Torteeka, Peerapong

Preferred name

Torteeka, Peerapong

Now showing 1 - 10 of 13

Some of the metrics are blocked by your
consent settings
PCL-PTD Net: Parallel Cross-Learning-Based Pixel Transferred Deconvolutional Network for Building Extraction in Dense Building Areas With Shadow
(2023-01-01)
Boonpook, Wuttichai
;
Tan, Yumin
;
Torsri, Kritanai
;
Kamsing, Patcharin
;
Torteeka, Peerapong
Urban building segmentation from remote sensed imageries is challenging because there usually exists a variety of building features. Furthermore, very high spatial resolution imagery can provide many details of the urban building, such as styles, small gaps among buildings, building shadows, etc. Hence, satisfactory accuracy in detecting and extracting urban features from highly detailed images still remains. Deep learning semantic segmentation using baseline networks works well on building extraction; however, their ability in building extraction in shadows area, unclear building feature, and narrow gaps among buildings in dense building zone is still limited. In this article, we propose parallel cross-learning-based pixel transferred deconvolutional network (PCL-PTD net), and then is used to segment urban buildings from aerial photographs. The proposed method is evaluated and intercompared with traditional baseline networks. In PCL-PTD net, it is composed of parallel network, cross-learning functions, residual unit in encoder part, and PTD in decoder part. The performance is applied to three datasets (Inria aerial dataset, international society for photogrammetry and remote sensing Potsdam dataset, and UAV building dataset), to evaluate its accuracy and robustness. As a result, we found that PCL-PTD net can improve learning capacities of the supervised learning model in differentiating buildings in dense area and extracting buildings covered by shadows. As compared to the baseline networks, we found that proposed network shows superior performance compared to all eight networks (SegNet, U-net, pyramid scene parsing network, PixelDCL, DeeplabV3+, U-Net++, context feature enhancement networ, and improved ResU-Net). The experiments on three datasets also demonstrate the ability of proposed framework and indicating its performance.
Some of the metrics are blocked by your
consent settings
Deep neural learning adaptive sequential monte carlo for automatic image and speech recognition
(2020-01-01)
Kamsing, Patcharin
;
Torteeka, Peerapong
;
Boonpook, Wuttichai
;
Cao, Chunxiang
To enhance the performance of image classification and speech recognition, the optimizer is considered an important factor for achieving high accuracy. The state-of-the-art optimizer can perform to serve in applications that may not require very high accuracy, yet the demand for high-precision image classification and speech recognition is increasing. This study implements an adaptive method for applying the particle filter technique with a gradient descent optimizer to improve model learning performance. Using a pretrained model helps reduce the computational time to deploy an image classification model and uses a simple deep convolutional neural network for speech recognition. The applied method results in a higher speech recognition accuracy score - 89.693% for the test dataset - than the conventional method, which reaches 89.325%. The applied method also performs well on the image classification task, reaching an accuracy of 89.860% on the test dataset, better than the conventional method, which has an accuracy of 89.644%. Despite a slight difference in accuracy, the applied optimizer performs well in this dataset overall.
Some of the metrics are blocked by your
consent settings
Ground software architecture for a lunar particle detector: Implementation with a double-sided silicon strip detector
(2026-04-15)
Panyalert, Thanayuth
;
Manuthasna, Shariff
;
Torteeka, Peerapong
;
He, Xu
;
Zhang, Ning
The Moon-Aiming Thai-Chinese Hodoscope (MATCH) is a particle detector developed for the Chang’E-7 mission, designed to support space weather monitoring and cosmic radiation studies within the Sun-Earth-Moon system. The primary observational objectives of MATCH include space weather science and alerts, as well as the detection of cosmic radiation, particularly Jovian and Galactic Cosmic Ray (GCR) electrons. Additionally, MATCH can detect lunar albedo ions (alpha particles and protons) and contribute to understanding cosmic-ray interactions with the lunar surface, including high-energy particle backscattering mechanisms. These studies are essential for improving our knowledge of cosmic ray propagation and energy distribution in the lunar environment. MATCH integrates a double-sided silicon strip detector (DSSD) for precise position tracking of incoming particles, along with a bismuth germanate (BGO) scintillator stack for accurate energy measurements. This study presents the development and validation of a scalable ground software architecture that supports event detection, signal processing, and data calibration, optimized for constrained onboard resources. The system has been validated through hardware-in-the-loop (HIL) testing using alpha-emitting sources under mission-equivalent conditions, demonstrating high accuracy and resource efficiency for on-orbit data acquisition modes. Once deployed, MATCH is expected to provide the first continuous MeV-range cosmic electron measurements from lunar orbit, enabling new insights into Jovian and Galactic cosmic ray propagation, space weather variability, and lunar albedo ion generation. The software architecture developed here plays a critical role in enabling astrophysical investigations during the upcoming Chang’E-7 mission.
Some of the metrics are blocked by your
consent settings
Sustainable Urban Healthcare Accessibility: Voronoi Screening and Travel-Time Coverage in Bangkok
(2025-12-01)
Boonprong, Sornkitja
;
Punturasan, Nathapat
;
Kamsing, Patcharin
;
Torteeka, Peerapong
;
Cao, Chunxiang
This study presents an integrated and reproducible framework for within-tier screening of potential healthcare accessibility in Bangkok. Facilities in three service tiers (primary 294 units, regular 75, referral 29) are analyzed using point-pattern diagnostics, Voronoi geometric partitions, population-weighted allocation from subdistrict controls, and cumulative network travel-time isochrones. Spatial diagnostics indicate clustering among primary care units, a near-random configuration for regular units, and modest dispersion for referral hospitals, summarized by observed-to-expected nearest-neighbor ratios of approximately 0.77, 1.05, and 1.19, respectively. Voronoi partitions translate these distributions into geometric units that enlarge with increasing inter-facility spacing, while population-weighted assignments reveal higher population-per-partition-area burdens in the outer east and southwest. Isochrone maps (5–60 min rings) show central corridors with short travel times and peripheral areas where potential access declines. Interpreted against statutory planning intent, the maps indicate broad consistency of siting with high-intensity zones, alongside residual gaps at residential fringes. Framed as repeatable indicators of access and coverage, the workflow contributes to measuring and monitoring urban health sustainability under universal health coverage and routine planning cycles. The framework yields transparent indicators that support monitoring, priority setting, and incremental adjustments within each tier. Limitations include planar proximity assumptions, uniform areal weighting, single-mode modeled travel times without temporal variation, and the absence of capacity measures, motivating future work on capacity-weighted partitions, minimal dasymetric refinements, and time-dependent multimodal scenarios.
Some of the metrics are blocked by your
consent settings
Vision-based spacecraft pose estimation via a deep convolutional neural network for noncooperative docking operations
(2020-09-01)
Phisannupawong, Thaweerath
;
Kamsing, Patcharin
;
Torteeka, Peerapong
;
Channumsin, Sittiporn
;
Sawangwit, Utane
The capture of a target spacecraft by a chaser is an on-orbit docking operation that requires an accurate, reliable, and robust object recognition algorithm. Vision-based guided spacecraft relative motion during close-proximity maneuvers has been consecutively applied using dynamic modeling as a spacecraft on-orbit service system. This research constructs a vision-based pose estimation model that performs image processing via a deep convolutional neural network. The pose estimation model was constructed by repurposing a modified pretrained GoogLeNet model with the available Unreal Engine 4 rendered dataset of the Soyuz spacecraft. In the implementation, the convolutional neural network learns from the data samples to create correlations between the images and the spacecraft’s six degrees-of-freedom parameters. The experiment has compared an exponential-based loss function and a weighted Euclidean-based loss function. Using the weighted Euclidean-based loss function, the implemented pose estimation model achieved moderately high performance with a position accuracy of 92.53 percent and an error of 1.2 m. The in-attitude prediction accuracy can reach 87.93 percent, and the errors in the three Euler angles do not exceed 7.6 degrees. This research can contribute to spacecraft detection and tracking problems. Although the finished vision-based model is specific to the environment of synthetic dataset, the model could be trained further to address actual docking operations in the future.
Some of the metrics are blocked by your
consent settings
Deep Learning Semantic Segmentation for Land Use and Land Cover Types Using Landsat 8 Imagery
(2023-01-01)
Boonpook, Wuttichai
;
Tan, Yumin
;
Nardkulpat, Attawut
;
Torsri, Kritanai
;
Torteeka, Peerapong
Using deep learning semantic segmentation for land use extraction is the most challenging problem in medium spatial resolution imagery. This is because of the deep convolution layer and multiple levels of deep steps of the baseline network, which can cause a degradation problem in small land use features. In this paper, a deep learning semantic segmentation algorithm which comprises an adjustment network architecture (LoopNet) and land use dataset is proposed for automatic land use classification using Landsat 8 imagery. The experimental results illustrate that deep learning semantic segmentation using the baseline network (SegNet, U-Net) outperforms pixel-based machine learning algorithms (MLE, SVM, RF) for land use classification. Furthermore, the LoopNet network, which comprises a convolutional loop and convolutional block, is superior to other baseline networks (SegNet, U-Net, PSPnet) and improvement networks (ResU-Net, DeeplabV3+, U-Net++), with 89.84% overall accuracy and good segmentation results. The evaluation of multispectral bands in the land use dataset demonstrates that Band 5 has good performance in terms of extraction accuracy, with 83.91% overall accuracy. Furthermore, the combination of different spectral bands (Band 1–Band 7) achieved the highest accuracy result (89.84%) compared to individual bands. These results indicate the effectiveness of LoopNet and multispectral bands for land use classification using Landsat 8 imagery.
Some of the metrics are blocked by your
consent settings
The dynamics of wetland cover change using a state estimation technique applied to time-series remote sensing imagery
(2017-12-15)
Insom, Patcharin
;
Cao, Chunxiang
;
Boonsrimuang, Pisit
;
Torteeka, Peerapong
;
Boonprong, Sornkitja
Monitoring the dynamics of inundation areas in wetlands over contiguous years is important because it influences wetland ecosystem monitoring. However, because the variable nature of wetlands tends to hamper monitoring change analyses, the potential for misinterpretation increases. The Kalman filter (KF) or extended Kalman filter (EKF), which uses recursive processing based on the former information, can be applied to time-series remote sensing imagery. In the experiment, a periodic triangle function of two modulated parameters is treated as the system model, and Normalized Difference Vegetation Index (NDVI) time-series data are used for the measurement model in the correction processes of the state estimation. A decision metric is computed from the mean and amplitude sequence, which results from the state estimation filter. Consequently, an optimal threshold is calculated using a minimum error thresholding algorithm based on a pre-labelled sample. NDVI time-series data from Poyang Lake, China–derived from 250-m Moderate Resolution Imaging Spectroradiometer satellite data obtained from January 2009 to December 2013–are applied to monitor the dynamics of inundation changes. The results show that the EKF achieves satisfactory results, with 85.52% accuracy in the year 2009, while the KF has an accuracy of 84.16% during that same time.
Some of the metrics are blocked by your
consent settings
An enhanced learning algorithm with a particle filter-based gradient descent optimizer method
(2020-08-01)
Kamsing, Patcharin
;
Torteeka, Peerapong
;
Yooyen, Soemsak
This experiment integrates a particle filter concept with a gradient descent optimizer to reduce loss during iteration and obtains a particle filter-based gradient descent (PF-GD) optimizer that can determine the global minimum with excellent performance. Four functions are applied to test optimizer deployment to verify the PF-GD method. Additionally, the Modified National Institute of Standards and Technology (MNIST) database is used to test the PF-GD method by implementing a logistic regression learning algorithm. The experimental results obtained with the four functions illustrate that the PF-GD method performs much better than the conventional gradient descent optimizer, although it has some parameters that must be set before modeling. The results of implementing the MNIST dataset demonstrate that the cross-entropy of the PF-GD method exhibits a smaller decrease than that of the conventional gradient descent optimizer, resulting in higher accuracy of the PF-GD method. The PF-GD method provides the best accuracy for the training model, 97.00%, and the accuracy of evaluating the model with the test dataset is 90.37%, which is higher than the accuracy of 90.08% obtained with the conventional gradient descent optimizer.
Some of the metrics are blocked by your
consent settings
Trajectory Planning for Multiple UAVs and Hierarchical Collision Avoidance Based on Nonlinear Kalman Filters
(2023-02-01)
Hematulin, Warunyu
;
Kamsing, Patcharin
;
Torteeka, Peerapong
;
Somjit, Thanaporn
;
Phisannupawong, Thaweerath
Fully autonomous trajectory planning for multiple unmanned aerial vehicles (UAVs) is significant for building the next generation of the logistics industry without human control. This paper presents a method to enable multiple UAVs to fly in the same trajectory without collision. It benefits several applications, such as smart cities and transfer goods, during the COVID-19 pandemic. Different types of nonlinear state estimation are deployed to test the position estimation of drones by treating the information from AirSim as offline dynamic data. The obtained global positioning system sensor data and magnetometer sensor data are determined as the measurement model. The experiment in the simulation is separated into (1) the localization state, (2) the rendezvous state, in which the proposed rendezvous strategy is presented by using the relation between velocity and displacement through the setting area, and (3) the full mission state, which combines both the localization and rendezvous states. The localization state results show the best RMSE in the case of full GPS available at 0.21477 m and 0.25842 m in the case of a GPS outage during a period of time by implementing the ensemble Kalman filter. Similarly, the ensemble Kalman filter performs well with an RMSE of 0.5112414 m in the rendezvous state and demonstrates exceptional performance in the full mission state. Moreover, the experiment is implemented in a real-world situation with some basic drone kits as proof that the proposed rendezvous strategy can truly operate.
Some of the metrics are blocked by your
consent settings
Signal Calibration and Energy Resolution Optimization of a Double-Sided Silicon Strip Detector for Lunar-Based Particle Detection
(2025-01-01)
Panyalert, Thanayuth
;
Manuthasna, Shariff
;
Torteeka, Peerapong
;
He, Xu
;
Zhang, Ning
This letter presents a signal calibration and energy resolution analysis of a double-sided silicon strip detector (DSSD) developed for charged particle detection in a lunar-based space environment. The detector is part of the Moon-Aiming Thai-Chinese Hodoscope, i.e., a proposed scientific payload for the Chang'E-7 lunar orbiter, aimed at monitoring space weather and lunar-surface particle interactions. To evaluate the DSSD's performance under vacuum conditions, alpha sources (Am-241 and Pu-239) were used to generate energy spectra, which were processed through baseline correction and histogram generation. Four peak models, i.e., Gaussian, Gaussian + Exponential Tail, exponentially modified Gaussian (EMG), and Hyper-EMG, were compared using nonlinear least squares. Results show that the Hyper-EMG model yields superior fits, especially for Am-241, achieving an average reduced chi-squared of 1.64 ± 4.44 and energy resolution of 3.09% ± 0.45%, with 22 out of 32 Akaike Information Criterion (AIC) wins. In contrast, Gaussian fits showed higher fitting errors (e.g., x<sup>2</sup>/DoF up to 10.5) and the poorest resolution. AIC selection further confirms Hyper-EMG's robustness, while Gaussian fits were consistently inadequate. These findings support the use of tail-aware models, such as Hyper-EMG, for accurate energy reconstruction in spaceborne silicon detectors.

Torteeka, Peerapong

Filters

Settings

Sort By

Results per page