arXiv subjects

Hung Cao

Publications and source records attributed to Hung Cao.

At least 55 records · Page 3Linked to original sources

arXiv2024

Efficient and Concise Explanations for Object Detection with Gaussian-Class Activation Mapping Explainer

To address the challenges of providing quick and plausible explanations in Explainable AI (XAI) for object detection models, we introduce the Gaussian Class Activation Mapping Explainer (G-CAME). Our method efficiently generates concise saliency maps by utilizing activation maps from selected layers and applying a Gaussian kernel to emphasize critical image regions for the predicted object. Compared with other Region-based approaches, G-CAME significantly reduces explanation time to 0.5 seconds without compromising the quality. Our evaluation of G-CAME, using Faster-RCNN and YOLOX on the MS-COCO 2017 dataset, demonstrates its ability to offer highly plausible and faithful explanations, especially in reducing the bias on tiny object detection.

cs.CV↗

arXiv2024

Achieving Pareto Optimality using Efficient Parameter Reduction for DNNs in Resource-Constrained Edge Environment

This paper proposes an optimization of an existing Deep Neural Network (DNN) that improves its hardware utilization and facilitates on-device training for resource-constrained edge environments. We implement efficient parameter reduction strategies on Xception that shrink the model size without sacrificing accuracy, thus decreasing memory utilization during training. We evaluate our model in two experiments: Caltech-101 image classification and PCB defect detection and compare its performance against the original Xception and lightweight models, EfficientNetV2B1 and MobileNetV2. The results of the Caltech-101 image classification show that our model has a better test accuracy (76.21%) than Xception (75.89%), uses less memory on average (847.9MB) than Xception (874.6MB), and has faster training and inference times. The lightweight models overfit with EfficientNetV2B1 having a 30.52% test accuracy and MobileNetV2 having a 58.11% test accuracy. Both lightweight models have better memory usage than our model and Xception. On the PCB defect detection, our model has the best test accuracy (90.30%), compared to Xception (88.10%), EfficientNetV2B1 (55.25%), and MobileNetV2 (50.50%). MobileNetV2 has the least average memory usage (849.4MB), followed by our model (865.8MB), then EfficientNetV2B1 (874.8MB), and Xception has the highest (893.6MB). We further experiment with pre-trained weights and observe that memory usage decreases thereby showing the benefits of transfer learning. A Pareto analysis of the models' performance shows that our optimized model architecture satisfies accuracy and low memory utilization objectives.

cs.LG↗

arXiv2024

LangXAI: Integrating Large Vision Models for Generating Textual Explanations to Enhance Explainability in Visual Perception Tasks

LangXAI is a framework that integrates Explainable Artificial Intelligence (XAI) with advanced vision models to generate textual explanations for visual recognition tasks. Despite XAI advancements, an understanding gap persists for end-users with limited domain knowledge in artificial intelligence and computer vision. LangXAI addresses this by furnishing text-based explanations for classification, object detection, and semantic segmentation model outputs to end-users. Preliminary results demonstrate LangXAI's enhanced plausibility, with high BERTScore across tasks, fostering a more transparent and reliable AI framework on vision tasks for end-users.

cs.CV↗

arXiv2024

Enhancing the Fairness and Performance of Edge Cameras with Explainable AI

The rising use of Artificial Intelligence (AI) in human detection on Edge camera systems has led to accurate but complex models, challenging to interpret and debug. Our research presents a diagnostic method using Explainable AI (XAI) for model debugging, with expert-driven problem identification and solution creation. Validated on the Bytetrack model in a real-world office Edge network, we found the training dataset as the main bias source and suggested model augmentation as a solution. Our approach helps identify model biases, essential for achieving fair and trustworthy models.

cs.CV↗

arXiv2024

XAI-Enhanced Semantic Segmentation Models for Visual Quality Inspection

Visual quality inspection systems, crucial in sectors like manufacturing and logistics, employ computer vision and machine learning for precise, rapid defect detection. However, their unexplained nature can hinder trust, error identification, and system improvement. This paper presents a framework to bolster visual quality inspection by using CAM-based explanations to refine semantic segmentation models. Our approach consists of 1) Model Training, 2) XAI-based Model Explanation, 3) XAI Evaluation, and 4) Annotation Augmentation for Model Enhancement, informed by explanations and expert insights. Evaluations show XAI-enhanced models surpass original DeepLabv3-ResNet101 models, especially in intricate object segmentation.

cs.CV↗

arXiv2023

Developing a Resource-Constraint EdgeAI model for Surface Defect Detection

Resource constraints have restricted several EdgeAI applications to machine learning inference approaches, where models are trained on the cloud and deployed to the edge device. This poses challenges such as bandwidth, latency, and privacy associated with storing data off-site for model building. Training on the edge device can overcome these challenges by eliminating the need to transfer data to another device for storage and model development. On-device training also provides robustness to data variations as models can be retrained on newly acquired data to improve performance. We, therefore, propose a lightweight EdgeAI architecture modified from Xception, for on-device training in a resource-constraint edge environment. We evaluate our model on a PCB defect detection task and compare its performance against existing lightweight models - MobileNetV2, EfficientNetV2B0, and MobileViT-XXS. The results of our experiment show that our model has a remarkable performance with a test accuracy of 73.45% without pre-training. This is comparable to the test accuracy of non-pre-trained MobileViT-XXS (75.40%) and much better than other non-pre-trained models (MobileNetV2 - 50.05%, EfficientNetV2B0 - 54.30%). The test accuracy of our model without pre-training is comparable to pre-trained MobileNetV2 model - 75.45% and better than pre-trained EfficientNetV2B0 model - 58.10%. In terms of memory efficiency, our model performs better than EfficientNetV2B0 and MobileViT-XXS. We find that the resource efficiency of machine learning models does not solely depend on the number of parameters but also depends on architectural considerations. Our method can be applied to other resource-constraint applications while maintaining significant performance.

cs.CV↗

arXiv2023

Evaluating Multi-Global Server Architecture for Federated Learning

Federated learning (FL) with a single global server framework is currently a popular approach for training machine learning models on decentralized environment, such as mobile devices and edge devices. However, the centralized server architecture poses a risk as any challenge on the central/global server would result in the failure of the entire system. To minimize this risk, we propose a novel federated learning framework that leverages the deployment of multiple global servers. We posit that implementing multiple global servers in federated learning can enhance efficiency by capitalizing on local collaborations and aggregating knowledge, and the error tolerance in regard to communication failure in the single server framework would be handled. We therefore propose a novel framework that leverages the deployment of multiple global servers. We conducted a series of experiments using a dataset containing the event history of electric vehicle (EV) charging at numerous stations. We deployed a federated learning setup with multiple global servers and client servers, where each client-server strategically represented a different region and a global server was responsible for aggregating local updates from those devices. Our preliminary results of the global models demonstrate that the difference in performance attributed to multiple servers is less than 1%. While the hypothesis of enhanced model efficiency was not as expected, the rule for handling communication challenges added to the algorithm could resolve the error tolerance issue. Future research can focus on identifying specific uses for the deployment of multiple global servers.

cs.LG↗

arXiv2023

ECAvg: An Edge-Cloud Collaborative Learning Approach using Averaged Weights

The use of edge devices together with cloud provides a collaborative relationship between both classes of devices where one complements the shortcomings of the other. Resource-constraint edge devices can benefit from the abundant computing power provided by servers by offloading computationally intensive tasks to the server. Meanwhile, edge devices can leverage their close proximity to the data source to perform less computationally intensive tasks on the data. In this paper, we propose a collaborative edge-cloud paradigm called ECAvg in which edge devices pre-train local models on their respective datasets and transfer the models to the server for fine-tuning. The server averages the pre-trained weights into a global model, which is fine-tuned on the combined data from the various edge devices. The local (edge) models are then updated with the weights of the global (server) model. We implement a CIFAR-10 classification task using MobileNetV2, a CIFAR-100 classification task using ResNet50, and an MNIST classification using a neural network with a single hidden layer. We observed performance improvement in the CIFAR-10 and CIFAR-100 classification tasks using our approach, where performance improved on the server model with averaged weights and the edge models had a better performance after model update. On the MNIST classification, averaging weights resulted in a drop in performance on both the server and edge models due to negative transfer learning. From the experiment results, we conclude that our approach is successful when implemented on deep neural networks such as MobileNetV2 and ResNet50 instead of simple neural networks.

cs.LG↗

arXiv2023

Reducing Intraspecies and Interspecies Covariate Shift in Traumatic Brain Injury EEG of Humans and Mice Using Transfer Euclidean Alignment

While analytics of sleep electroencephalography (EEG) holds certain advantages over other methods in clinical applications, high variability across subjects poses a significant challenge when it comes to deploying machine learning models for classification tasks in the real world. In such instances, machine learning models that exhibit exceptional performance on a specific dataset may not necessarily demonstrate similar proficiency when applied to a distinct dataset for the same task. The scarcity of high-quality biomedical data further compounds this challenge, making it difficult to evaluate the model's generality comprehensively. In this paper, we introduce Transfer Euclidean Alignment - a transfer learning technique to tackle the problem of the dearth of human biomedical data for training deep learning models. We tested the robustness of this transfer learning technique on various rule-based classical machine learning models as well as the EEGNet-based deep learning model by evaluating on different datasets, including human and mouse data in a binary classification task of detecting individuals with versus without traumatic brain injury (TBI). By demonstrating notable improvements with an average increase of 14.42% for intraspecies datasets and 5.53% for interspecies datasets, our findings underscore the importance of the use of transfer learning to improve the performance of machine learning and deep learning models when using diverse datasets for training.

cs.LG↗

arXiv2023

Fostering new Vertical and Horizontal IoT Applications with Intelligence Everywhere

Intelligence Everywhere is predicated on the seamless integration of IoT networks transporting a vast amount of data streams through many computing resources across an edge-to-cloud continuum, relying on the orchestration of distributed machine learning models. The result is an interconnected and collective intelligent ecosystem where devices, systems, services, and users work together to support IoT applications. This paper discusses the state-of-the-art research and the principles of the Intelligence Everywhere framework for enhancing IoT applications in vertical sectors such as Digital Health, Infrastructure, and Transportation/Mobility in the context of intelligent society (Society 5.0). It also introduces a novel perspective for the development of horizontal IoT applications, capable of running across various IoT networks while fostering collective intelligence across diverse sectors. Finally, this paper provides comprehensive insights into the challenges and opportunities for harnessing collective knowledge from real-time insights, leading to optimised processes and better overall collaboration across different IoT sectors.

cs.NI↗

arXiv2023

TransferD2: Automated Defect Detection Approach in Smart Manufacturing using Transfer Learning Techniques

Quality assurance is crucial in the smart manufacturing industry as it identifies the presence of defects in finished products before they are shipped out. Modern machine learning techniques can be leveraged to provide rapid and accurate detection of these imperfections. We, therefore, propose a transfer learning approach, namely TransferD2, to correctly identify defects on a dataset of source objects and extend its application to new unseen target objects. We present a data enhancement technique to generate a large dataset from the small source dataset for building a classifier. We then integrate three different pre-trained models (Xception, ResNet101V2, and InceptionResNetV2) into the classifier network and compare their performance on source and target data. We use the classifier to detect the presence of imperfections on the unseen target data using pseudo-bounding boxes. Our results show that ResNet101V2 performs best on the source data with an accuracy of 95.72%. Xception performs best on the target data with an accuracy of 91.00% and also provides a more accurate prediction of the defects on the target images. Throughout the experiment, the results also indicate that the choice of a pre-trained model is not dependent on the depth of the network. Our proposed approach can be applied in defect detection applications where insufficient data is available for training a model and can be extended to identify imperfections in new unseen data.

cs.CV↗

arXiv2022

A Novel ECG Denoising Scheme Using the Ensemble Kalman Filter

Monitoring of electrocardiogram (ECG) provides vital information as well as any cardiovascular anomalies. Recent advances in the technology of wearable electronics have enabled compact devices to acquire personal physiological signals in the home setting; however, signals are usually contaminated with high level noise. Thus, an efficient ECG filtering scheme is a dire need. In this paper, a novel method using Ensemble Kalman Filter (EnKF) is developed for denoising ECG signals. We also intensively explore various filtering algorithms, including Savitzky-Golay (SG) filter, Ensemble Empirical mode decomposition (EEMD), Normalized Least-Mean-Square (NLMS), Recursive least squares (RLS) filter, Total variation denoising (TVD), Wavelet and extended Kalman filter (EKF) for comparison. Data from the MIT-BIH Noise Stress Test database were used. The proposed methodology shows the average signal to noise ratio (SNR) of 10.96, the Percentage Root Difference of 150.45, and the correlation coefficient of 0.959 from the modified MIT-BIH database with added motion artifacts.

eess.SP↗

arXiv2021

Deep learning-based framework for cardiac function assessment in embryonic zebrafish from heart beating videos

Zebrafish is a powerful and widely-used model system for a host of biological investigations including cardiovascular studies and genetic screening. Zebrafish are readily assessable during developmental stages; however, the current methods for quantification and monitoring of cardiac functions mostly involve tedious manual work and inconsistent estimations. In this paper, we developed and validated a Zebrafish Automatic Cardiovascular Assessment Framework (ZACAF) based on a U-net deep learning model for automated assessment of cardiovascular indices, such as ejection fraction (EF) and fractional shortening (FS) from microscopic videos of wildtype and cardiomyopathy mutant zebrafish embryos. Our approach yielded favorable performance with accuracy above 90% compared with manual processing. We used only black and white regular microscopic recordings with frame rates of 5-20 frames per second (fps); thus, the framework could be widely applicable with any laboratory resources and infrastructure. Most importantly, the automatic feature holds promise to enable efficient, consistent and reliable processing and analysis capacity for large amounts of videos, which can be generated by diverse collaborating teams.

eess.IV↗

arXiv2021

A Raspberry Pi-based Traumatic Brain Injury Detection System for Single-Channel Electroencephalogram

Traumatic Brain Injury (TBI) is a common cause of death and disability. However, existing tools for TBI diagnosis are either subjective or require extensive clinical setup and expertise. The increasing affordability and reduction in size of relatively high-performance computing systems combined with promising results from TBI related machine learning research make it possible to create compact and portable systems for early detection of TBI. This work describes a Raspberry Pi based portable, real-time data acquisition, and automated processing system that uses machine learning to efficiently identify TBI and automatically score sleep stages from a single-channel Electroen-cephalogram (EEG) signal. We discuss the design, implementation, and verification of the system that can digitize EEG signal using an Analog to Digital Converter (ADC) and perform real-time signal classification to detect the presence of mild TBI (mTBI). We utilize Convolutional Neural Networks (CNN) and XGBoost based predictive models to evaluate the performance and demonstrate the versatility of the system to operate with multiple types of predictive models. We achieve a peak classification accuracy of more than 90% with a classification time of less than 1 s across 16 s - 64 s epochs for TBI vs control conditions. This work can enable development of systems suitable for field use without requiring specialized medical equipment for early TBI detection applications and TBI research. Further, this work opens avenues to implement connected, real-time TBI related health and wellness monitoring systems.

eess.SP↗

arXiv2018

Developing an edge computing platform for real-time descriptive analytics

The Internet of Mobile Things encompasses stream data being generated by sensors, network communications that pull and push these data streams, as well as running processing and analytics that can effectively leverage actionable information for transportation planning, management, and business advantage. Edge computing emerges as a new paradigm that decentralizes the communication, computation, control and storage resources from the cloud to the edge of the network. This paper proposes an edge computing platform where mobile edge nodes are physical devices deployed on a transit bus where descriptive analytics is used to uncover meaningful patterns from real-time transit data streams. An application experiment is used to evaluate the advantages and disadvantages of our proposed platform to support descriptive analytics at a mobile edge node and generate actionable information to transit managers.

cs.CY↗

arXiv2018

An edge-fog-cloud platform for anticipatory learning process designed for Internet of Mobile Things

This paper presents a novel architecture for data analytics targeting an anticipatory learning process in the context of the Internet of Mobile Things. The architecture is geo-distributed and composed by edge, fog, and cloud resources that operate collectively to support such an anticipatory learning process. We designed the architecture to manage large volumes of data streams coming from the IoMT devices, analyze in successive phases climbing up in the hierarchy of resources from edge, fog and cloud. We discuss the characteristics of the analytical tasks at each layer. We notice that the amount of data being transported in the network decreases going from the edge, to the fog and finally to the cloud, while the complexity of the computation increases. Such design allows to support different kind of analytical needs, from real-time to historical according to the type of resource being utilized. We have implemented the proposed architecture as a proof-of-concept using the transit data feeds from the area of Greater Moncton, Canada.

cs.CY↗

arXiv2018

The design of a streaming analytical workflow for processing massive transit feeds

Retrieving and analyzing transit feeds relies on working with analytical workflows that can handle the massive volume of data streams that are relevant to understand the dynamics of transit networks which are entirely deterministic in the geographical space in which they takes place. In this paper, we consider the fundamental issues in developing a streaming analytical workflow for analyzing the continuous arrival of multiple, unbounded transit data feeds for automatically processing and enriching them with additional information containing higher level concepts accordingly to a particular mobility context. This workflow consists of three tasks: (1) stream data retrieval for creating time windows; (2) data cleaning for handling missing data, overlap data or redundant data; and (3) data contextualization for computing actual arrival and departure times as well as the stops and moves during a bus trip, and also performing mobility context computation. The workflow was implemented in a Hadoop cloud ecosystem using data streams from the CODIAC Transit System of the city of Moncton, NB. The Map() function of MapReduce is used to retrieve and bundle data streams into numerous clusters which are subsequently handled in a parallel manner by the Reduce() function in order to execute the data contextualization step. The results validate the need for cloud computing for achieving high performance and scalability, however, due to the delay in computing and networking, it is clear that data cleaning tasks should not only be deployed using a cloud environment, paving the way to combine it with fog computing in the near future.

cs.CY↗

arXiv2017

Implementing an Edge-Fog-Cloud architecture for stream data management

The Internet of Moving Things (IoMT) requires support for a data life cycle process ranging from sorting, cleaning and monitoring data streams to more complex tasks such as querying, aggregation, and analytics. Current solutions for stream data management in IoMT have been focused on partial aspects of a data life cycle process, with special emphasis on sensor networks. This paper aims to address this problem by developing streaming data life cycle process that incorporates an edge/fog/cloud architecture that is needed for handling heterogeneous, streaming and geographically-dispersed IoMT devices. We propose a 3-tier architecture to support an instant intra-layer communication that establishes a stream data flow in real-time to respond to immediate data life cycle tasks in the system. Communication and process are thus the defining factors in the design of our stream data management solution for IoMT. We describe and evaluate our prototype implementation using real-time transit data feeds. Preliminary results are showing the advantages of running data life cycle tasks for reducing the volume of data streams that are redundant and should not be transported to the cloud.

cs.DC↗