arXiv subjects

Luan Nguyen

Publications and source records attributed to Luan Nguyen.

4 recordsLinked to original sources

arXiv2025

Fast and Simple Densest Subgraph with Predictions

We study the densest subgraph problem and its NP-hard densest at-most-$k$ subgraph variant through the lens of learning-augmented algorithms. We show that, given a reasonably accurate predictor that estimates whether a node belongs to the solution (e.g., a machine learning classifier), one can design simple linear-time algorithms that achieve a $(1-\epsilon)$approximation. Finally, we present experimental results demonstrating the effectiveness of our methods for the densest at-most-$k$ subgraph problem on real-world graphs.

cs.DS↗

arXiv2019

Small Target Detection for Search and Rescue Operations using Distributed Deep Learning and Synthetic Data Generation

It is important to find the target as soon as possible for search and rescue operations. Surveillance camera systems and unmanned aerial vehicles (UAVs) are used to support search and rescue. Automatic object detection is important because a person cannot monitor multiple surveillance screens simultaneously for 24 hours. Also, the object is often too small to be recognized by the human eye on the surveillance screen. This study used UAVs around the Port of Houston and fixed surveillance cameras to build an automatic target detection system that supports the US Coast Guard (USCG) to help find targets (e.g., person overboard). We combined image segmentation, enhancement, and convolution neural networks to reduce detection time to detect small targets. We compared the performance between the auto-detection system and the human eye. Our system detected the target within 8 seconds, but the human eye detected the target within 25 seconds. Our systems also used synthetic data generation and data augmentation techniques to improve target detection accuracy. This solution may help the search and rescue operations of the first responders in a timely manner.

cs.CV↗

arXiv2019

Improved visible to IR image transformation using synthetic data augmentation with cycle-consistent adversarial networks

Infrared (IR) images are essential to improve the visibility of dark or camouflaged objects. Object recognition and segmentation based on a neural network using IR images provide more accuracy and insight than color visible images. But the bottleneck is the amount of relevant IR images for training. It is difficult to collect real-world IR images for special purposes, including space exploration, military and fire-fighting applications. To solve this problem, we created color visible and IR images using a Unity-based 3D game editor. These synthetically generated color visible and IR images were used to train cycle consistent adversarial networks (CycleGAN) to convert visible images to IR images. CycleGAN has the advantage that it does not require precisely matching visible and IR pairs for transformation training. In this study, we discovered that additional synthetic data can help improve CycleGAN performance. Neural network training using real data (N = 20) performed more accurate transformations than training using real (N = 10) and synthetic (N = 10) data combinations. The result indicates that the synthetic data cannot exceed the quality of the real data. Neural network training using real (N = 10) and synthetic (N = 100) data combinations showed almost the same performance as training using real data (N = 20). At least 10 times more synthetic data than real data is required to achieve the same performance. In summary, CycleGAN is used with synthetic data to improve the IR image conversion performance of visible images.

eess.IV↗

arXiv2011

Fabrication of Rare Earth-Doped Transparent Glass Ceramic Optical Fibers by Modified Chemical Vapor Deposition

Rare earth (RE) doped silica-based optical fibers with transparent glass ceramic (TGC) core was fabricated through the well-known modified chemical vapor deposition (MCVD) process without going through the commonly used stage of post-ceramming. The main characteristics of the RE-doped oxyde nanoparticles namely, their density and mean diameter in the fibers are dictated by the concentration of alkaline earth element used as phase separating agent. Magnesium and erbium co-doped fibers were fabricated. Optical transmission in term of loss due to scattering as well as some spectroscopic characteristics of the erbium ions was studied. For low Mg content, nano-scale particles could be grown with and relatively low scattering losses were obtained, whereas large Mg-content causes the growth of larger particles resulting in much higher loss. However in the latter case, certain interesting alteration of the spectroscopic properties of the erbium ions were observed. These initial studies should be useful in incorporating new doped materials in order to realize active optical fibers for constructing lasers and amplifiers.

physics.optics↗