Udomchaiporn, Akadej

Loading...
Thumbnail Image
Preferred name
Udomchaiporn, Akadej
Main Affiliation
Email
akadej.ud@kmitl.ac.th

Settings

Now showing 1 - 4 of 4
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Software requirements retrieval using use case terms and structure similarity computation
    (2006-12-01)
    Udomchaiporn, Akadej
    ;
    Prompoon, Nakornthip
    ;
    Kanongchaiyos, Pizzanu
    For a large scale of software development, there is a tremendous number of software requirements documents in a collection which may be produced for different domains by different developer teams. They may be later reused to reduce cost and time for the next development. Thus, there is a need to retrieve ones that meet user's need efficiently. This paper presents an approach for software requirements specification retrieval in a form of use case description using use case structure and similarity computation between terms of use case query and use cases in the collection. The contribution of the paper has five main points; 1) the approach for retrieving use case description is proposed, 2) the developed tool supporting the approach is presented, 3) the experiment is designed to measure effectiveness of the approach, 4) the results of the experiment are shown to compare effectiveness of the approach to that of a general approach, and finally 5) the recommended use case description query template is proposed. © 2006 IEEE.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Cross-Domain Robust Liveness Detection: A Transfer Learning Approach for Combating Sophisticated Presentation Attacks in Mobile Authentication
    (2025-01-01)
    Kerdpramote, Phuvis
    ;
    Poomdeesittinon, Akeanant
    ;
    Jamsri, Rapeeploy
    ;
    Krueyos, Phanuwit
    ;
    Muangkan, Ronnakorn
    As biometric authentication systems become ubiquitous in Southeast Asia's digital economy, sophisticated presentation attacks using deepfakes, high-resolution displays, and 3D masks pose critical security threats. This paper presents a comprehensive cross-domain liveness detection framework that addresses the generalization challenges plaguing current systems. Our approach leverages MobileNetV2-based transfer learning with a novel two-phase training strategy, achieving superior cross-domain performance while maintaining computational efficiency for mobile deployment. We introduce domain-aware augmentation techniques and evaluate our system across multiple benchmark datasets including NUAA and a locally-collected Thai demographic dataset. Experimental results demonstrate 84.35% accuracy on NUAA and 78.62% cross-domain accuracy, with significant improvements in Attack Presentation Classification Error Rate (APCER) reduction from 28.7% to 15.4% compared to baseline methods. The system successfully detects emerging attack vectors including deepfake videos and tablet-based spoofing attempts. We provide comprehensive analysis of deployment challenges in resource-constrained environments demonstrating practical applicability for Thailand's mobile banking and digital identity verification ecosystem.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    A search-free intersection algorithm
    (2008-05-27)
    Udomchaiporn, Akadej
    ;
    Boonjing, Veera
    This paper proposes a new intersection algorithm of sorted sets. The new algorithm employs a comparison-and-elimination approach to the intersection problem instead of using search algorithms as existing intersection solutions. It takes (1) O(1) times for the best case; (2) 0(λ) times for the average case, where λ is an average size of sorted sets; and (3) 0(kn) times for the worst case, where k is the number of sorted sets and n is the total elements of k sorted sets. © 2008 IEEE.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Ensemble Model for Segmentation of Lateral Ventricles from 3D Magnetic Resonance Imaging
    (2020-01-01)
    Udomchaiporn, Akadej
    ;
    Lertrungwichean, Khitichai
    ;
    Klinkasen, Pokpakorn
    ;
    Nuchprasert, Chawanwut
    The paper proposes an ensemble model to segment lateral ventricles from 3D Magnetic Resonance Imaging (MRI) brain scan. Thresholding and Active Contour techniques combining with a noise removal method were applied to segment lateral ventricles from the brain images. The experiments were conducted by segmenting 73 MRI brain scans using our proposed model and then comparing their volumes to those using manual model conducted by an expert. The experimental results indicated that the proposed model segmented lateral ventricles as excellent as the manual model in terms of accuracy but outperformed the manual model in terms of time performance. The contribution of the paper is that the segmented lateral ventricles can be used for further analysis such as medical condition classification.