generator (AI-2) then forecasts short-horizon carrier trajectories and drives a discrete-time phase accumulator to maintain continuous phase evolution during binaural beat embedding. Objective evaluation under multiple acoustic profiles and noise conditions shows that the proposed framework maintains strong phase continuity, with a Phase Coherence Factor greater than 0.91, and low artifact levels, with a Signal-to-Artifact Ratio greater than 39.8 dB, under the evaluated conditions. Additional comparisons with conventional DSP baselines, stronger classical F0 estimators, a lightweight neural F0 tracker, and component-wise ablation variants further demonstrate that the performance improvement arises from the combination of adaptive carrier estimation and predictive phase-coherent actuation, rather than from carrier estimation alone. Hardware profiling shows a combined INT8 inference time of 2.4 ms per frame on a resource-constrained Raspberry Pi Zero 2W-class edge device. Importantly, this inference time and the sub-millisecond phase-accumulator resolution should not be interpreted as sub-millisecond end-to-end physical audio latency. The complete system still includes buffering, framing, neural inference, and output processing delay; the proposed method instead reduces effective phase-boundary misalignment through short-horizon predictive compensation. These results support the proposed framework as a lightweight engineering solution for real-time phase-continuous auditory synthesis in dynamic listening environments. The reported PCF and SAR values should be interpreted as signal-level indicators of phase continuity and artifact suppression, rather than as evidence of listener comfort, perceptual preference, or neurophysiological efficacy.
Some of the metrics are blocked by your
consent settings
Cross-Sensor and Cross-Population Generalization of Deep Learning Models for Digital Mammography: A Controlled Four-Country Benchmark of Five Backbone Architectures with Statistical Significance Testing
(2026-06-01)
Gabbualoy, Somprasonk
;
Phasukkit, Pattarapong
;
Tungjitkusolmun, Supan
Background/Objectives: Deep learning models for digital mammography sensor data are increasingly deployed across hospitals using different X-ray detector technologies and patient populations. Whether models trained on one sensor platform and population maintain accuracy when transferred to another has not been tested for the latest generation of mammography-specific foundation models under one controlled protocol. Methods: We fine-tuned five backbone architectures (ResNet-50, DINOv2-B14, Rad-DINO, Mammo-CLIP B5, and Mammo-FM) on CBIS-DDSM (film-digitized, USA, n = 714 validation) with three seeds, ablated a density-aware focal loss across three auxiliary weights, and evaluated transfer to three external sensor cohorts: CMMD (full-field digital, China, n = 1032), DMID (mixed digital, India, n = 509), and MIAS (film-digitized, UK, n = 322). Significance used paired DeLong z-tests with Benjamini–Hochberg FDR correction; temperature scaling tested post hoc recalibration at all transfer targets. Results: Within this single-source three-seed evaluation, ResNet-50 outperformed all four foundation models on CBIS-DDSM (AUC 0.867 vs. 0.847, 0.846, 0.813, and 0.703; all gaps p_adj < 0.05). The density-aware focal loss degraded both AUC and calibration at every weight tested. At transfer, every model lost 0.165 to 0.320 AUC points relative to in-distribution performance, with sensitivity at 95% specificity collapsing from 0.31 to 0.47 in-distribution to 0.11 to 0.22 across the three external targets. A per-seed Stouffer meta-analysis confirms that Mammo-CLIP B5 and Mammo-FM significantly outperformed ResNet-50 on DMID and Mammo-CLIP on CMMD, after BH-FDR; MIAS comparisons remained directional only. In the extremely dense subgroup (BI-RADS D4), Mammo-FM reached AUC 0.870 versus ResNet-50 at 0.842, a directional observation whose 95% CIs overlap heavily at the n = 140 sample size and which we do not interpret as a statistically supported advantage. Conclusions: In this single training-source, three-seed protocol, mammography-specific pretraining did not deliver the in-distribution AUC premium reported in the originating papers, and no architecture reached a level at which transfer deployment without local validation would be defensible. We frame these as observations specific to the present protocol rather than as broader conclusions about foundation models for mammography classification. The findings argue for sensor-stratified and population-stratified external validation and for local recalibration as practical prerequisites before clinical use. Code and weights are released under MIT license.
Some of the metrics are blocked by your
consent settings
Smart Diagnostics: Hierarchical Deep Learning of Acoustic Emission Signals for Early Crack Detection in Zirconia Dental Structures
(2026-05-01)
Tuntiwong, Kuson
;
Wangman, Rangsinee
;
Kanchanatawewat, Kanchana
;
Anucul, Boonjira
;
Sritart, Hiranya
Monolithic zirconia restorations are frequently affected by the unnoticed growth of subcritical cracks, a failure process that is not captured by traditional imaging methods like radiographs and ultrasounds in sophisticated dental architectures. To address this evaluative inadequacy, this research introduces a hierarchical deep learning framework for microcrack detection and spatial localization. We promote a hierarchical deep learning system that integrates Acoustic Emission (AE) detection alongside signal processing. Raw AE signals utilized during dynamic loading are enhanced via Kalman filtering and Continuous Wavelet Transform (CWT) to construct high-fidelity time–frequency scalograms. The diagnostic pipeline operates in two stages: first, a hybrid CNN–BiGRU network with temporal attention fulfills zirconia component-level classification; second, a ResNet-18 backbone integrated with Bidirectional LSTM and Multi-Head Attention precisely localizes defects across five anatomical crown regions. This hierarchical design effectively captures the non-stationary, transient nature of fracture-induced stress waves. The framework achieved an F1-score of 99.00% and an AUC of 0.994, significantly outperforming conventional convolutional networks. By enabling predictive maintenance through early, non-invasive damage localization, this study demonstrates a promising laboratory framework for AE-based crack detection in zirconia dental structures and prosthetics and toward enhanced clinical reliability in digital dentistry.
Some of the metrics are blocked by your
consent settings
Multi-Entropy Feature Extraction With LSTM Networks for Acoustic Emission-Based Railway Crack Localization
(2026-01-01)
Laon, Popphon
;
Pourbunthidkul, Supavee
;
Rattan, Praphaporn
;
Sahavisit, Tanawit
;
Suwansin, Wara
Railway infrastructure security is contingent upon the prompt identification of structural anomalies within steel tracks. This research establishes a framework that merges acoustic emission (AE) sensing with advanced machine learning for prompt fracture identification and location. To improve the quality of Raw AE signals, they are first cleaned using Kalman filtering to tackle environmental disturbances and ambiguous readings. The characteristics that arise from entropy comprising approximate entropy, Shannon entropy, and dimensional entropy are derived to delineate the temporal patterns of fracture-induced emissions. Density-based spatial clustering of applications with noise (DBSCAN) is applied to remove outliers during preprocessing. Long short-term memory (LSTM) networks classify crack locations into three anatomical regions: rail head, web, and foot. Experimental validation with 3,000 labeled AE signals (1,000 per class) under laboratory conditions, the full pipeline that integrates Kalman filtering, entropy features, DBSCAN, and an LSTM classifier attains an accuracy of 98.83%. With an entropy-based variant, the accuracy drops to 96.67%, confirming the incremental value of temporal denoising and outlier rejection. While using mel-frequency cepstral coefficient (MFCC) baseline achieves 97.67% accuracy, a deep neural network (DNN) trained on Kalman-filtered, entropy-based, and DBSCAN-processed inputs reaches 96.33% accuracy, underscoring the advantages of temporal modeling for AE. A GRU using the same Kalman-filtered, entropy-based, and DBSCAN-processed inputs, achieves 97.67% accuracy, but trails the LSTM overall. When deployed on actual railway lines with a mobile inspection platform, the system maintained robust performance, correctly identifying 84.67% of cracks at 3 km/h and 80.67% at 5 km/h. The LSTM configuration consistently outperformed all alternative approaches, including the entropy-only variant, MFCC-based method, DNN classifier, and GRU model. This confirms the LSTM’s enhanced capability to capture the temporal dynamics of AE signals, establishing it as the most effective framework for AE-based crack localization in railway structural health monitoring.
Some of the metrics are blocked by your
consent settings
Skip-Feedback Neural Network With Unsupervised Feature Learning for Railway Crack Distance Estimation
(2026-01-01)
Pavitpok, Siwagorn
;
Phasukkit, Pattarapong
Accurate estimation of crack distance is critical for railway safety and structural health monitoring. This paper proposes a Skip-Feedback Neural Network (SFNN) for single-sensor acoustic emission (AE)–based crack distance estimation in railway rails. Unlike conventional multi-sensor or high-cost inspection systems, the proposed framework employs a single AE sensor together with entropy-based signal representations to capture distance-dependent crack characteristics under noisy operating conditions. The SFNN integrates learnable skip and feedback pathways to enhance feature reuse, stabilize gradient propagation, and improve robustness against signal attenuation. Experimental evaluation was performed using controlled pencil-lead break simulations at multiple crack-to-sensor distances, as well as field measurements at real crack locations validated by phased-array ultrasonic testing. Results demonstrate that the proposed SFNN consistently outperforms baseline neural-network models across different distance ranges. The proposed framework offers a cost-efficient, noise-resilient, and scalable solution for AE-based railway crack monitoring and future structural health-monitoring applications.
Some of the metrics are blocked by your
consent settings
Hybrid Unsupervised–Supervised Learning Framework for Rainfall Prediction Using Satellite Signal Strength Attenuation
(2026-01-01)
Laon, Popphon
;
Sahavisit, Tanawit
;
Pourbunthidkul, Supavee
;
Puangragsa, Sarut
;
Wichittrakarn, Pattharin
Satellite communication systems experience significant signal degradation during rain events, a phenomenon that can be leveraged for meteorological applications. This study introduces a novel hybrid machine learning framework combining unsupervised clustering with cluster-specific supervised deep learning models to transform satellite signal attenuation into a predictive tool for rainfall prediction. Unlike conventional single-model approaches treating all atmospheric conditions uniformly, our methodology employs K-Means Clustering with the Elbow Method to identify four distinct atmospheric regimes based on Signal-to-Noise Ratio (SNR) patterns from a 12-m Ku-band satellite ground station at King Mongkut’s Institute of Technology Ladkrabang (KMITL), Bangkok, Thailand, combined with absolute pressure and hourly rainfall measurements. The dataset comprises 98,483 observations collected with 30-s temporal resolutions, providing comprehensive coverage of diverse tropical atmospheric conditions. The experimental platform integrates three subsystems: a receiver chain featuring a Low-Noise Block (LNB) converter and Software-Defined Radio (SDR) platform for real-time data acquisition; a control system with two-axis motorized pointing incorporating dual-encoder feedback; and a preprocessing workflow implementing data cleaning, K-Means Clustering (k = 4), Synthetic Minority Over-Sampling Technique (SMOTE) for balanced representation, and standardization. Specialized Long Short-Term Memory (LSTM) networks trained for each identified cluster enable capture of regime-specific temporal dynamics. Experimental validation demonstrates substantial performance improvements, with cluster-specific LSTM models achieving R<sup>2</sup> values exceeding 0.92 across all atmospheric regimes. Comparative analysis confirms LSTM superiority over RNN and GRU. Classification performance evaluation reveals exceptional detection capabilities with Probability of Detection ranging from 0.75 to 0.99 and False Alarm Ratios below 0.23. This work presents a scalable approach to weather radar systems for tropical regions with limited ground-based infrastructure, particularly during rapid meteorological transitions characteristic of tropical climates.
Some of the metrics are blocked by your
consent settings
Reinforcement Learning-Driven Framework for High-Precision Target Tracking in Radio Astronomy
(2025-12-01)
Sahavisit, Tanawit
;
Laon, Popphon
;
Pourbunthidkul, Supavee
;
Wichittrakarn, Pattharin
;
Phasukkit, Pattarapong
Radio astronomy requires precise target localization and tracking to ensure accurate observations. Conventional regulation methodologies, encompassing PID controllers, frequently encounter difficulties due to orientation inaccuracies precipitated by mechanical limitations, environmental fluctuations, and electromagnetic interferences. To tackle these obstacles, this investigation presents a reinforcement learning (RL)-oriented framework for high-accuracy monitoring in radio telescopes. The suggested system amalgamates a localization control module, a receiver, and an RL tracking agent that functions in scanning and tracking stages. The agent optimizes its policy by maximizing the signal-to-noise ratio (SNR), a critical factor in astronomical measurements. The framework employs a reconditioned 12-m radio telescope at King Mongkut’s Institute of Technology Ladkrabang (KMITL), originally constructed as a satellite earth station antenna for telecommunications and was subsequently refurbished and adapted for radio astronomy research. It incorporates dual-axis servo regulation and high-definition encoders. Real-time SNR data and streaming are supported by a HamGeek ZedBoard with an AD9361 software-defined radio (SDR). The RL agent leverages the Proximal Policy Optimization (PPO) algorithm with a self-attention actor–critic model, while hyperparameters are tuned via Optuna. Experimental results indicate strong performance, successfully maintaining stable tracking of randomly moving, non-patterned targets for over 4 continuous hours without any external tracking assistance, while achieving an SNR improvement of up to 23.5% compared with programmed TLE-based tracking during live satellite experiments with Thaicom-4. The simplicity of the framework, combined with its adaptability and ability to learn directly from environmental feedback, highlights its suitability for next-generation astronomical techniques in radio telescope surveys, radio line observations, and time-domain astronomy. These findings underscore RL’s potential to enhance telescope tracking accuracy and scalability while reducing control system complexity for dynamic astronomical applications.
Some of the metrics are blocked by your
consent settings
Implementation of Deep Reinforcement Learning for Radio Telescope Control and Scheduling
(2025-12-01)
Puangragsa, Sarut
;
Sahavisit, Tanawit
;
Laon, Popphon
;
Puangragsa, Utumporn
;
Phasukkit, Pattarapong
The proliferation of terrestrial and space-based communication systems introduces significant radio frequency interference (RFI), which severely compromises data acquisition for radio telescopes, necessitating robust and dynamic scheduling solutions. This study addresses this challenge by implementing a Deep Recurrent Reinforcement Learning (DRL) framework for the control and dynamic scheduling of the X-Y pedestal-mounted KMITL radio telescope, explicitly trained for RFI avoidance. The methodology involved developing a custom simulation environment with a domain-specific Convolutional Neural Network (CNN) feature extractor and a Long Short-Term Memory (LSTM) network to model temporal dynamics and long-horizon planning. Comparative evaluation demonstrated that the recurrent DRL agent achieved a mean effective survey coverage of 475 deg<sup>2</sup>/h, representing a 72.7% superiority over the non-recurrent baseline, and maintained exceptional stability with only 1.0% degradation in median coverage during real-world deployment. The DRL framework offers a highly reliable and adaptive solution for telescope scheduling that is capable of maintaining survey efficiency while proactively managing dynamic RFI sources.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results