KMITL

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

Browse

Search Results

Now showing 1 - 10 of 11

Some of the metrics are blocked by your
consent settings
Fractal Dimension Based Color Texture Analysis for Mangosteen Ripeness Grading
(2019-12-01)
Phothisonothai, Montri
;
Tantisatirapong, Suchada
Mangosteen is one of the fruits that has an enormous export potential in Thailand. However, it contains numerous undesirable external as well as internal conditions which result in the shipment rejection and decrease the reliability of the export. Therefore, in this paper for the first time, we propose the method for mangosteen ripeness grading using the spatiotemporal properties of external rind texture analysis on the basis of fractal dimension (FD) approach for three classes: i.e., Glossy (GS), Medium Rough (MR) and Extreme Rough (ER). In this study, for the first time, the five stages of ripening have been extracted using FD based feature with Gaussian Mixture Model (GMM) classifier. The obtained results showed that the proposed method can perform the better results compared with the classical texture feature, i.e., average accuracy rates of 88.0%, 82.0%, and 90.0% for GS, MR, and ER classes, respectively.
Some of the metrics are blocked by your
consent settings
Integrated Human-Machine Interaction System: ERP-SSVEP and Eye Tracking Based Technologies
(2019-04-10)
Phothisonothai, Montri
;
Tantisatirapong, Suchada
Human-Machine Interaction (HMI) requires a multidisciplinary research study mainly focused on interaction modalities between humans and machines. In this paper, we introduced an integrated HMI system using electrical brainwave signal (or called Electroencephalography: EEG) and eye tracking of pupil movement (or called ET), whose are used as an alternative channel to communicate with others for people with disabilities. In this experiment, the target and non-Target visual stimuli of EEG-based HMI system on the basis of event-related potential (ERP) and steady state visually evoked potential (SSVEP) signals have been performed. For the ET based framework, we proposed the user-friendly virtual keyboard typing for Thai language, i.e., free-form and automatic typing modes. The results showed that the integrated HMI using ERP-SSVEP yielded an average accuracy of 97.4% and reaction time approximately was 724.2 millisecond for control commands. The automatic typing mode performed an average accuracy of 97%, with an average printing time of 6.17 seconds per word for ET based virtual Thai keyboard.
Some of the metrics are blocked by your
consent settings
Design of User-Friendly Virtual Thai Keyboard Based on Eye-Tracking Controlled System
(2018-12-24)
Tantisatirapong, Suchada
;
Phothisonothai, Montri
Eye trackers are noninvasive and portable devices allowing the application of various assistive technologies. We aim to improve the quality of life of Thai people with disabilities whose eyes are the only mean of communication. Therefore, this paper presents a gaze-controlled virtual Thai keyboard. We compare two typing modes. automatic typing and free-form typing. A survey of often used words for Thai patients is conducted for word prediction system. The system is evaluated with 10 adult participants. The results show that the automatic typing provides faster responses and obtains higher user satisfaction than free-form typing.
Some of the metrics are blocked by your
consent settings
Comparison of feature extraction for accent dependent Thai speech recognition system
(2018-09-13)
Tantisatirapong, Suchada
;
Prasoproek, Chalisa
;
Phothisonothai, Montri
This paper aims to compare the feature extraction methods for accent dependent Thai speech from three regions including central, southern and northeastern regions. We investigate four frequency analysis methods: i.e., Energy Spectral Density (ESD), Power Spectral Density (PSD), Mel-Frequency Cepstral Coefficients (MFCC) and Spectrogram (SPT). Radial basis function kernel based on support vector machine is used as a classifier with 5-fold cross validation. The isolated speech data sets are recorded from 30 male and 30 female participants speaking the 10 Thai digits from 0 to 9. The MFCC-based feature gives better accuracy than ESD, PSD and SPT respectively. For within the same region, the MFCC-based feature provides average accuracy of 94.9% and 99.1% for male and female voices respectively. For the three regions, the MFCC-based feature provides average accuracy of 89.34% and 93.81% for male and female voices, respectively.
Some of the metrics are blocked by your
consent settings
Classification of in Vitro Blood Stages of Plasmodium Falciparum Based on Fuzzy Inference System
(2018-08-06)
Tantisatirapong, Suchada
;
Phothisonothai, Montri
This paper proposes the automated texture based classification of Malaria parasites in Giemsa-stained thin blood film images based on fuzzy inference system (FIS). The proposed expert and knowledge based framework includes the segmentation, feature extraction and classification of erythrocytes. First-order statistical analysis includes mean, standard deviation, skewness and kurtosis have been proposed as input parameters of FIS. The effectiveness of classifier is compared to find appropriate parame- ters for classification of normal cells and infected cells, both ring and trophozoite stages. The proposed method can provide 96.28% accuracy rate for binary classification of normal and infected cells. The results also yield 97.55% accuracy for ring stage classification, and 98.54% accuracy for trophozoite stage classification.
Some of the metrics are blocked by your
consent settings
Surface roughness classification of mangosteen with gray level co-occurrence matrix based texture analysis
(2018-07-02)
Acharya, Anjali
;
Phothisonothai, Montri
;
Tantisatirapong, Suchada
Mangosteen is one of the fruits that has an enormous export potential in Thailand. It is well-known as the queen of fruit. Mangosteen export generates large revenue; however, fruit is not defect free it contains many undesirable external as well as internal condition which results in the shipment rejection and decrease the reliability of the export. Therefore, this research investigates an approach for texture image analysis based surface roughness detection and classification into 3 classes: i.e., Glossy Surface, Mid Rough Surface and Extreme Rough Surface. In this study, for the first time, we propose the textural features extracted using Gray-Level Co-occurrence Matrix (GLCM) for surface roughness classification of mangosteen.
Some of the metrics are blocked by your
consent settings
Time-frequency based coherence analysis of red and green flickering visual stimuli for EEG-controlled applications
(2017-03-23)
Tantisatirapong, Suchada
;
Dechwechprasit, Panisa
;
Senavongse, Wongwit
;
Phothisonothai, Montri
The stimulus flickering at specific frequencies, or as known as steady-state visually evoked potential (SSVEP), can be recorded on an occipital area of the brain. SSVEP is used to interpret the EEG signal to detect the desired goal of the experiment. In this paper, we aim to investigate SSVE signal by means of magnitude-squared coherence (MSC) analysis between the red and green visual stimuli. In the experimental paradigm, we considered two parameters that are chromatic color and flickering frequency where an epoch interval was 10 seconds. The obtained results showed the statistically significant frequency-domain response and its maximum MSC coefficient in the theta and alpha bands for green and red flickers, respectively.
Some of the metrics are blocked by your
consent settings
Time-frequency analysis of red-green visual flickers based on steady-state visual evoked potential recording
(2017-02-21)
Dechwechprasit, Panisa
;
Phothisonothai, Montri
;
Tantisatirapong, Suchada
The study of brain activity can be done by visual stimulus flickering at specific frequencies, Steady-State Visual Evoked Potential or as known as SSVEP. SSVEP is to stimulate the EEG signal to locate the desired goal of the experiment when a visual stimulus flickering with different constant frequencies and same duration. We aim to present a case study of electroencephalogram (EEG) signal by analyzing the frequency response of the red and green light. The stimulation is based on SSVEP by dividing the trial into two trials: single light and two lights. We considered three parameters that are light color, frequency and epoch interval. The optimal experimental results showed the classification accuracy rate of 74% and 75% for single and two color lights, respectively. The results can be considerably applied to the brain-computer interface (BCI) system.
Some of the metrics are blocked by your
consent settings
Automated determination of watermelon ripeness based on image color segmentation and rind texture analysis
(2017-01-18)
Phothisonothai, Montri
;
Tantisatirapong, Suchada
;
Aurasopon, Apinan
Watermelons are popularly grown and consumed in most tropical areas of agricultural countries especially in the Asian countries. Quality control is important to standardize the production especially the procedure of automatic system based on computer vision. In this paper, therefore, we objectively investigated the ripeness of watermelon based on color segmentation using k-means clustering and rind texture analysis using Laplacian of Gaussian (LoG) filter. We captured each image of 20 watermelons (Kinnaree variety), which are divided into ten ripe and unripe groups by an experienced farmer. Different experimental conditions were compared to achieve the optimal outcome. The experimental results showed that the proposed features could extract different ripeness levels statistically with p < 0.001.
Some of the metrics are blocked by your
consent settings
Automated segmentation of erythrocytes from Giemsa-stained thin blood films
(2016-03-23)
Puttapirat, Pargorn
;
Phothisonothai, Montri
;
Tantisatirapong, Suchada
This paper investigates automated segmentation of malaria parasites in images of Giemsa-stained thin blood film specimens. The Giemsa staining exhibits not only on the malaria parasites, but also platelets and artifacts. We aim to extract erythrocytes both normal and infected cells from other particles and separate overlapping cells. Our approach is compared with manual cell counting and existing program named CELLCOUNTER. Our processing framework provides 97% accuracy, which yields predominant detection more accurate than the CELLCOUNTER. The results also indicate high correlation between our proposed method and the manual cell counting.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results