Long Tonthat

Long Tonthat

Auto-generated profile, based on publicly available information

Long Tonthat

School of Electrical Engineering, International University, Ho Chi Minh City, Vietnam

Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 December 2022 in IEEE Access
Verified
Journal Article
Open access
THTri Huynh
Tri Huynh
TLTuan Minh Le
Tuan Minh Le
LTLong Ton That
Long Ton That
LTLy Van Tran
Ly Van Tran
SDSon Dao
Son Dao
Fruit and vegetable identification and classification system is always necessary and advantageous for the agriculture business, the food processing sector, as well as the convenience shops and hypermarkets where these products are sold. Therefore, it is necessary to build an effective automated tool to meet the needs of the market by boosting the outcome, in order to improve economic efficiency. In this paper, a two-stage model is proposed to recognize fruits using camera images. We employed a Densnet121 to get the features from the fruits dataset in the first module. In the second stage, we utilize a feature subset selection method to choose the most significant features for recognizing fruits from the images of the fruits. In this study, Adaptive particle - Grey Wolf Optimization (APGWO) has been applied for choosing the most pertinent features. The final subset feature has been used for recognizing fruits using several machine learning classifiers, namely K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), and Multilayer Perceptron (MLP). The proposed research’s experimental results are highly effective; the training time of proposed models is reduced to over 50%, and the classification accuracy reaches 99%.

4 August 2020 in IEEE Transactions on Magnetics
Verified
Journal Article
Closed access
Gia Minh Thao NGUYENGia Minh Thao NGUYEN
Gia Minh Thao NGUYEN
KFKeisuke Fujisaki
Keisuke Fujisaki
LTLong Ton That
Long Ton That
SMSatoshi Motozuka
Satoshi Motozuka
This paper is the first work to carefully investigate the characteristics of iron loss per weight and relative permeability between the flake and spherical powders with the sinusoidal excitation up to a high frequency of 3.5 MHz in experiments using three separate measurement systems. The composition of the considered powder samples contains 0.001 wt% C, 0.012 wt% Si, 0.09 wt% Mn, 0.0004 wt% P, 0.008 wt% S, and 0.06 wt% O. The flake particle has the preferred orientation that the easy axes are normal to the platelet face, and the water atomized iron powder was processed with ball milling with the lubrication. Key objectives of the utilization of the three independent measurement systems with different specialized equipment and probes are to cross-check and overcome unexpected measurement errors of the related devices in use for validating measured results thoroughly. Experimental results demonstrate the much better performances on the iron loss per weight and relative permeability of the flake powder than that of the spherical powder at high frequencies. Physics-based explanations for the measured results are also presented. Furthermore, two estimation functions based on a six-parameter Steinmetz equation for the measured iron loss of the powders are developed by using MATLAB.

1 March 2012 in Journal of Biological Systems
Verified
Journal Article
Closed access
LTLong Tonthat
Long Tonthat
Zhengtao DingZhengtao Ding
Zhengtao Ding

27 March 2025 in IEEE Access
Verified
Journal Article
Open access
UNUyen L. P. Nguyen
Uyen L. P. Nguyen
LTLong Tonthat
Long Tonthat
VPVo-Tan Phuoc
Vo-Tan Phuoc
LNLap Luat Nguyen
Lap Luat Nguyen
This paper proposes an image encryption-decryption scheme with two-layer security. In the first layer, an image is encrypted by a switched system, using the concept from control theory. Also, from this switched system, a random key matrix will be generated while encrypting in the first layer. This key matrix is used to diffuse the image pixels from the first layer, providing the second layer of security. Then, in the decryption stage, the random key matrix plays a role as a cryptographic key and the optimization method will be used to decrypt the second layer encrypted image. The flatness property of the switched system that makes the plain image to be recovered from the encrypted image of the first layer. This scheme is to enhance the security in image encryption and also fortifies the encryption process against sophisticated attacks while maintaining computational efficiency. Finally, the effectiveness of the proposed scheme is evaluated by simulation, then assessing it to different types of cryptanalytic analyses such as information entropy analysis, adjacent pixel correlation, differential attack analysis, key sensitivity analysis, and histogram analysis. The numerical values that can achieved by the proposed schemes are entropy 7.996, adjacent pixel correlation for horizontal −0.0012, vertical −0.0039, diagonal 0.0005; and differential attack analysis with Number of Pixels Change Rate 0.9963 and Unified Average Changing Intensity 0.3341.

1 December 2022 in IAES International Journal of Artificial Intelligence (IJ-AI)
Verified
Journal Article
Open access
PDF file
KTKien Trang
Kien Trang
ANAn Hoang Nguyen
An Hoang Nguyen
LTLong TonThat
Long TonThat
Bao Quoc VuongBao Quoc Vuong
Bao Quoc Vuong
Millions of fatal cases have been reported worldwide as a result of the Coronavirus disease 2019 (COVID-19) outbreak. In order to stop the spreading of disease, early diagnosis and quarantine of infected people are one of the most essential steps. Therefore, due to the similar symptoms of SARS-CoV-2 virus and other pneumonia, identifying COVID-19 still exists some challenges. Reverse transcription-polymerase chain reaction (RT-PCR) is known as a standard method for the COVID-19 diagnosis process. Due to the shortage of RT-PCR toolkit in global, Chest X-Ray (CXR) image is introduced as an initial step to support patient’s classification. Applying deep learning in medical imaging becomes an advanced research trend in many applications. In this research, RepVGG pre-trained model is considered to be used as the main backbone of the network. Besides, variational autoencoder (VAE) is firstly trained to perform lung segmentation. Afterwards, the encoder part in VAE is preserved as an additional feature extractor to combine with RepVGG performing classification. A COVID-19 radiography database consisting of 3 classes COVID-19, Normal and Viral Pneumonia is conducted. The obtained average accuracy of the proposed model is 95.4% and other evaluation metrics also show better results compared with the original RepVGG model.

16 July 2025 in IEEE Access
Verified
Journal Article
Open access
UNUyen L. P. Nguyen
Uyen L. P. Nguyen
LTLong Tonthat
Long Tonthat
VTVu Dai Tran
Vu Dai Tran
LNLap Luat Nguyen
Lap Luat Nguyen
This paper proposes a pre-processing encryption then hiding scheme that enhances image security in steganography. The method combines a Walsh code with switched linear system encryption, followed by Discrete Wavelet Transform and Singular Value Decomposition (DWT-SVD) for secure image hiding. The Walsh code is considered as the switching signal of the switched linear system, which offers benefits in image encryption due to its characteristics of left invertibility and flatness. Additionally, with the aid of the Walsh code generator, the key matrix will be formed to enhance the image encryption security. To hide the encrypted secret image in another image, the DWT-SVD method will be applied to obtain the frequency-domain representation, which is then transformed into the singular values of the secret image. Consequently, the matrix manipulations between SVD matrices of both the encrypted secret image and cover image complete the embedding process of steganography. The contribution of this scheme lies in its simplicity of hardware use while achieving high security. Finally, the simulations are shown to demonstrate the effectiveness of this scheme. The metrics such as Mean Squared Error (MSE), Peak Signal-to-Noise Ratio (PSNR), Bit Correction Ratio (BCR), and Structural Similarity Index Measurement (SSIM) are examined to evaluate the security level of the proposed scheme.