KMITL

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

Browse

Search Results

Now showing 1 - 8 of 8

Some of the metrics are blocked by your
consent settings
On-edge 2D-to-3D generative pipeline for seamless instance transformation
(2025-12-01)
Petchhan, Jirayu
;
Doungtap, Surasachai
Despite ongoing challenges with fragmented workflows, latency in device imports, and the main issue of limitations in object reconstruction functionality, relying on imperfect extraction networks remains an impractical solution for scalable object generation. To deal with these constraints, we proposed an end-to-end pipeline that leverages a re-designed self-consistency mechanism—aimed at reducing discrimination, along with the beneficial enhancement from level-set projection and gradient-surface orthogonality. In addition, our approach designs dynamic 3D object creation with minimal manual effort by unifying surface topology and optimizing data loading, enabling a streamlined reconstruction process and more flexible object projection. Our method supports rapid, resource-efficient mesh reconstruction and consistently demonstrates performance improvements across multiple instance benchmarks, covering virtual projection tasks. Improvements in mesh topology reconstruction, as measured by the L1 Chamfer distance (CD) metric, are consistently higher, while the system also achieves significant transmission speedups—up to 56.5×—near-instant importing—along with lowering latency in practical rendering on virtual reality (VR) devices. This result highlights that refining mesh binding improves re-creation fidelity. Our approach to scalability leads to faster user engagement and allows automated deployment without requiring human intervention during importing.
Some of the metrics are blocked by your
consent settings
Unknown Computerized-Attack Recognition via Open-Set Circumstance-Adaptive Learning Approach
(2025-01-01)
Petchhan, Jirayu
;
Kulworatit, Chaichana
Cyber threats come in all forms and constantly evolve into newly unknown malicious attacks. The status quo is that current mobile and/or embedded devices are unaware of embedded data with newly unknown threats, making it difficult to predict new threat types. To access much higher mobile security, we have deployed open-set domain adaptation to understand existing information yet still be realizable and recognizable to the novel unseen class instances. Our demonstrations are validated against visual benchmarks, such as handwritten data, and applied to our datasets to verify the embedding of unfamiliar data on embedded devices. As a result, it can gather a lot of information for both known and unknown instances from both datasets. Besides, it raises the issue of what cyber threat data to use in the circumstance like embedding over system-level firmware updates.
Some of the metrics are blocked by your
consent settings
Analysis of Urban Heat Islands Using Nighttime Land Surface Temperature Data
(2025-01-01)
Kulworatit, Chaichana
;
Petchhan, Jirayu
This study analyzes Urban Heat Island (UHI) in Thailand using Nighttime Land Surface Temperature (LST) data from the MODIS MYD11A2 dataset, covering 2002-2025, during the winter months (December-January). The research aims to assess heat accumulation in urban areas and identify factors such as heat-absorbing materials, energy consumption, and the lack of green spaces. It also addresses the misconception that water bodies reduce heat, showing that they can contribute to UHI when not managed properly. The data was processed using Google Earth Engine and QGIS to identify UHI zones and calculate Zonal Statistics at the provincial level.
Some of the metrics are blocked by your
consent settings
Better Learning to Drive Autonomously with Proximal-Policy Reinforcement Learning and Visual Perception Representations
(2025-01-01)
Sittigorn, Jirasak
;
Tungtrakool, Racha
;
Petchhan, Jirayu
This study presents an approach based on a deep reinforcement learning framework for vision-based autonomous driving in the CARLA environment, focusing on urban driving tasks. Our study implements a sub-policy Proximal Policy Optimization (PPO) algorithm, demonstrating its effectiveness in navigating complex scenarios including lane following, straight driving, and left/right turns, and outperforming a single-policy approach for intersection maneuvers. To enhance learning efficiency, our representation learning leverages a deep mobile network for state representation, which significantly reduces image feature complexity. Furthermore, the integration of a single-shot multi-box detector enables the agent to perform realistic tasks such as responding to traffic lights and maintaining safe distances from leading vehicles, without compromising training speed. While the system demonstrates stable driving in various scenarios, current limitations include handling highly complex decisions and adapting to diverse speed limits due to environmental constraints. Future work will focus on expanding training environments and exploring more advanced network architectures to improve real-world applicability and learning efficiency.
Some of the metrics are blocked by your
consent settings
Enhancing X-Resolution Integration through Federated Learning-based Building Learning Ecosystems
(2025-01-01)
Hongsuwan, Thana
;
Rueangaram, Chaksawat
;
Kongsmoot, Thanith
;
Petchhan, Jirayu
The challenges of running federated learning (FL) in resource-constrained browser environments, such as limited GPU access and interruptions, are caused by tab closures or page refreshes. To address this, the FLEXIBLE learning ecosystem introduces a browser-based federated learning system that enables decentralized model training without sharing raw data, preserving user privacy and enhancing accessibility. Built on a microservices architecture with Kafka for asynchronous, fault-tolerant communication, the system allows users to participate without software installation. Experiments on a super-resolution task reveal performance disparities across devices - mid-range performance mobile laptops achieved complete zero-shot super-resolution (ZSSR) training-inference cycles under 3 seconds, while mobile devices took up to 13.78 seconds. Notably, image patching proved essential for enabling large-image processing in browsers, offering a key insight into optimizing FL in low-resource environments.
Some of the metrics are blocked by your
consent settings
Modular LLM Architecture with Pluggable Reasoning Heads: A Scalable Approach to Multi-Modal AI Reasoning
(2025-01-01)
Pathak, Anurag
;
Sharma, Dilip Kumar
;
Agrawal, Harshada
;
Chawuthai, Rathachai
;
Petchhan, Jirayu
This paper introduces a modular architecture for Large Language Models (LLMs) that incorporates pluggable, domain-specific reasoning heads to augment the model's capabilities beyond conventional text generation. Central to our approach is an attention-routing controller that intelligently classifies and dispatches user prompts to appropriate reasoning modules - namely symbolic, logical, or graph - based heads-based on the prompt's structure and intent. This design enables hybrid, multi-paradigm reasoning without requiring retraining or finetuning of the base LLM. By decoupling reasoning tasks from general language understanding, our system improves both computational efficiency and interpretability. We demonstrate the architecture using Groq as the base LLM and integrate lightweight engines such as SymPy for symbolic mathematics and custom modules for logical and graph-based reasoning. Experiments across a suite of structured and unstructured prompts show a significant reduction in inference latency and token usage, along with higher accuracy and better explainability. The proposed framework offers a scalable foundation for embedding modular reasoning capabilities into modern LLM-driven applications.
Some of the metrics are blocked by your
consent settings
Detecting AI-Generated Scientific Abstracts Using Galactica and Graph Neural Networks
(2025-01-01)
Pathak, Anurag
;
Sharma, Dilip Kumar
;
Agrawal, Harshada
;
Chawuthai, Rathachai
;
Petchhan, Jirayu
The rise of large language models has introduced new challenges in maintaining research integrity, particularly through the potential proliferation of AI-generated scientific content. This paper presents a novel hybrid framework that combines Galactica-a scientific language model developed by Meta AI-with Graph Neural Networks (GNNs) to detect AI-generated research abstracts. Leveraging the AI-GA dataset comprising 28,662 labeled abstracts, we extract domain-specific semantic embeddings using Galactica and construct a semantic similarity graph to approximate citation-like relationships. A two- layer GCN is then trained to classify each abstract as human- or AI-authored. Experimental results demonstrate that our method outperforms traditional baselines such as TF-IDF, RoBERTa, and perplexity-based detectors, while offering interpretable and scalable detection suitable for editorial screening pipelines.
Some of the metrics are blocked by your
consent settings
PLC-Based Wheelchair Control with Integration of the Internet of Things
(2018-10-15)
Gulpanich, Suphan
;
Petchhan, Jirayu
;
Wongvanich, Napasool
This paper presents a PLC-based mobile mecanum-wheeled wheelchair control algorithm with the integration of the Internet of Things (IoT). The design aims to enhance comfort to the handicapped by offering wireless wheelchair control in addition to the Human Machine Interface (HMI) over the Ethernet port. The bottom-up approach was taken where proportional mobility control in the vector form of speed and direction is firstly configured and displayed over the touch screen. Remote access is then designed to access through the use of the HMI Remote Viewer application which supports and displays the corresponding speed and direction on a smartphone, upon wireless communication. Results show that the communication speed is within the rapidity range, whereas the locomotion test shows an average of 8.5% drop in speed from the no-load case to the loaded case. The proposed design thus provides a complete range of wheelchair control which can be used to assist the handicapped.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results