Long Nguyen The

Long Nguyen The


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
11 November 2019 in Algorithms
Verified
Journal Article
Open access
PDF file
HTHuong Nguyen Thu
Huong Nguyen Thu
LTLong Nguyen The
Long Nguyen The
The system that automatically identifies the anthropometric fingerprint is one of the systems that interact directly with the user, which every day will be provided with a diverse database. This requires the system to be optimized to handle the process to meet the needs of users such as fast processing time, almost absolute accuracy, no errors in the real process. Therefore, in this paper, we propose the application of machine learning methods to develop fingerprint classification algorithms based on the singularity feature. The goal of the paper is to reduce the number of comparisons in automatic fingerprint recognition systems with large databases. The combination of using computer vision algorithms in the image pre-processing stage increases the calculation time, improves the quality of the input images, making the process of feature extraction highly effective and the classification process fast and accurate. The classification results on 3 datasets with the criteria for Precision, Recall, Accuracy evaluation and ROC analysis of algorithms show that the Random Forest (RF) algorithm has the best accuracy (≥96.75%) on all 3 databases, Support Vector Machine (SVM) has the best results (≥95.5%) 2 / 3 databases.

3 January 2021 in Journal of the Operations Research Society of China
Verified
Journal Article
Closed access
PDF file
LTLong Nguyen The
Long Nguyen The
HTHuong Nguyen Thu
Huong Nguyen Thu
AAAlexander Diomidovich Afanasiev
Alexander Diomidovich Afanasiev
TNTao Van Nguyen
Tao Van Nguyen

28 April 2019 in BRAIN. BROAD RESEARCH IN ARTIFICIAL INTELLIGENCE AND NEUROSCIENCE
Verified
Journal Article
Open access
PDF file
HTHuong Nguyen Thu
Huong Nguyen Thu
LTLong Nguyen The
Long Nguyen The
ROC analysis is a visual and numerical method used to evaluate the performance of classification algorithms, such as those used to predict the structure and functions from string data. The main objective of the paper is to use the ROC analysis to evaluate the accuracy of the Random Forest algorithm to classify road surface defects on three different sets of data collected from Portugal, Irkutsk city - Russia Federation and Thai Nguyen city - Vietnam. This article summarizes the basics of ROC analysis and interprets the results analysis with other thresholds to build ROC curve. In addition, we present the steps to build a system to automatically classify road surface defects based on visual techniques and machine learning methods.

1 May 2020 in Intelligent Decision Technologies
Verified
Journal Article
Closed access
PDF file
TNT.H. Nguyen
T.H. Nguyen
LTLong Nguyen The
Long Nguyen The
DSD.N. Sidorov
D.N. Sidorov
ADA.I. Dreglea
A.I. Dreglea
Bubble detection is a challenging problem in automatic process control in the power and energy industry, medical and pharmaceutical industry and many other fields. Computer vision methods applications for bubble detection and measurement is the principal step of robust bubbles monitoring systems development. In various applications the input image may include a diverse and image background, especially in different environments. This paper presents a new and effective bubble detection approach. The main steps of this proposed approach are as follows: image preprocessing, background subtraction, and contour detection. The graph cut algorithm is used for image segmentation. The Haar wavelet transform is applied to collect bubble component points. The developed approach is evaluated based on the real data set.

30 June 2026 in TNU Journal of Science and Technology
Verified
Journal Article
Closed access
PDF file
PLPhạm Thị Liên
Phạm Thị Liên
HTHuong Nguyen Thu
Huong Nguyen Thu
LTLong Nguyen The
Long Nguyen The
TVTrần Tuấn Việt
Trần Tuấn Việt
Cháy rừng đang là vấn đề nghiêm trọng tại Việt Nam, nhất là trong mùa khô. Nghiên cứu này đề xuất sử dụng mạng nơ-ron ConvLSTM để phát hiện sớm cháy rừng từ dữ liệu ảnh. Khác với các phương pháp truyền thống chỉ xử lý từng khung hình riêng lẻ, ConvLSTM có khả năng kết hợp trích xuất đặc điểm không gian (qua lớp tích chập) và mô hình hóa sự phụ thuộc thời gian (qua bộ nhớ dài-ngắn LSTM). Chúng tôi sử dụng bộ dữ liệu công khai gồm 999 ảnh có và không có cháy. Để áp dụng ConvLSTM trên ảnh tĩnh, các chuỗi thời gian được tạo bằng cửa sổ trượt trên các biến thể tăng cường dữ liệu. Kết quả thử nghiệm cho thấy mô hình đạt độ chính xác 98,3%, độ chính xác dự báo dương tính 98,1%, độ nhạy 96,8% và F1-score 98,1%, cao hơn rõ rệt so với các mô hình CNN hay LSTM độc lập. Dù vậy, nghiên cứu còn hạn chế về kích thước bộ dữ liệu (nhỏ, mất cân bằng, 75% là ảnh có cháy) và thiếu dữ liệu đặc thù cho Việt Nam. Các bước phát triển tiếp theo cần hướng tới bộ dữ liệu lớn hơn, phù hợp với điều kiện địa phương và triển khai thời gian thực trên các thiết bị biên.

13 December 2023
Verified
Book Chapter
Closed access
PDF file
LTLong Nguyen The
Long Nguyen The
HTHuong Nguyen Thu
Huong Nguyen Thu
SGShmeleva Anna. G.
Shmeleva Anna. G.
PLPham Thi Lien
Pham Thi Lien