Engineering Commons^™

A Survey On Few-Shot Class-Incremental Learning, Songsong Tian, Lusi Li, Weijun Li, Hang Ran, Xin Ning, Prayag Tiwari

Computer Science Faculty Publications

Large deep learning models are impressive, but they struggle when real-time data is not available. Few-shot class-incremental learning (FSCIL) poses a significant challenge for deep neural networks to learn new tasks from just a few labeled samples without forgetting the previously learned ones. This setup can easily leads to catastrophic forgetting and overfitting problems, severely affecting model performance. Studying FSCIL helps overcome deep learning model limitations on data volume and acquisition time, while improving practicality and adaptability of machine learning models. This paper provides a comprehensive survey on FSCIL. Unlike previous surveys, we aim to synthesize few-shot learning and incremental …

Multimodal Fusion For Audio-Image And Video Action Recognition, Muhammad B. Shaikh, Douglas Chai, Syed M. S. Islam, Naveed Akhtar

Research outputs 2022 to 2026

Multimodal Human Action Recognition (MHAR) is an important research topic in computer vision and event recognition fields. In this work, we address the problem of MHAR by developing a novel audio-image and video fusion-based deep learning framework that we call Multimodal Audio-Image and Video Action Recognizer (MAiVAR). We extract temporal information using image representations of audio signals and spatial information from video modality with the help of Convolutional Neutral Networks (CNN)-based feature extractors and fuse these features to recognize respective action classes. We apply a high-level weights assignment algorithm for improving audio-visual interaction and convergence. This proposed fusion-based framework utilizes …

A Wavegan Approach For Mmwave-Based Fanet Topology Optimization, Enas Odat, Hakim Ghazzai, Ahmad Alsharoa

Electrical and Computer Engineering Faculty Research & Creative Works

The integration of dynamic Flying Ad hoc Networks (FANETs) and millimeter Wave (mmWave) technology can offer a promising solution for numerous data-intensive applications, as it enables the establishment of a robust flying infrastructure with significant data transmission capabilities. However, to enable effective mmWave communication within this dynamic network, it is essential to precisely align the steerable antennas mounted on Unmanned Aerial Vehicles (UAVs) with their corresponding peer units. Therefore, it is important to design a novel approach that can quickly determine an optimized alignment and network topology. In this paper, we propose a Generative Adversarial Network (GAN)-based approach, called WaveGAN, …

Sub-Band Backdoor Attack In Remote Sensing Imagery, Kazi Aminul Islam, Hongyi Wu, Chunsheng Xin, Rui Ning, Liuwan Zhu, Jiang Li

Electrical & Computer Engineering Faculty Publications

Remote sensing datasets usually have a wide range of spatial and spectral resolutions. They provide unique advantages in surveillance systems, and many government organizations use remote sensing multispectral imagery to monitor security-critical infrastructures or targets. Artificial Intelligence (AI) has advanced rapidly in recent years and has been widely applied to remote image analysis, achieving state-of-the-art (SOTA) performance. However, AI models are vulnerable and can be easily deceived or poisoned. A malicious user may poison an AI model by creating a stealthy backdoor. A backdoored AI model performs well on clean data but behaves abnormally when a planted trigger appears in …

Urban Flood Extent Segmentation And Evaluation From Real-World Surveillance Camera Images Using Deep Convolutional Neural Network, Yidi Wang, Yawen Shen, Behrouz Salahshour, Mecit Cetin, Khan Iftekharuddin, Navid Tahvildari, Guoping Huang, Devin K. Harris, Kwame Ampofo, Jonathan L. Goodall

Civil & Environmental Engineering Faculty Publications

This study explores the use of Deep Convolutional Neural Network (DCNN) for semantic segmentation of flood images. Imagery datasets of urban flooding were used to train two DCNN-based models, and camera images were used to test the application of the models with real-world data. Validation results show that both models extracted flood extent with a mean F1-score over 0.9. The factors that affected the performance included still water surface with specular reflection, wet road surface, and low illumination. In testing, reduced visibility during a storm and raindrops on surveillance cameras were major problems that affected the segmentation of flood extent. …

Understanding The Impact Of Trade Policy Effect Uncertainty On Firm-Level Innovation Investment: A Deep Learning Approach, Daniel Chang, Nan Hu, Peng Liang, Morgan Swink

Research Collection School Of Computing and Information Systems

Integrating the real options perspective and resource dependence theory, this study examines how firms adjust their innovation investments to trade policy effect uncertainty (TPEU), a less studied type of firm specific, perceived environmental uncertainty in which managers have difficulty predicting how potential policy changes will affect business operations. To develop a text-based, context-dependent, time-varying measure of firm-level perceived TPEU, we apply Bidirectional Encoder Representations from Transformers (BERT), a state-of-the-art deep learning approach. We apply BERT to analyze the texts of mandatory Management Discussion and Analysis (MD&A) sections of annual reports for a sample of 22,669 firm-year observations from 3,181 unique …

Sc-Fuse: A Feature Fusion Approach For Unpaved Road Detection From Remotely Sensed Images, Aniruddh Saxena

Department of Computer Science and Engineering: Dissertations, Theses, and Student Research

Road network extraction from remote sensing imagery is crucial for numerous applications, ranging from autonomous navigation to urban and rural planning. A particularly challenging aspect is the detection of unpaved roads, often underrepresented in research and data. These roads display variability in texture, width, shape, and surroundings, making their detection quite complex. This thesis addresses these challenges by creating a specialized dataset and introducing the SC-Fuse model.

Our custom dataset comprises high resolution remote sensing imagery which primarily targets unpaved roads of the American Midwest. To capture the diverse seasonal variation and their impact, the dataset includes images from different …

Statistical And Deep Learning Models For Reference Evapotranspiration Time Series Forecasting: A Comparison Of Accuracy, Complexity, And Data Efficiency, Arman Ahmadi, Andre Daccache, Mojtaba Sadegh, Richard L. Snyder

Civil Engineering Faculty Publications and Presentations

Reference evapotranspiration (ETo) is an essential variable in agricultural water resources management and irrigation scheduling. An accurate and reliable forecast of ETo facilitates effective decision-making in agriculture. Although numerous studies assessed various methodologies for ETo forecasting, an in-depth multi-dimensional analysis evaluating different aspects of these methodologies is missing. This study systematically evaluates the complexity, computational cost, data efficiency, and accuracy of ten models that have been used or could potentially be used for ETo forecasting. These models range from well-known statistical forecasting models like seasonal autoregressive integrated moving average (SARIMA) to state-of-the-art deep learning (DL) algorithms like temporal fusion transformer …

An In-Depth Analysis Of Domain Adaptation In Computer And Robotic Vision, Muhammad Hassan Tanveer, Zainab Fatima, Shehnila Zardari, David A. Guerra-Zubiaga

Faculty and Research Publications

This review article comprehensively delves into the rapidly evolving field of domain adaptation in computer and robotic vision. It offers a detailed technical analysis of the opportunities and challenges associated with this topic. Domain adaptation methods play a pivotal role in facilitating seamless knowledge transfer and enhancing the generalization capabilities of computer and robotic vision systems. Our methodology involves systematic data collection and preparation, followed by the application of diverse assessment metrics to evaluate the efficacy of domain adaptation strategies. This study assesses the effectiveness and versatility of conventional, deep learning-based, and hybrid domain adaptation techniques within the domains of …

Prediction Of Drilling Fluid Lost-Circulation Zone Based On Deep Learning, Yili Kang, Chenglin Ma, Chengyuan Xu, Lijun You, Zhenjiang You

Research outputs 2022 to 2026

Lost circulation has become a crucial technical problem that restricts the quality and efficiency improvement of the drilling operation in deep oil and gas wells. The lost-circulation zone prediction has always been a hot and difficult research topic on the prevention and control of lost circulation. This study applied machine learning and statistical methods to deeply mine 105 groups and 29 features of loss data from typical loss block M. After removing 10 sets of noise data, the methods of mean removal, range scaling and normalization were used to pre-treat the 95 sets of the loss data. The multi-factor analysis …

Learning Deep Time-Index Models For Time Series Forecasting, Jiale Gerald Woo, Chenghao Liu, Doyen Sahoo, Akshat Kumar, Steven Hoi

Research Collection School Of Computing and Information Systems

Deep learning has been actively applied to time series forecasting, leading to a deluge of new methods, belonging to the class of historicalvalue models. Yet, despite the attractive properties of time-index models, such as being able to model the continuous nature of underlying time series dynamics, little attention has been given to them. Indeed, while naive deep timeindex models are far more expressive than the manually predefined function representations of classical time-index models, they are inadequate for forecasting, being unable to generalize to unseen time steps due to the lack of inductive bias. In this paper, we propose DeepTime, a …

Distributed Deep Learning Optimization Of Heat Equation Inverse Problem Solvers, Zhuowei Wang, Le Yang, Haoran Lin, Genping Zhao, Zixuan Liu, Xiaoyu Song

Electrical and Computer Engineering Faculty Publications and Presentations

The inversion problem of partial differential equation plays a crucial role in cyber-physical systems applications. This paper presents a novel deep learning optimization approach to constructing a solver of heat equation inversion. To improve the computational efficiency in large-scale industrial applications, data and model parallelisms are incorporated on a platform of multiple GPUs. The advanced Ring-AllReduce architecture is harnessed to achieve an acceleration ratio of 3.46. Then a new multi-GPUs distributed optimization method GradReduce is proposed based on Ring-AllReduce architecture. This method optimizes the original data communication mechanism based on mechanical time and frequency by introducing the gradient transmission scheme …

Terrain Characterization Via Machine Vs. Deep Learning Using Remote Sensing, Jordan Ewing, Thomas Oommen, Jobin Thomas, Anush Kasaragod, Richard Dobson, Colin Brooks, Paramsothy Jayakumar, Michael Cole, Tulga Ersal

Michigan Tech Publications

Terrain traversability is critical for developing Go/No-Go maps for ground vehicles, which significantly impact a mission’s success. To predict the mobility of terrain, one must understand the soil characteristics. In-situ measurements performed in the field are the current method of collecting this information, which is time-consuming, costly, and can be lethal for military operations. This paper investigates an alternative approach using thermal, multispectral, and hyperspectral remote sensing from an unmanned aerial vehicle (UAV) platform. Remotely sensed data combined with machine learning (linear, ridge, lasso, partial least squares (PLS), support vector machines (SVM), and k nearest neighbors (KNN)) and deep learning …

Tempnet – Temporal Super-Resolution Of Radar Rainfall Products With Residual Cnns, Muhammed Ali Sit, Bongchul Seo, Ibrahim Demir

Civil, Architectural and Environmental Engineering Faculty Research & Creative Works

The temporal and spatial resolution of rainfall data is crucial for environmental modeling studies in which its variability in space and time is considered as a primary factor. Rainfall products from different remote sensing instruments (e.g., radar, satellite) have different space-time resolutions because of the differences in their sensing capabilities and post-processing methods. In this study, we developed a deep-learning approach that augments rainfall data with increased time resolutions to complement relatively lower-resolution products. We propose a neural network architecture based on Convolutional Neural Networks (CNNs), namely TempNet, to improve the temporal resolution of radar-based rainfall products and compare the …

Noise2clean: Cross-Device Side-Channel Traces Denoising With Unsupervised Deep Learning, Honggang Yu, Mei Wang, Xiyu Song, Haoqi Shan, Hongbing Qiu, Junyi Wang, Kaichen Yang

Michigan Tech Publications

Deep learning (DL)-based side-channel analysis (SCA) has posed a severe challenge to the security and privacy of embedded devices. During its execution, an attacker exploits physical SCA leakages collected from profiling devices to create a DL model for recovering secret information from victim devices. Despite this success, recent works have demonstrated that certain countermeasures, such as random delay interrupts or clock jitters, would make these attacks more complex and less practical in real-world scenarios. To address this challenge, we present a novel denoising scheme that exploits the U-Net model to pre-process SCA traces for “noises” (i.e., countermeasures) removal. Specifically, we …

Efficient Training On Alzheimer’S Disease Diagnosis With Learnable Weighted Pooling For 3d Pet Brain Image Classification, Xin Xing, Muhammad Usman Rafique, Gongbo Liang, Hunter Blanton, Zu Zhang, Chris Wang, Nathan Jacobs, Ai-Ling Lin

Computer Science Faculty Publications

Three-dimensional convolutional neural networks (3D CNNs) have been widely applied to analyze Alzheimer’s disease (AD) brain images for a better understanding of the disease progress or predicting the conversion from cognitively impaired (CU) or mild cognitive impairment status. It is well-known that training 3D-CNN is computationally expensive and with the potential of overfitting due to the small sample size available in the medical imaging field. Here we proposed a novel 3D-2D approach by converting a 3D brain image to a 2D fused image using a Learnable Weighted Pooling (LWP) method to improve efficient training and maintain comparable model performance. By …

An Empirical Study Of Pre-Trained Model Reuse In The Hugging Face Deep Learning Model Registry, Wenxin Jiang, Nicholas Synovic, Matt Hyatt, Taylor R. Schorlemmer, Rohan Sethi, Yung-Hsiang Lu, George K. Thiruvathukal, James C. Davis

Department of Electrical and Computer Engineering Faculty Publications

Deep Neural Networks (DNNs) are being adopted as components in software systems. Creating and specializing DNNs from scratch has grown increasingly difficult as state-of-the-art architectures grow more complex. Following the path of traditional software engineering, machine learning engineers have begun to reuse large-scale pre-trained models (PTMs) and fine-tune these models for downstream tasks. Prior works have studied reuse practices for traditional software packages to guide software engineers towards better package maintenance and dependency management. We lack a similar foundation of knowledge to guide behaviors in pre-trained model ecosystems.

In this work, we present the first empirical investigation of PTM reuse. …

An Effective Transfer Learning Based Landmark Detection Framework For Uav-Based Aerial Imagery Of Urban Landscapes, Bishwas Praveen, Vineetha Menon, Tathagata Mukherjee, Bryan Mesmer, Sampson Gholston, Steven Corns

Engineering Management and Systems Engineering Faculty Research & Creative Works

Aerial imagery captured through airborne sensors mounted on Unmanned Aerial Vehicles (UAVs), aircrafts, satellites, etc. in the form of RGB, LiDAR, multispectral or hyperspectral images provide a unique perspective for a variety of applications. These sensors capture high-resolution images that can be used for applications related to mapping, surveying, and monitoring of crops, infrastructure, and natural resources. Deep learning based algorithms are often the forerunners in facilitating practical solutions for such data-centric applications. Deep learning-based landmark detection is one such application which involves the use of deep learning algorithms to accurately identify and locate landmarks of interest in images captured …

Cooperative Deep $Q$ -Learning Framework For Environments Providing Image Feedback, Krishnan Raghavan, Vignesh Narayanan, Sarangapani Jagannathan

Electrical and Computer Engineering Faculty Research & Creative Works

In This Article, We Address Two Key Challenges in Deep Reinforcement Learning (DRL) Setting, Sample Inefficiency and Slow Learning, with a Dual-Neural Network (NN)-Driven Learning Approach. in the Proposed Approach, We Use Two Deep NNs with Independent Initialization to Robustly Approximate the Action-Value Function in the Presence of Image Inputs. in Particular, We Develop a Temporal Difference (TD) Error-Driven Learning (EDL) Approach, Where We Introduce a Set of Linear Transformations of the TD Error to Directly Update the Parameters of Each Layer in the Deep NN. We Demonstrate Theoretically that the Cost Minimized by the EDL Regime is an Approximation …

Instantaneous Frequency Estimation Of Fm Signals Under Gaussian And Symmetric Alpha-Stable Noise: Deep Learning Versus Time-Frequency Analysis, Huda Saleem Razzaq, Zahir M. Hussain

Research outputs 2022 to 2026

Deep learning (DL) and machine learning (ML) are widely used in many fields but rarely used in the frequency estimation (FE) and slope estimation (SE) of signals. Frequency and slope estimation for frequency-modulated (FM) and single-tone sinusoidal signals are essential in various applications, such as wireless communications, sound navigation and ranging (SONAR), and radio detection and ranging (RADAR) measurements. This work proposed a novel frequency estimation technique for instantaneous linear FM (LFM) sinusoidal wave using deep learning. Deep neural networks (DNN) and convolutional neural networks (CNN) are classes of artificial neural networks (ANNs) used for the frequency and slope estimation …

Embedded Ai For Wheat Yellow Rust Infection Type Classification, Uferah Shafi, Rafia Mumtaz, Muhammad Deedahwar Mazhar Qureshi, Zahid Mahmood, Sikander Khan Tanveer, Ihsan Ul Haq, Syed Mohammad Hassan Zaidi

Articles

Wheat is the most important and dominating crop in Pakistan in terms of production and acreage, which is grown on 37% of the cultivated area, accounting for 70% of the total production. However, wheat yield is highly affected by stripe rust, which is considered the most devastating fungal disease, causing 5.5 million tonnes of loss per year globally. In order to minimize this loss, the accurate and timely detection of rust disease is crucial instead of manual inspection. Towards this end, we propose a system to detect wheat rust disease and classify its infection types into four classes, including healthy, …

Interpreting Energy Utilisation With Shapley Additive Explanations By Defining A Synthetic Data Generator For Plausible Charging Sessions Of Electric Vehicles, Prasant Kumar Mohanty, Gayadhar Panda, Malabika Basu, Diptendu Sinha Roy

Articles

Electric vehicles (EVs) are an effective solution for reducing reliance on non-renewable energy sources. However, the lack of charging infrastructure and concerns over their range are some of the biggest hurdles to adopting EVs. Charging infrastructure for EVs is, however, on the rise. Proper planning of charging stations vis-`a-vis road networks and related points of interest such as transportation hubs, schools, shopping centres, etc., alongside such roads become vital to laying out a plan for such infrastructure, particularly for developing countries like India where EV adoption is relatively in a nascent stage. Synthetic datasets can help overcome these hurdles and …

Cooperative Deep Q -Learning Framework For Environments Providing Image Feedback, Krishnan Raghavan, Vignesh Narayanan, Sarangapani Jagannathan

Publications

In this article, we address two key challenges in deep reinforcement learning (DRL) setting, sample inefficiency, and slow learning, with a dual-neural network (NN)-driven learning approach. In the proposed approach, we use two deep NNs with independent initialization to robustly approximate the action-value function in the presence of image inputs. In particular, we develop a temporal difference (TD) error-driven learning (EDL) approach, where we introduce a set of linear transformations of the TD error to directly update the parameters of each layer in the deep NN. We demonstrate theoretically that the cost minimized by the EDL regime is an approximation …

Class Activation Mapping And Uncertainty Estimation In Multi-Organ Segmentation, Md. Shibly Sadique, Walia Farzana, Ahmed Temtam, Khan Iftekharuddin, Khan Iftekharuddin (Ed.), Weijie Chen (Ed.)

Electrical & Computer Engineering Faculty Publications

Deep learning (DL)-based medical imaging and image segmentation algorithms achieve impressive performance on many benchmarks. Yet the efficacy of deep learning methods for future clinical applications may become questionable due to the lack of ability to reason with uncertainty and interpret probable areas of failures in prediction decisions. Therefore, it is desired that such a deep learning model for segmentation classification is able to reliably predict its confidence measure and map back to the original imaging cases to interpret the prediction decisions. In this work, uncertainty estimation for multiorgan segmentation task is evaluated to interpret the predictive modeling in DL …

Hardware-In-The-Loop And Digital Twin Enabled Autonomous Robotics-Assisted Environment Inspection, Johnny Li, Bo Shang, Iresh Jayawardana, Genda Chen

Civil, Architectural and Environmental Engineering Faculty Research & Creative Works

Empowered by the advanced 3D sensing, computer vision and AI algorithm, autonomous robotics provide an unprecedented possibility for close-up infrastructure environment inspection in an efficient and reliable fashion. Deep neural network (DNN) learning algorithms, pretrained on the large database can empower real-time object detection as well as fully autonomous, safe robotic navigation in unstructured environments while avoiding the potential obstacle. However, the development and deployment of the robots, inspection planning and operation procedures are still tedious and segmented with tremendous manual intervention during environmental inspection and anomaly monitoring. The proposed digital twin approach is able to provide a virtual representation …

How Visual Stimuli Evoked P300 Is Transforming The Brain–Computer Interface Landscape: A Prisma Compliant Systematic Review, Jai Kalra, Prashasti Mittal, Nirmiti Mittal, Abhishek Arora, Utkarsh Tewari, Aviral Chharia, Rahul Upadhyay, Vinay Kumar, Luca Longo

Articles

Non-invasive Visual Stimuli evoked-EEGbased P300 BCIs have gained immense attention in recent years due to their ability to help patients with disability using BCI-controlled assistive devices and applications. In addition to the medical field, P300 BCI has applications in entertainment, robotics, and education. The current article systematically reviews 147 articles that were published between 2006-2021*. Articles that pass the pre-defined criteria are included in the study. Further, classification based on their primary focus, including article orientation, participants’ age groups, tasks given, databases, the EEG devices used in the studies, classification models, and application domain, is performed. The application-based classification considers …

Advances And Applications Of Dsmt For Information Fusion. Collected Works, Volume 5, Florentin Smarandache, Jean Dezert, Albena Tchamova

Branch Mathematics and Statistics Faculty and Staff Publications

This fifth volume on Advances and Applications of DSmT for Information Fusion collects theoretical and applied contributions of researchers working in different fields of applications and in mathematics, and is available in open-access. The collected contributions of this volume have either been published or presented after disseminating the fourth volume in 2015 (available at fs.unm.edu/DSmT-book4.pdf or www.onera.fr/sites/default/files/297/2015-DSmT-Book4.pdf) in international conferences, seminars, workshops and journals, or they are new. The contributions of each part of this volume are chronologically ordered.

First Part of this book presents some theoretical advances on DSmT, dealing mainly with modified Proportional Conflict Redistribution Rules (PCR) of …

Skin Lesion Segmentation In Dermoscopic Images With Noisy Data, Norsang Lama, Jason Hagerty, Anand Nambisan, Ronald Joe Stanley, William Van Stoecker

Electrical and Computer Engineering Faculty Research & Creative Works

We Propose a Deep Learning Approach to Segment the Skin Lesion in Dermoscopic Images. the Proposed Network Architecture Uses a Pretrained Efficient Net Model in the Encoder and Squeeze-And-Excitation Residual Structures in the Decoder. We Applied This Approach on the Publicly Available International Skin Imaging Collaboration (ISIC) 2017 Challenge Skin Lesion Segmentation Dataset. This Benchmark Dataset Has Been Widely Used in Previous Studies. We Observed Many Inaccurate or Noisy Ground Truth Labels. to Reduce Noisy Data, We Manually Sorted All Ground Truth Labels into Three Categories — Good, Mildly Noisy, and Noisy Labels. Furthermore, We Investigated the Effect of Such …

Machine Learning Based Pcb/Package Stack-Up Optimization For Signal Integrity, Wenchang Huang, Jiahuan Huang, Minseok Kim, Bumhee Bae, Chulsoon Hwang, Subin Kim

Electrical and Computer Engineering Faculty Research & Creative Works

PCB/package stack-up design optimization is time-consuming and requiring a great deal of experience. Although some iterative optimization algorithms are applied to implement automatic stack-up design, evaluating the results of each iteration is still time-intensive. This paper proposes a combined Bayesian optimization-artificial neural network (BO-ANN) algorithm, utilizing a trained ANN-based surrogate model to replace a 2D cross-section analysis tool for fast PCB/package stack-up design optimization. With the acceleration of ANN, the proposed BO-ANN algorithm can finish 100 iterations in 40 seconds while achieving the target characteristic impedance. To better generalize the BO-ANN algorithm, a strategy of effective dielectric calculation is applied …

Aerial Lidar-Based 3d Object Detection And Tracking For Traffic Monitoring, Baya Cherif, Hakim Ghazzai, Ahmad Alsharoa, Hichem Besbes, Yehia Massoud

Electrical and Computer Engineering Faculty Research & Creative Works

The proliferation of Light Detection and Ranging (LiDAR) technology in the automotive industry has quickly promoted its use in many emerging areas in smart cities and internet-of-things. Compared to other sensors, like cameras and radars, LiDAR provides up to 64 scanning channels, vertical and horizontal field of view, high precision, high detection range, and great performance under poor weather conditions. In this paper, we propose a novel aerial traffic monitoring solution based on Light Detection and Ranging (LiDAR) technology. By equipping unmanned aerial vehicles (UAVs) with a LiDAR sensor, we generate 3D point cloud data that can be used for …