Ba-Viet Ngo

Mr. Ba-Viet Ngo

Ho Chi Minh City University of Technology and Education, Ho Chi Minh City, Hồ Chí Minh, Vietnam
orcid

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

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

Representative Publications

View All Publications
5 October 2022 in Agriengineering
Verified
Journal Article
Open access
PDF file
Thanh-Hai NguyenThanh-Hai Nguyen
Thanh-Hai Nguyen
Nghia Thanh NguyenNghia Thanh Nguyen
Nghia Thanh Nguyen
Ba-Viet NgoBa-Viet Ngo
Ba-Viet Ngo
Tomato leaves can have different diseases which can affect harvest performance. Therefore, accurate classification for the early detection of disease for treatment is very important. This article proposes one classification model, in which 16,010 tomato leaf images obtained from the Plant Village database are segmented before being used to train a deep convolutional neural network (DCNN). This means that this classification model will reduce training time compared with that of the model without segmenting the images. In particular, we applied a VGG-19 model with transfer learning for re-training in later layers. In addition, the parameters such as epoch and learning rate were chosen to be suitable for increasing classification performance. One highlight point is that the leaf images were segmented for extracting the original regions and removing the backgrounds to be black using a hue, saturation, and value (HSV) color space. The segmentation of the leaf images is to synchronize the black background of all leaf images. It is obvious that this segmentation saves time for training the DCNN and also increases the classification performance. This approach improves the model accuracy to 99.72% and decreases the training time of the 16,010 tomato leaf images. The results illustrate that the model is effective and can be developed for more complex image datasets.

13 February 2025 in Electronics
Verified
Journal Article
Open access
PDF file
Thanh-Hai NguyenThanh-Hai Nguyen
Thanh-Hai Nguyen
Ba-Viet NgoBa-Viet Ngo
Ba-Viet Ngo
Nghia Thanh NguyenNghia Thanh Nguyen
Nghia Thanh Nguyen
Electric wheelchairs are the primary means of transportation that enable individuals with disabilities to move independently to their desired locations. This paper introduces a novel, low-cost smart wheelchair system designed to enhance the mobility of individuals with severe disabilities through hand gesture recognition. Additionally, the system aims to support low-income individuals who previously lacked access to smart wheelchairs. Unlike existing methods that rely on expensive hardware or complex systems, the proposed system utilizes an affordable webcam and an Nvidia Jetson Nano embedded computer to process and recognize six distinct hand gestures—“Forward 1”, “Forward 2”, “Backward”, “Left”, “Right”, and “Stop”—to assist with wheelchair navigation. The system employs the “You Only Look Once version 8n” (YOLOv8n) model, which is well suited for low-spec embedded computers, trained on a self-collected hand gesture dataset containing 12,000 images. The pre-processing phase utilizes the MediaPipe library to generate landmark hand images, remove the background, and then extract the region of interest (ROI) of the hand gestures, significantly improving gesture recognition accuracy compared to previous methods that relied solely on hand images. Experimental results demonstrate impressive performance, achieving 99.3% gesture recognition accuracy and 93.8% overall movement accuracy in diverse indoor and outdoor environments. Furthermore, this paper presents a control circuit system that can be easily installed on any existing electric wheelchair. This approach offers a cost-effective, real-time solution that enhances the autonomy of individuals with severe disabilities in daily activities, laying the foundation for the development of affordable smart wheelchairs.

13 July 2023 in Micromachines
Verified
Journal Article
Open access
PDF file
Thanh-Hai NguyenThanh-Hai Nguyen
Thanh-Hai Nguyen
Ba-Viet NgoBa-Viet Ngo
Ba-Viet Ngo
Nghia Thanh NguyenNghia Thanh Nguyen
Nghia Thanh Nguyen
Chi Cuong VuChi Cuong Vu
Chi Cuong Vu
Soft sensors are attracting much attention from researchers worldwide due to their versatility in practical projects. There are already many applications of soft sensors in aspects of life, consisting of human-robot interfaces, flexible electronics, medical monitoring, and healthcare. However, most of these studies have focused on a specific area, such as fabrication, data analysis, or experimentation. This approach can lead to challenges regarding the reliability, accuracy, or connectivity of the components. Therefore, there is a pressing need to consider the sensor’s placement in an overall system and find ways to maximize the efficiency of such flexible sensors. This paper proposes a fabrication method for soft capacitive pressure sensors with spacer fabric, conductive inks, and encapsulation glue. The sensor exhibits a good sensitivity of 0.04 kPa−1, a fast recovery time of 7 milliseconds, and stability of 10,000 cycles. We also evaluate how to connect the sensor to other traditional sensors or hardware components. Some machine learning models are applied to these built-in soft sensors. As expected, the embedded wearables achieve a high accuracy of 96% when recognizing human walking phases.

4 September 2022 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Ba-Viet NgoBa-Viet Ngo
Ba-Viet Ngo
Thanh-Hai NguyenThanh-Hai Nguyen
Thanh-Hai Nguyen
A semi-automatic wheelchair allows disabled people to possibly control in an indoor environment with obstacles and targets. The paper proposes an EEG-based control system for the wheelchair based on a grid map designed to allow disabled people to reach any preset destination. In particular, the grid map is constructed by dividing it into grid cells that may contain free spaces or obstacles. The map with the grid cells is simulated to find the optimal paths to the target positions using a Deep Q-Networks (DQNs) model with the Parametric Rectified Linear Unit (PReLU) activation function, in which a novel algorithm for finding the optimal path planning by converting wheelchair actions is applied using the output parameters of the DQNs. For the wheelchair movement in one real indoor environment corresponding to the virtual 2D grid map, the initial position of the wheelchair will be determined based on natural landmarks and a graphical user interface designed for on-screen display can support disabled people in selecting the desired destination from a list of predefined locations using Electroencephalogram (EEG) signals by blinking eyes. Therefore, one user can easily and safely control the wheelchair using an EEG system to reach the desired target when the wheelchair position and destination are determined in the indoor environment. As a result, a grid map was developed and experiments for the semi-automatic wheelchair control were performed in real indoor environments to illustrate the effectiveness of the proposed method. In addition, the system is a platform to develop different types of controls depending on the types of user disabilities and different environmental maps built.

2 March 2026 in IEEE Access
Verified
Journal Article
Open access
TNThanh-Tam Nguyen
Thanh-Tam Nguyen
Thanh-Hai NguyenThanh-Hai Nguyen
Thanh-Hai Nguyen
Nghia Thanh NguyenNghia Thanh Nguyen
Nghia Thanh Nguyen
Ba-Viet NgoBa-Viet Ngo
Ba-Viet Ngo
Breast cancer diagnosis using mammography can be significantly enhanced by integrating information from multi-source fusion and complementary imaging perspectives. In this paper, we propose a novel deep learning framework with dataset fusion for breast lesion classification. In particular, our approach fuses mammogram datasets of both Medio-Lateral-Oblique (MLO) and Cranio-Caudal (CC) views in four breast lesion categories from two data sources to ensure comprehensive lesion representation with total 5,906 images, and a combined EfficientNet-B7 and Multi-Head Attention (MHA) network is applied to classify these breast lesion categories. Therefore, the largest Region of Interest (ROI) in each image is extracted and then colorized before inserted into the network for lesion classification. Two colorized MLO and CC ROIs are separately processed by two parallel EfficientNet-B7 networks to produce the combined high-level features. The combined feature vectors from the dual EfficientNet-B7 network are then concatenated to yield a unified feature representation, which is subsequently refined using the MHA network to prioritize the most salient patterns and suppress background noise before classified using a fully connected head. To select an optimal classification model for these breast lesion categories, we performed four different models for two cases with and without an MHA network. Therefore, extensive experiments on integrated multi-source mammogram datasets demonstrate that the proposed framework outperforms conventional single-view and non-attentive models, achieving around 93.9% accuracy for breast lesion classification. It is obvious that the use of MHA improves accuracy by about 2% compared to not using it. The results highlight the value of dataset fusion and MHA-based feature selection in advancing computer-aided mammography.

4 June 2025 in Engineering, Technology & Applied Science Research
Verified
Journal Article
Open access
PDF file
Nghia Thanh NguyenNghia Thanh Nguyen
Nghia Thanh Nguyen
Ba-Viet NgoBa-Viet Ngo
Ba-Viet Ngo
The number of breast cancer patients is rapidly increasing worldwide, with Asia currently accounting for 45% of global breast cancer cases. In addition, the number of breast cancer cases is expected to increase by 21.0%, and the mortality rate is projected to increase by 27.8% during the 2020-2030 period. This paper proposes a method for classifying breast cancer from Digital Imaging and Communications in Medicine (DICOM) images. In particular, an image segmentation technique was developed for extracting the breast region from DICOM images of varying sizes. The extracted images were then enhanced using multiple augmentation techniques to improve classification performance. Finally, a deep learning network was applied to classify breast cancer from the processed DICOM images. The VinDr-Mammo dataset was used to evaluate the effectiveness of the proposed method, and the experimental results showed an accuracy of 81.45%, demonstrating that the proposed approach is highly suitable for breast cancer detection and classification.