Titijaroonroj, Taravichet

Preferred name

Alternative Name

Titijaroonrog, Taravichet

Main Affiliation

School of Information Technology

taravichet.ti@kmitl.ac.th

Now showing 1 - 10 of 32

Some of the metrics are blocked by your
consent settings
Automatic Thai Ticket Classification By Using Machine Learning For IT Infrastructure Company
(2022-01-01)
Khowongprasoed, Kraidet
;
Titijaroonroj, Taravichet
Ticket classification is a process to define the category name of each ticket before assigning the resolution team to serve each ticket. It is an important process to support the customers inside and outside the company. It can make customer dissatisfaction if the processing time is high or delayed. Based on the recording data in 2019-2021 at the studying company, we found that the manual ticket classification got an error rate about 53 percent because the office workers misunderstand. To alleviate this problem, we propose the methodology for automatic Thai ticket classification by using Term Frequency-Inverse Document Frequency with Support Vector Machine. The experimental result shows that the performance of the proposed methodology is higher than the manual classification by 2 times or 41 percent.
Some of the metrics are blocked by your
consent settings
Proposed Structural Validation-based Testing for Object-Oriented Programming
(2021-01-21)
Lumyong, Chayanin
;
Tachasriburapha, Natthawut
;
Titijaroonroj, Taravichet
Nowadays, Java program testing is widely used in an academic e-learning system. It can help a lecturer to give feedback and suggestion to learners automatically and immediately. Especially, an object-oriented programming (OOP) class should have a system to support learners. It is difficult and labor-intensive, if there are many learners. The test case-based approach is widely used in programming testing, because they are simple and convenient. This approach requires input-output pairs to evaluate the submitted code by comparing the output from their code with the expected output. However, it considers the output only, thus making it not appropriate for an OOP class. To develop the Java program testing to support an OOP class, a proposed structural validation-based testing (PSVT) is proposed. The experiments showed that our method can evaluate the correctness and the relation validity of the given code which is corresponding to the human. Moreover, its complexity is O(N^{2}).
Some of the metrics are blocked by your
consent settings
Seven Segment Display Detection and Recognition using Predefined HSV Color Slicing Technique
(2019-07-01)
Popayorm, Sorawee
;
Titijaroonroj, Taravichet
;
Phoka, Thanathorn
;
Massagram, Wansuree
Detection and recognition of LED seven segment panels present a particular challenge of locating characters among background clutter. This study proposed a framework based on a predefined HSV color slicing technique. The results demonstrate the framework's superiority over other color slicing methods with 94.46% precision, 92.24% recall, and 87.17% accuracy rates. The digit detection and recognition with the proposed predefined HSV color slicing is simple and straightforward-making it an attractive solution for the future deployment for edge-computing.
Some of the metrics are blocked by your
consent settings
Enhancing HR Support in a Thai Organization with LLM-Based Question Answering
(2026-01-01)
Taemkaeo, Chinnatip
;
Saetia, Chanatip
;
Chalothorn, Tawunrat
;
Titijaroonroj, Taravichet
Large language models (LLMs) enhanced with retrieval-augmented generation (RAG) and multimodal inputs are increasingly used as interfaces to organizational knowledge. However, their effectiveness in specialized, non-English enterprise settings-such as Thai HR support-remains largely unclear. In many Thai organizations, employees frequently ask detailed HR-related questions, but the relevant information is scattered across internal webpages, PDF manuals, announcements, and images, making it difficult for generic LLMs to provide accurate, policy-consistent responses. To address this issue, we develop a multimodal RAG pipeline that combines hybrid dense-sparse retrieval over a vector database and evaluate six LLM models on a private Thai Visual Question Answering (VQA) HR dataset consisting of 226 questions and reference images across five HR topics. The results show that recent multimodal models, especially Qwen2.5-VL, achieve the best performance, with the highest averages in correctness (0.54), relevance (0.75), and helpfulness (0.64), clearly outperforming older vision-language systems and a text-only reasoning model. For large-scale answer evaluation, we apply an LLM-as-a-judge approach using GPT-4.1 and Gemini 2.5 Flash. We found that it serves as a generally reliable, though imperfect, substitute for human evaluation.
Some of the metrics are blocked by your
consent settings
Image Denoising Based on Structural BIMF
(2018-08-21)
Titijaroonroj, Taravichet
;
Woraratpanya, Kuntpong
Image denoising is an important process for image analysis. It is always a pre-processing step before feeding its result to the next process. Although many papers based on statistical filter are presented for noise reduction, those still need to improve the quality of image. Bidimensional empirical mode decomposition (BEMD) is an adaptive image analysis method without a prior function for non-linear and non-stationary images in many applications. One of them is image denoising. The BEMD method can reduce the information loss by separating the noised and fundamental components into different bidimensional intrinsic mode function (BIMF) components. This makes it to be effective and flexible method. Therefore, this paper proposes image denoising based on structural BIMF. The proposed method not only removes the noise from noised image but also retrieves the main structure from noised image. Based on 44 noised images, the experimental results demonstrate that the performance of the proposed method outperforms that of baseline methods in terms of image quality assessments: PSNR, Qi, and TAA.
Some of the metrics are blocked by your
consent settings
Modified Scale-Space Analysis in Frequency Domain Based on Adaptive Multiscale Gaussian Filter for Saliency Detection
(2019-07-01)
Jaemsiri, Jenjira
;
Titijaroonroj, Taravichet
;
Rungrattanaubol, Jaratsri
The salient region is an area in the image that is paid attention from a human. It is a distinctive feature from neighbors such as color, shape or pattern. Saliency detection is a model that imitates the human visual system to perceive the scene. It has been widely used in many vision systems. Many papers use the smoothing or suppressing technique to extract a desirable output as so-called saliency map. Even though most of these researches achieve saliency detection based on the filter, the size of the filter is fixed. This leads to a filter ineffective when applying to the whole area of each image. In order to solve this issue, an adaptive multiscale Gaussian filter (MSS) for scale-space analysis in the frequency domain is proposed. The proposed filter is extended from an adaptive median filter which is a powerful method to remove the noise from the input image. This paper proposes the method that offers the appropriate filter to suppress the repeated pattern spectrum of each region in each image before extracting the saliency map. The experimental result shows that the proposed method outperforms the baseline methods, which includes HFT, Itti, SAL, SR and SUN.
Some of the metrics are blocked by your
consent settings
Adaptive Histogram of Oriented Gradient for Printed Thai Character Recognition
(2014-01-01)
Woraratpanya, Kuntpong
;
Titijaroonroj, Taravichet
A similarity of printed Thai characters is a grand challenge of optical character recognition (OCR), especially in case of a variety of font types, sizes, and styles. This paper proposes an effective feature extraction, adaptive histogram of oriented gradient (AHOG), for overcoming the character similarity. The proposed method improves the conventional histogram of oriented gradient (HOG) in two principal phases, which are (i) adaptive partition for gradient images and (ii) adaptive binning for oriented histograms. The former is implemented with quadtree partition based on gradient image variance so as to provide for an effective local feature extraction. The later is implemented with non-uniform mapping technique, so that the AHOG descriptor can be constructed with minimal errors. Based on 59,408 single character images equally divided into training and testing samples, the experimental results show that the AHOG method outperforms the conventional HOG and state-of-the-art methods, including scale space histogram of oriented gradient (SSHOG), pyramid histogram of oriented gradient (PHOG), multilevel histogram of oriented gradient (MHOG), and HOG column encoding algorithm (HOG-Column). © Springer International Publishing Switzerland 2014.
Some of the metrics are blocked by your
consent settings
Skin Video-based Blood Pressure Approximation Using CHROM with LSTM-NN
(2023-01-01)
Lumyong, Chayanin
;
Yodrabum, Nutcha
;
Winaikosol, Kengkart
;
Titijaroonroj, Taravichet
The measurement of blood pressure (BP) is an essential step in clinical practice. It is used to determine the patient's BP, which reflects the condition of the patient. Recently, there is a solution for extracting, non-invasively and with no contact, a blood pressure indicator from electrical signal like Photoplethysmography (PPG), called remote-Photoplethysmography (rPPG). This rPPG signal can be used to estimate from a video clip several vital physiological indicators for humans, especially, systolic blood pressure (SBP), diastolic blood pressure (DBP), and mean arterial pressure (MAP). This paper proposed a computer method for blood pressure approximation from an input video. A chrominance method, or CHROM, was used to extract rPPG signal from a given video before forwarding it to estimate SBP and DBP values by LSTM-NN. Afterwards, MAP value was determined from SBP and DBP values by a weighting score technique. Experimental results showed that CHROM achieved the lowest mean absolute error (MAE) at 14.04, 8.37, and 9.78 for the SBP, DBP, and MAP, respectively, when compared among NN, RNN, and GRU.
Some of the metrics are blocked by your
consent settings
Comparative performance of deep learning models and non-dermatologists in diagnosing psoriasis, dermatophytosis, and eczema
(2026-12-01)
Yodrabum, Nutcha
;
Wongpraparut, Chanisada
;
Titijaroonroj, Taravichet
;
Chularojanamontri, Leena
;
Bunyaratavej, Sumanas
Accurately differentiating scaly erythematous rashes among psoriasis, eczema, and dermatophytosis remains a clinical challenge, particularly for non-dermatologists. This study aimed to develop and evaluate deep learning models using macroscopic clinical images to classify these conditions and compare their performance with that of non-specialists. A total of 2940 images were sourced from public datasets, the Siriraj Dermatology databank, and newly collected images from Thai participants. Among sixteen evaluated models, the Swin demonstrated the best performance and interpretability. Gradient-weighted Class Activation Mapping (Grad-CAM) visualizations confirmed that the model focused on clinically relevant lesion features. Most importantly, in a pilot comparison, the Swin outperformed non-specialists in diagnostic accuracy. However, given the limited sample size of 30 images and 30 evaluators, these results should be interpreted as exploratory. Future studies with larger datasets and diverse clinician cohorts are warranted to confirm these findings and to support clinical integration.
3
Some of the metrics are blocked by your
consent settings
Automatic Lymph Node Classification with Convolutional Neural Network
(2022-01-01)
Uthatham, Ason
;
Yodrabum, Nutcha
;
Sinmaroeng, Chanya
;
Titijaroonroj, Taravichet
Manual lymph node classification is a tedious and time-consuming task. It requires a histopathologist to discriminate a lymph node from other look-alike kinds of tissues. The lymph node is easily misunderstood with other tissues because its shape and color might be similar to the others tissue around it. To automate this task, we present an automatic lymph node classification with convolutional neural network (CNN). In addition, we compared eight existing CNNs to ensure that we discover the best architecture for discriminating lymph node. DenseNet architecture provided the highest performance among AlexNet, VGG, GoogLeNet, ResNet, SqueezeNet, MobileNet, and EfficientNet, the highest accuracy at 0.994 and an F1score of 0.996. DenseNet accomplished the highest performance from two advantages: (i) fewer parameters and (ii) Dense connectivity.
1

Titijaroonroj, Taravichet

Filters

Settings

Sort By

Results per page