KMITL

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

Browse

Settings

Subject: search.filters.subject.Detection

Search Results

Now showing 1 - 5 of 5
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Eye Landmarks Detection using RT-DETR with Rules
    (2024-01-01)
    Boonnithititikul, Chatree
    ;
    Jaknamon, Teetouch
    ;
    Chawuthai, Rathachai
    In order to help ophthalmologists diagnose eye problems, it is necessary to scan for eye landmarks such as the pupil, the reflection point on the retina, and the boundary of the eye. An individual's eye landmarks on their face can be obtained via some facial landmarks' detection methods, including Haar Cascade. Two problematic aspects of the current approaches, however, are that the pupil and reflection point information is not provided, and the detection is ineffective when confronted with a picture of the upper half of the face or a person wearing a mask. In this study, we intend to develop a deep learning model for eye landmark identification using the Realtime identification Transformer (RT-DETR) approach together with our rules. As a consequence, nine landmark points-two for the eye, six for the pupil, and one for the reflection, are computed with an accuracy of 0.974. Since the focus of this paper is on eye landmark recognition, the next stage will be to build an application and a machine learning model for the diagnosis of eye disorders. - Keywords Deep Learning, Detection, Eye Landmarks, Facial Landmarks, Ophthalmology, RT-DETR
      1
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Evaluation of Deep Transfer Learning Models in Glaucoma Detection for Clinical Application
    (2021-01-01)
    Phankokkruad, Manop
    The clinical information supports the doctors in diagnosing the diseases and making the right decisions. Glaucoma is the leading cause of irreversible blindness disease. Vision loss can be avoided by early stage detection and right treatment. This study has proposed the deep transfer learning of the CNN model for detecting the glaucoma using ResNet50V2, VGG16, InceptionV3, and Xception. The proposed models help in the diagnosis of the patients who have glaucoma. The model with CNN architecture was used to learn from training the Glaucoma image dataset. Since the existing dataset has a small number of images, this study uses the data augmentation techniques to increase the virtual number of images. The results reveal that the proposed models have performed the classification task for detecting glaucoma. The proposed model achieved an accuracy level of VGG16, RestNet50V2, InceptionV3, and Xception are 97.27%, 94.53%, 95.31%, and 94.92%, respectively. Furthermore, this study evaluates the models by considering the clinical performance parameters include accuracy, precision, specificity, sensitivity, and F1 score. All models provide the high confidence values. The evaluation reveals that the deep transfer learning model with VGG16 architecture is the highest performance in tests. The VGG16 model achieved the average AUC-ROC value of 98.94%.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Evaluation of NID2 promoter methylation for screening of Oral squamous cell carcinoma
    (2020-03-14)
    Srisuttee, Ratakorn
    ;
    Arayataweegool, Areeya
    ;
    Mahattanasakul, Patnarin
    ;
    Tangjaturonrasme, Napadon
    ;
    Kerekhanjanarong, Virachai
    Background: Oral squamous cell carcinoma (OSCC) is an aggressive human malignancy. Because of late diagnosis and recurrence of OSCC, the treatment of patients with OSCC is often ineffective. Thus, finding novel biomarkers of OSCC are essential. Here we derived a methylation marker by utilizing methylation microarray data and testing its capacity in cross-sectional study designed for OSCC detection and screening. Methods: According to bioinformatics analysis of total of 27,578 cg sites, cg22881914 of Nidogen 2 (NID2) methylation was selected for evaluation. Next, we confirmed the methylation status by bisulfite sequencing from the microdissected OSCC cells in comparison with the microdissected oral epithelia. Subsequently, we developed a simple technique using real-time PCR with the specific probe to examine the ability for the detection of OSCC in the oral epithelial samples, which included 103 oral rinse and 82 oral swab samples. Results: Based on the comparison of microdissected tissue, cg22881914 of NID2 was proved to be methylated in most OSCC cells but unmethylated in the normal oral epithelia. Furthermore, the methylated NID2-relied quantitative PCR approach has demonstrated that this marker assists in distinguishing among patients with OSCC from normal oral epithelia, smokers, and patients with oral lichen planus using the non-invasive oral rinse and swab samples. Conclusions: Specific methylation at cg22881914 of NID2 of OSCC could be used as an important potential marker for detecting OSCC. Thus, to certify the utility of this marker, further studies with a larger sample size are needed.
      2
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Classification of file duplication by hierarchical clustering based on similarity relations
    (2018-06-21)
    Phankokkruad, Manop
    This paper have proposed the classification of the duplicate file by measuring the similarity score between the couple of files. This work examined the distance between the pairwise of files by the Smith-Waterman algorithm. In addition, the make use of the Euclidean distance matrix could identify the relativity between the persons who often copies the files each other. Since the regularity of the duplication happens, this work could classify the proximity to the persons, and a group of person who positioned closely together by applying the hierarchical clustering. The result revealed that the Smith-Waterman algorithms could measure the similarity between files effectively. Also, this work could analyze the relativity of the persons, classifies the person who positioned closely together, and the person between nearest related members of the group. Finally, this work represented the amount of time that person duplicated the files.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Effect of variation factors on the processing time of the face recognition algorithms in video sequence
    (2016-12-23)
    Phankokkruad, Manop
    ;
    Jaturawat, Phichaya
    Face recognition is one of the most popular research topics of the biometric identification. Sometimes, face recognition work under the unstable environments. Therefore, the study on the factors that affect face recognition performance is the very interesting topic. The recognition accuracy can be affected by many factors, but the quality of image database and system environmental factors, are mainly factors that need to be focused. This research creates the image database to represent each user by controlled the environmental factors strictly, to ensure that the images have minimum confounding and have the best quality. The performance of the processing was tested by the variation of factors. These are variant variables that using for testing the algorithms accuracy and processing speed. The standard processing times are done on the still images. In the video, the experiment was conducted by using the test set with variation of confounding factors. All factors are similar to the still images, but it has one addition factors that is video resolution. The results found that, the face recognition processing time, the factors that are face recognition algorithms, video resolution, and the number of input image affected to the processing speed. As results from the experiment, it is relevant for the hypothesis that is the smaller number of input image, and smaller image size can reduce the processing time. In the case of face recognition algorithm, Fisherfaces is the fastest algorithms which took a shorter processing time than Eigenfaces, and LBPH, respectively. The results from this experiment can indicate the most effective face recognition algorithm. After that, it would be applied for the as-is sys- tem.