KMITL

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

Browse

Search Results

Now showing 1 - 10 of 20

Some of the metrics are blocked by your
consent settings
Near-Infrared Hyperspectral Imaging to Predict Intact Sweet Tamarind Fruit Quality
(2026-07-01)
Sahachairungrueng, Woranitta
;
Aozora, Wayan Dipasasri
;
Tantinantrakun, Achiraya
;
Suwapanich, Rachit
;
Workhwa, Saranya
The quality of sweet tamarind fruit, as determined by its total soluble solids (TSS), titratable acidity (TA), and TSS/TA ratio, is important for consumer satisfaction. Nondestructive techniques are therefore required to assess the quality of sweet tamarind fruit. This study investigated whether near-infrared hyperspectral imaging (NIR-HSI) in the wavelength range of 935–1720 nm can be used as a non-destructive method to assess TSS, TA, and the TSS/TA ratio of sweet tamarind fruit and to classify it under commercial standards. NIR-HSI-based chemometric and machine-learning modeling was applied for quantification and qualification analyses. Calibration models for determining TSS, TA, and the TSS/TA ratio were developed using partial least squares regression (PLSR) and support vector machine regression (SVMR). A combination of first derivative and SNV spectral pretreatment was optimized to establish an SVMR model for TSS determination. MSC spectral pretreatment was optimized to develop the SVMR model for TA assessment, and the first derivative spectral pretreatment was optimized to establish an SVMR model for the TSS/TA ratio. Correlation coefficients of prediction (R<inf>p</inf>) of 0.959, 0.961 and 0.957 were obtained with root mean square errors of prediction (RMSEP) of 1.102%, 0.369% and 7.850, and a ratio of performance to deviation (RPD) of 3.29, 3.52 and 3.34 for the TSS, TA, and TSS/TA ratio evaluations, respectively. Partial least squares–discriminant analysis (PLS-DA) and support vector machine classification (SVMC) were used for classifying sweet tamarind fruit under a commercial acidity standard (≤4%). The SVMC with SNV spectral pretreatment produced the best prediction results for distinguishing standard and off-standard sweet tamarind fruit with an 82.86% accuracy. NIR HSI can be used to non-destructively predict the quality of tamarind fruit. It can be applied for online sorting to evaluate individual sweet tamarind fruits for grading and quality control in factory environments.
Some of the metrics are blocked by your
consent settings
A Review of Thin-Film Growth, Properties, Applications, and Future Prospects
(2025-02-01)
Sakthinathan, Subramanian
;
Meenakshi, Ganesh Abinaya
;
Vinothini, Sivaramakrishnan
;
Yu, Chung Lun
;
Chen, Ching Lung
This review article’s primary aim is to discuss different thin-film deposition technique methods and their important uses. The histories of thin-film technology, thin-film growth, thin-film classification, and thin-film preparation techniques are also covered in this review article. The preparation and characterization of functional thin films and nanostructured materials, as well as various devices based on these materials and recent developments are also focused on in this review. The properties of the materials and several thin-film deposition techniques are also covered in this article. This review article also discusses the classification and application of thin-film sensors. Furthermore, the formation of thin films and their physical properties are impacted by deposition conditions such as pH, temperature, deposition time, and deposition parameters, which are analyzed. This article discusses how a wide range of potential uses in structural, mechanical, and protective coatings; sensing; energy storage systems; catalysis; optoelectronics; and biomedicine are made possible by the special qualities of thin films and nanostructured materials, including their high surface area to volume ratio, structure, surface charge, anisotropic nature, and tunable functionalities.
Some of the metrics are blocked by your
consent settings
Classification of Thai Rice Varieties Using Image Processing and Deep Learning Techniques
(2025-01-01)
Kongmanee, Panpatsorn
;
Puengpradith, Sorapojana
;
Boongasame, Laor
The methods for identifying Thai rice varieties are complex, time-consuming, and require high expertise to achieve accurate results. This research explores different deep learning techniques to efficiently classify the strains of Thai rice that optimize accuracy and speed. The focus rice varieties are Khao Hom-Mali Thai and Thai Hom Pathum Thani 1 fragrant rice; both have similar shapes and characteristics but differ in price, market value, and recognition. The proposed model is based on an instance segmentation model of YOLOv8, which is compared against popular instance segmentation models such as YOLACT, SOLOv2, and Mask R-CNN. Additionally, hyperparameter tuning is performed to ascertain the most optimal values. The evaluation of the model performance reports in the form of mean average precision (mAP), inference time, and model stability. Experimental results indicate that YOLOv8n-seg, with the fewest parameters, achieves the highest accuracy comparable to other YOLOv8-based models with more parameters. The proposed model demonstrates superior accuracy and processing speed performance compared to other state-of-the-art models.
Some of the metrics are blocked by your
consent settings
Investigation variable star classification through light curve analysis using machine learning approach
(2025-01-01)
Tongleak, Chutipon
;
Thongsuwan, Setthanun
;
Srithongtae, Kewalee
;
Kitrattana, Borirak
;
Tanirat, Purin
With the development of space technology, wide-field sky surveys using telescopes have expanded the range of new data available for time-domain astronomical research. Traditional data analysis methods can no longer respond quickly and accurately enough to the growing volume of data. Thus, classifying time-series data, such as light curves, has become a significant challenge in the era of big data. In modern times, analyzing light curves has become essential for using machine learning techniques to handle and filter through massive amounts of data. Machine learning algorithms can be divided into two categories: shallow learning and deep learning. Numerous researchers have proposed and developed a variety of algorithms for light curve classification. In this study, we experimented with Support Vector Machine (SVM) and XGBoost, which are shallow machine learning algorithms, as well as 1D-CNN and Long Short-Term Memory (LSTM), which are deep learning algorithms, which are branches of deep machine learning, to classify variable stars. The training and testing data used in this study were from the Optical Gravitational Lensing Experiment-III (OGLE-III), consisting of variable star data from the Large Magellanic Cloud (LMC), categorized into five main classes: Classical Cepheids, δ Scutis, eclipsing binaries, RR Lyrae stars, and Long-period variables. The results demonstrate the performance analysis of each machine learning algorithm type applied to light curve data, while also highlighting the accuracy and statistical metrics of the algorithms used in the experiments.
Some of the metrics are blocked by your
consent settings
Enhanced Feature Selection via Hierarchical Concept Modeling
(2024-12-01)
Saelee, Jarunee
;
Wetchapram, Patsita
;
Wanichsombat, Apirat
;
Intarasit, Arthit
;
Muangprathub, Jirapond
The objectives of feature selection include simplifying modeling and making the results more understandable, improving data mining efficiency, and providing clean and understandable data preparation. With big data, it also allows us to reduce computational time, improve prediction performance, and better understand the data in machine learning or pattern recognition applications. In this study, we present a new feature selection approach based on hierarchical concept models using formal concept analysis (FCA) and a decision tree (DT) for selecting a subset of attributes. The presented methods are evaluated based on all learned attributes with 10 datasets from the UCI Machine Learning Repository by using three classification algorithms, namely decision trees, support vector machines (SVM), and artificial neural networks (ANN). The hierarchical concept model is built from a dataset, and it is selected by top-down considering features (attributes) node for each level of structure. Moreover, this study is considered to provide a mathematical feature selection approach with optimization based on a paired-samples t-test. To compare the identified models in order to evaluate feature selection effects, the indicators used were information gain (IG) and chi-squared (CS), while both forward selection (FS) and backward elimination (BS) were tested with the datasets to assess whether the presented model was effective in reducing the number of features used. The results show clearly that the proposed models when using DT or using FCA, needed fewer features than the other methods for similar classification performance.
Some of the metrics are blocked by your
consent settings
Differential Evolution for Classification: A Novel Classifier Technique in Data Mining
(2024-01-01)
Farda, Irfan
;
Thammano, Arit
In the realm of Data Mining, the pursuit of innovative classification methodologies remains crucial for advancing robust techniques in handling complex and diverse datasets. This paper explores the application of Differential Evolution (DE), a powerful optimization algorithm, as a unique and effective optimization-based standalone classifier. Our exploration focused on harnessing DE's intrinsic capabilities, adapting it into a classifier while preserving its distinctiveness. The fundamental principle of DE for classification involves iteratively optimizing the center point for each class using DE operators, which are specialized mechanisms for exploring and refining solutions, and utilizing these optimized points for making predictions. The effectiveness of our model was evaluated on ten classification datasets from the UCI Machine Learning Repository and compared against three other classification methods: KNN, ZMP, and BPNN. Experimental results underscore the competitive performance of our proposed model, emphasizing the potential of DE in effectively addressing classification challenges.
Some of the metrics are blocked by your
consent settings
Partial Discharge Classification with Transformer Neural Networks
(2024-01-01)
Cheypoca, Thepjit
;
Promphanich, Wiboon
;
Thway, Aung Ye
;
Hankae, Angelina Phimpissada
;
Jeenmuang, Siwakorn
This paper introduces an approach with the Transformer Neural Networks model for partial discharge patterns classification, that consists of corona discharge, internal discharge and surface discharge. The PD measuring circuit suggested in IEC 60270:2000 is used to record Partial discharge signals. Independent parameters such as phase and charge of PD patterns were recorded. The phase value will be encoded into the charge array and Transformer Neural Network is constructed using Positional Embedding and Transformer Encoder Layer. 80% of the recorded data will be used as a training data and 20% recorded data was used for testing of the classification models. Impacts of neuron numbers and network architecture on the PD classification performance will be observed
Some of the metrics are blocked by your
consent settings
Partial Discharge Classification With 1D Convolutional Neural Network
(2024-01-01)
Cheypoca, Thepjit
;
Promphanich, Wiboon
;
Thway, Aung Ye
;
Hankae, Angelina Phimpissada
;
Jeenmuang, Siwakorn
This paper introduces a novel approach for classifying with the 1D Convolutional Neural Network model for partial discharge patterns, that consists of corona discharge, surface discharge and internal discharge. The PD measuring circuit suggested in IEC 60270:2000 is used to record Partial discharge signals. Independent parameters such as phase and charge of PD patterns were recorded. The Artificial Neural Network for the classification model was constructed. Moreover, 2×1D CNN feature extraction was utilized to reduce the curse of dimensionality in the dense layer of the proposed PD classification model. 80% of the recorded data will be used as a training data and 20% recorded data was used for testing of the classification models. Impacts of neuron numbers and network architecture on the PD classification performance will be observed.
Some of the metrics are blocked by your
consent settings
Assessment of Nitrite Content in Vienna Chicken Sausages Using Near-Infrared Hyperspectral Imaging
(2023-07-01)
Tantinantrakun, Achiraya
;
Thompson, Anthony Keith
;
Terdwongworakul, Anupun
;
Teerachaichayut, Sontisuk
Sodium nitrite is a food additive commonly used in sausages, but legally, the unsafe levels of nitrite in sausage should be less than 80 mg/kg, since higher levels can be harmful to consumers. Consumers must rely on processors to conform to these levels. Therefore, the determination of nitrite content in chicken sausages using near infrared hyperspectral imaging (NIR-HSI) was investigated. A total of 140 chicken sausage samples were produced by adding sodium nitrite in various levels. The samples were divided into a calibration set (n = 94) and a prediction set (n = 46). Quantitative analysis, to detect nitrate in the sausages, and qualitative analysis, to classify nitrite levels, were undertaken in order to evaluate whether individual sausages had safe levels or non-safe levels of nitrite. NIR-HSI was preprocessed to obtain the optimum conditions for establishing the models. The results showed that the model from the partial least squares regression (PLSR) gave the most reliable performance, with a coefficient of determination of prediction (R<inf>p</inf>) of 0.92 and a root mean square error of prediction (RMSEP) of 15.603 mg/kg. The results of the classification using the partial least square-discriminant analysis (PLS-DA) showed a satisfied accuracy for prediction of 91.30%. It was therefore concluded that they were sufficiently accurate for screening and that NIR-HSI has the potential to be used for the fast, accurate and reliable assessment of nitrite content in chicken sausages.
Some of the metrics are blocked by your
consent settings
Image Enhancement and 27 Pretrained Convolutional Neural Network Models for Diabetic Retinopathy Grading
(2023-01-01)
Kanjanasurat, Isoon
;
Anuwongpinit, Thanavit
;
Purahong, Boonchana
Diabetic retinopathy (DR) affects the retina's blood vessels and causes vision loss. Fundus images are used to diagnose DR, which is a lengthy process because experienced clinicians must accurately diagnose the disease and identify microlesions early to prevent blindness. Computer vision can be used for retinal image classification. The APTOS dataset contains 5990 normal, moderate, mild, proliferate, and severe retinal images. In this study, we proposed a convolutional neural network (CNN) ensemble for DR fundus grading. Each image channel was enhanced by contrast-limited adaptive histogram equalization (CLAHE) and gamma correction and then fed to 27 pretrained CNN models for one-time training to examine the DR grading. The results showed that MobileNet's green channel with the CLAHE technique is sufficiently fast and accurate for disease classification. The grading retinal images had an accuracy of 96.95%, a precision of 96.17%, a sensitivity of 97.80%, an F1 score of 96.98%, and a specificity of 97.75%. In addition, the proposed method improves the speed and robustness of retinal DR grading.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results