KMITL

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

Browse

Search Results

Now showing 1 - 6 of 6

Some of the metrics are blocked by your
consent settings
WILDFIRE CLASSIFICATION WITH DEEP LEARNING MODEL
(2026-07-01)
Tankue, Puwanai
;
Kruekaew, Boonhatai
;
Kimpan, Warangkhana
This research aims to develop models for wildfire classification from images. The objective is to provide decision-making support for wildfire control planning, prevention and management. Convolutional Neural Networks techniques are used to analyze a dataset consisting of three image groups: no fire images, fire images that are not wildfires, and wildfire images. The experimental results compared ResNet group, DenseNet group, MobileNet group, and EfficientNet group. The research findings indicate the best-performing model in this study is ResNet152V2, which achieved an accuracy of 92.75%. Furthermore, Precision, Recall, and F1-Score are within a satisfactory range. A web application has also been developed to facilitate users to detect and classify wildfire more conveniently.
Some of the metrics are blocked by your
consent settings
Improving the Sound Classification Accuracy Using CNN-LSTM and MFCC with Audio Augmentation for Diagnosing Respiratory Disease
(2025-01-01)
Phankokkruad, Manop
;
Wacharawichanant, Sirirat
Audio is vital information data for understanding various situations. A multitude of sound features can be explained by analysis through the audio signals. Numerous classification methods have been developed to study audio classification. This work studies the improvement of audio classification for the diagnosis of respiratory disease through the integration of audio data augmentation and CNN in conjunction with LSTM (CNN-LSTM). Furthermore, this paper focuses on audio data augmentation and feature extraction in the deep learning approach. This study proposed the CNN-LSTM model to diagnose respiratory disease by learning from the different audio datasets. The results reveal that the CNN-LSTM model attained an accuracy of 81.48%, precision of 0.8340, sensitivity of 0.6948, and F1-score of 0.7225. Considering the achieved F1-score, the CNN-LSTM model demonstrates a high level of diagnotic accuracy. Therefore, all evaluation evaluation parameters collectively indicate the robust performance of the proposed disease classification model.
Some of the metrics are blocked by your
consent settings
Gender classification using convolutional neural networks based on fingerprint analysis with in-line digital holography
(2024-01-01)
Thonglim, Pachara
;
Thongsuwan, Setthanun
;
Buranasiri, Prathan
Gender classification has found applications in various fields, including criminology, biometrics, and surveillance. Historically, different methods for gender identification have been employed, such as analyzing hand shape, gait, iris, and facial features. Fingerprints, being unique to each individual, are formed based on the control of multiple genes on chromosomes. After the 24th embryonic week, a person's fingerprint pattern remains unchanged throughout their life. Numerous studies have explored the use of fingerprints for various purposes, such as investigating mental characteristics, characteristics of hereditary diseases, and cancer screening. This paper focuses on studying fingerprints for the identification and classification of human gender through fingerprint analysis using in-line digital holography. The deep learning model constructed for this study includes two convolutional layers, pooling layers, and dense layers. It was trained on a biometric fingerprint database containing 6,000 images, achieving an impressive 99% accuracy. The model was then utilized to classify human gender based on fingerprint analysis, and its accuracy was tested using fingerprint images obtained through Inline Digital Holography (IDH) technique, achieving an 83% accuracy. The performance of the proposed system demonstrates that fingerprints contain vital features for effectively discriminating a person's gender.
Some of the metrics are blocked by your
consent settings
A Low-Cost Digital Stethoscope for Normal and Abnormal Heart Sound Classification
(2022-01-01)
Khoruamkid, Sorawit
;
Visitsattapongse, Sarinporn
Heart disease is a major problem in most deaths. To conquer this situation, heartbeat sound analysis is a convenient method for diagnosing heart disease. Heartbeat sound classification remains a challenging problem in heart sound division and feature extraction. A stethoscope is a medical device widely used by physicians to listen to the heartbeat. An acoustic stethoscope operates on the chest piece to the ears of the listener. The main problem is in listening to heart sounds that the low signal level and are difficult to be analyzed. Adding electronic circuitry and software to acoustic stethoscopes will strengthen the heart rate signal and can minimize error analysis of the state of the patient's heart. Machine learning is used to efficiently analyze and classify heart sounds. Convolutional Neural Network (CNN) models and Support Vector Machine (SVM) with feature extractors were effective methods and were used in this research. First, the Phonocardiogram (PCG) files are fragmented into pieces of equivalent length. Then, we convert the PCG files to a spectrogram. The spectrogram images are fed into a convolutional neural network and support vector machine. The best result is using an Inception V3 model with the CNN classifier which has an accuracy of 0.909, with 0.948 sensitivity and 0.869 specificity.
Some of the metrics are blocked by your
consent settings
Enhancement of Anime Imaging Enlargement using Modified Super-Resolution CNN
(2021-01-01)
Intaniyom, Tanakit
;
Thananporn, Warinthorn
;
Woraratpanya, Kuntpong
Anime is a storytelling medium similar to movies and books. Anime images are a kind of artworks, which are almost entirely drawn by hand. Hence, reproducing existing Anime with larger sizes and higher quality images is expensive. Therefore, we proposed a model based on convolutional neural networks to extract outstanding features of images, enlarge those images, and enhance the quality of Anime images. We trained the model with a training set of 160 images and a validation set of 20 images. We tested the trained model with a testing set of 20 images. The experimental results indicated that our model successfully enhanced the image quality with a larger image-size when compared with the common existing image enlargement and the original SRCNN method.
Some of the metrics are blocked by your
consent settings
Municipal solid waste segregation with CNN
(2019-07-01)
Srinilta, Chutimet
;
Kanharattanachai, Sivakorn
Pollution from municipal solid waste has been a problem in Thailand for a long time. People generate waste in every minute. Ineffective waste segregation does increase difficulties in solid waste management. The Pollution Control Department of Thailand provides segregation guideline for municipal solid waste. Household wastes should be separated into four types-general waste, compostable waste, recyclable waste and hazardous waste. This paper explored performance of CNN-based waste-type classifiers (VGG-16, ResNet-50, MobileNet V2 and DenseNet-121) in classifying waste types of 9,200 municipal solid waste images. Waste type can be identified directly from waste-type classifier or derived from waste-item class. Derived classifiers outperformed their corresponding direct classifiers in the experiment. The highest waste-type classification accuracy was 94.86% from the derived ResNet-50 classifier.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results