arXiv subjects

Hae Young Noh

Publications and source records attributed to Hae Young Noh.

31 records · Page 2Linked to original sources

arXiv2022

PigV$^2$: Monitoring Pig Vital Signs through Ground Vibrations Induced by Heartbeat and Respiration

Pig vital sign monitoring (e.g., estimating the heart rate (HR) and respiratory rate (RR)) is essential to understand the stress level of the sow and detect the onset of parturition. It helps to maximize peri-natal survival and improve animal well-being in swine production. The existing approach mainly relies on manual measurement, which is labor-intensive and only provides a few points of information. Other sensing modalities such as wearables and cameras are developed to enable more continuous measurement, but are still limited due to animal discomfort, data transfer, and storage challenges. In this paper, we introduce PigV$^2$, the first system to monitor pig heart rate and respiratory rate through ground vibrations. Our approach leverages the insight that both heartbeat and respiration generate ground vibrations when the sow is lying on the floor. We infer vital information by sensing and analyzing these vibrations. The main challenge in developing PigV$^2$ is the overlap of vital- and non-vital-related information in the vibration signals, including pig movements, pig postures, pig-to-sensor distances, and so on. To address this issue, we first characterize their effects, extract their current status, and then reduce their impact by adaptively interpolating vital rates over multiple sensors. PigV$^2$ is evaluated through a real-world deployment with 30 pigs. It has 3.4% and 8.3% average errors in monitoring the HR and RR of the sows, respectively.

eess.SP↗

arXiv2022

GaitVibe+: Enhancing Structural Vibration-based Footstep Localization Using Temporary Cameras for In-home Gait Analysis

In-home gait analysis is important for providing early diagnosis and adaptive treatments for individuals with gait disorders. Existing systems include wearables and pressure mats, but they have limited scalability. Recent studies have developed vision-based systems to enable scalable, accurate in-home gait analysis, but it faces privacy concerns due to the exposure of people's appearances. Our prior work developed footstep-induced structural vibration sensing for gait monitoring, which is device-free, wide-ranged, and perceived as more privacy-friendly. Although it has succeeded in temporal gait event extraction, it shows limited performance for spatial gait parameter estimation due to imprecise footstep localization. In particular, the localization error mainly comes from the estimation error of the wave arrival time at the vibration sensors and its error propagation to wave velocity estimations. Therefore, we present GaitVibe+, a vibration-based footstep localization method fused with temporarily installed cameras for in-home gait analysis. Our method has two stages: fusion and operating. In the fusion stage, both cameras and vibration sensors are installed to record only a few trials of the subject's footstep data, through which we characterize the uncertainty in wave arrival time and model the wave velocity profiles for the given structure. In the operating stage, we remove the camera to preserve privacy at home. The footstep localization is conducted by estimating the time difference of arrival (TDoA) over multiple vibration sensors, whose accuracy is improved through the reduced uncertainty and velocity modeling during the fusion stage. We evaluate GaitVibe+ through a real-world experiment with 50 walking trials. With only 3 trials of multi-modal fusion, our approach has an average localization error of 0.22 meters, which reduces the spatial gait parameter error from 111% to 27%.

eess.SP↗

arXiv2022

Vibration-Based Bridge Health Monitoring using Telecommunication Cables

Bridge Health Monitoring (BHM) enables early damage detection of bridges and is thus critical for avoiding more severe damages that might result in major financial and human losses. However, conventional BHM systems require dedicated sensors on bridges, which is costly to install and maintain and hard to scale up. To overcome this challenge, we introduce a new system that uses existing telecommunication cables for Distributed Acoustic Sensing (DAS) to collect bridge dynamic strain responses. In addition, we develop a two-module physics-guided system identification method to extract bridge damage-sensitive information (e.g., natural frequencies and mode shapes) from noisy DAS data by constraining strain and displacement mode shapes by bridge dynamics. This approach does not require installation and maintenance of dedicated sensors on bridges. We evaluate our system with field experiments on a concrete bridge with fiber cable running in a conduit under the deck. Our system successfully identified modal frequencies and reconstructed meter-scale mode shapes.

eess.SY↗

arXiv2022

Bayesian Updating of Seismic Ground Failure Estimates via Causal Graphical Models and Satellite Imagery

Earthquake-induced secondary ground failure hazards, such as liquefaction and landslides, result in catastrophic building and infrastructure damage as well as human fatalities. To facilitate emergency responses and mitigate losses, the U.S. Geological Survey provides a rapid hazard estimation system for earthquake-triggered landslides and liquefaction using geospatial susceptibility proxies and ShakeMap ground motion estimates. In this study, we develop a generalized causal graph-based Bayesian network that models the physical interdependencies between geospatial features, seismic ground failures, and building damage, as well as DPMs. Geospatial features provide physical insights for estimating ground failure occurrence while DPMs contain event-specific surface change observations. This physics-informed causal graph incorporates these variables with complex physical relationships in one holistic Bayesian updating scheme to effectively fuse information from both geospatial models and remote sensing data. This framework is scalable and flexible enough to deal with highly complex multi-hazard combinations. We then develop a stochastic variational inference algorithm to jointly update the intractable posterior probabilities of unobserved landslides, liquefaction, and building damage at different locations efficiently. In addition, a local graphical model pruning algorithm is presented to reduce the computational cost of large-scale seismic ground failure estimation. We apply this framework to the September 2018 Hokkaido Iburi-Tobu, Japan (M6.6) earthquake and January 2020 Southwest Puerto Rico (M6.4) earthquake to evaluate the performance of our algorithm.

cs.CE↗

arXiv2022

Predicting Peak Stresses In Microstructured Materials Using Convolutional Encoder-Decoder Learning

This work presents a machine learning approach to predict peak-stress clusters in heterogeneous polycrystalline materials. Prior work on using machine learning in the context of mechanics has largely focused on predicting the effective response and overall structure of stress fields. However, their ability to predict peak stresses -- which are of critical importance to failure -- is unexplored, because the peak-stress clusters occupy a small spatial volume relative to the entire domain, and hence requires computationally expensive training. This work develops a deep-learning based Convolutional Encoder-Decoder method that focuses on predicting peak-stress clusters, specifically on the size and other characteristics of the clusters in the framework of heterogeneous linear elasticity. This method is based on convolutional filters that model local spatial relations between microstructures and stress fields using spatially weighted averaging operations. The model is first trained against linear elastic calculations of stress under applied macroscopic strain in synthetically-generated microstructures, which serves as the ground truth. The trained model is then applied to predict the stress field given a (synthetically-generated) microstructure and then to detect peak-stress clusters within the predicted stress field. The accuracy of the peak-stress predictions is analyzed using the cosine similarity metric and by comparing the geometric characteristics of the peak-stress clusters against the ground-truth calculations. It is observed that the model is able to learn and predict the geometric details of the peak-stress clusters and, in particular, performed better for higher (normalized) values of the peak stress as compared to lower values of the peak stress. These comparisons showed that the proposed method is well-suited to predict the characteristics of peak-stress clusters.

math.AP↗

arXiv2021

HierMUD: Hierarchical Multi-task Unsupervised Domain Adaptation between Bridges for Drive-by Damage Diagnosis

Monitoring bridge health using vibrations of drive-by vehicles has various benefits, such as no need for directly installing and maintaining sensors on the bridge. However, many of the existing drive-by monitoring approaches are based on supervised learning models that require labeled data from every bridge of interest, which is expensive and time-consuming, if not impossible, to obtain. To this end, we introduce a new framework that transfers the model learned from one bridge to diagnose damage in another bridge without any labels from the target bridge. Our framework trains a hierarchical neural network model in an adversarial way to extract task-shared and task-specific features that are informative to multiple diagnostic tasks and invariant across multiple bridges. We evaluate our framework on experimental data collected from 2 bridges and 3 vehicles. We achieve accuracies of 95% for damage detection, 93% for localization, and up to 72% for quantification, which are ~2 times improvements from baseline methods.

cs.AI↗

arXiv2020

Knowledge Transfer between Buildings for Seismic Damage Diagnosis through Adversarial Learning

Automated structural damage diagnosis after earthquakes is important for improving the efficiency of disaster response and rehabilitation. In conventional data-driven frameworks which use machine learning or statistical models, structural damage diagnosis models are often constructed using supervised learning. The supervised learning requires historical structural response data and corresponding damage states (i.e., labels) for each building to learn the building-specific damage diagnosis model. However, in post-earthquake scenarios, historical data with labels are often not available for many buildings in the affected area. This makes it difficult to construct a damage diagnosis model. Further, directly using the historical data from other buildings to construct a damage diagnosis model for the target building would lead to inaccurate results. This is because each building has unique physical properties and thus unique data distribution. To this end, we introduce a new framework to transfer the model learned from other buildings to diagnose structural damage states in the target building without any labels. This framework is based on an adversarial domain adaptation approach that extracts domain-invariant feature representations of data from different buildings. The feature extraction function is trained in an adversarial way, which ensures that the extracted feature distributions are robust to changes in structures while being predictive of the damage states. With the extracted domain-invariant feature representations, the data distributions become consistent across different buildings. We evaluate our framework on both numerical simulation and field data collected from multiple building structures, which outperforms the state-of-the-art benchmark methods.

eess.SP↗

arXiv2020

O-MedAL: Online Active Deep Learning for Medical Image Analysis

Active Learning methods create an optimized labeled training set from unlabeled data. We introduce a novel Online Active Deep Learning method for Medical Image Analysis. We extend our MedAL active learning framework to present new results in this paper. Our novel sampling method queries the unlabeled examples that maximize the average distance to all training set examples. Our online method enhances performance of its underlying baseline deep network. These novelties contribute significant performance improvements, including improving the model's underlying deep network accuracy by 6.30%, using only 25% of the labeled dataset to achieve baseline accuracy, reducing backpropagated images during training by as much as 67%, and demonstrating robustness to class imbalance in binary and multi-class tasks.

cs.LG↗

arXiv2020

Knowledge transfer between bridges for drive-by monitoring using adversarial and multi-task learning

Monitoring bridge health using the vibrations of drive-by vehicles has various benefits, such as low cost and no need for direct installation or on-site maintenance of equipment on the bridge. However, many such approaches require labeled data from every bridge, which is expensive and time-consuming, if not impossible, to obtain. This is further exacerbated by having multiple diagnostic tasks, such as damage quantification and localization. One way to address this issue is to directly apply the supervised model trained for one bridge to other bridges, although this may significantly reduce the accuracy because of distribution mismatch between different bridges'data. To alleviate these problems, we introduce a transfer learning framework using domain-adversarial training and multi-task learning to detect, localize and quantify damage. Specifically, we train a deep network in an adversarial way to learn features that are 1) sensitive to damage and 2) invariant to different bridges. In addition, to improve the error propagation from one task to the next, our framework learns shared features for all the tasks using multi-task learning. We evaluate our framework using lab-scale experiments with two different bridges. On average, our framework achieves 94%, 97% and 84% accuracy for damage detection, localization and quantification, respectively. within one damage severity level.

cs.CV↗

arXiv2020

LaNet: Real-time Lane Identification by Learning Road SurfaceCharacteristics from Accelerometer Data

The resolution of GPS measurements, especially in urban areas, is insufficient for identifying a vehicle's lane. In this work, we develop a deep LSTM neural network model LaNet that determines the lane vehicles are on by periodically classifying accelerometer samples collected by vehicles as they drive in real time. Our key finding is that even adjacent patches of road surfaces contain characteristics that are sufficiently unique to differentiate between lanes, i.e., roads inherently exhibit differing bumps, cracks, potholes, and surface unevenness. Cars can capture this road surface information as they drive using inexpensive, easy-to-install accelerometers that increasingly come fitted in cars and can be accessed via the CAN-bus. We collect an aggregate of 60 km driving data and synthesize more based on this that capture factors such as variable driving speed, vehicle suspensions, and accelerometer noise. Our formulated LSTM-based deep learning model, LaNet, learns lane-specific sequences of road surface events (bumps, cracks etc.) and yields 100% lane classification accuracy with 200 meters of driving data, achieving over 90% with just 100 m (correspondingly to roughly one minute of driving). We design the LaNet model to be practical for use in real-time lane classification and show with extensive experiments that LaNet yields high classification accuracy even on smooth roads, on large multi-lane roads, and on drives with frequent lane changes. Since different road surfaces have different inherent characteristics or entropy, we excavate our neural network model and discover a mechanism to easily characterize the achievable classification accuracies in a road over various driving distances by training the model just once. We present LaNet as a low-cost, easily deployable and highly accurate way to achieve fine-grained lane identification.

cs.CV↗

arXiv2018

MedAL: Deep Active Learning Sampling Method for Medical Image Analysis

Deep learning models have been successfully used in medical image analysis problems but they require a large amount of labeled images to obtain good performance.Deep learning models have been successfully used in medical image analysis problems but they require a large amount of labeled images to obtain good performance. However, such large labeled datasets are costly to acquire. Active learning techniques can be used to minimize the number of required training labels while maximizing the model's performance.In this work, we propose a novel sampling method that queries the unlabeled examples that maximize the average distance to all training set examples in a learned feature space. We then extend our sampling method to define a better initial training set, without the need for a trained model, by using ORB feature descriptors. We validate MedAL on 3 medical image datasets and show that our method is robust to different dataset properties. MedAL is also efficient, achieving 80% accuracy on the task of Diabetic Retinopathy detection using only 425 labeled images, corresponding to a 32% reduction in the number of required labeled examples compared to the standard uncertainty sampling technique, and a 40% reduction compared to random sampling.

cs.CV↗

arXiv2018

An Incentive Mechanism for Crowd Sensing with Colluding Agents

Vehicular mobile crowd sensing is a fast-emerging paradigm to collect data about the environment by mounting sensors on vehicles such as taxis. An important problem in vehicular crowd sensing is to design payment mechanisms to incentivize drivers (agents) to collect data, with the overall goal of obtaining the maximum amount of data (across multiple vehicles) for a given budget. Past works on this problem consider a setting where each agent operates in isolation---an assumption which is frequently violated in practice. In this paper, we design an incentive mechanism to incentivize agents who can engage in arbitrary collusions. We then show that in a "homogeneous" setting, our mechanism is optimal, and can do as well as any mechanism which knows the agents' preferences a priori. Moreover, if the agents are non-colluding, then our mechanism automatically does as well as any other non-colluding mechanism. We also show that our proposed mechanism has strong (and asymptotically optimal) guarantees for a more general "heterogeneous" setting. Experiments based on synthesized data and real-world data reveal gains of over 30\% attained by our mechanism compared to past literature.

cs.GT↗

arXiv2016

Automated Synchronization of Driving Data Using Vibration and Steering Events

We propose a method for automated synchronization of vehicle sensors useful for the study of multi-modal driver behavior and for the design of advanced driver assistance systems. Multi-sensor decision fusion relies on synchronized data streams in (1) the offline supervised learning context and (2) the online prediction context. In practice, such data streams are often out of sync due to the absence of a real-time clock, use of multiple recording devices, or improper thread scheduling and data buffer management. Cross-correlation of accelerometer, telemetry, audio, and dense optical flow from three video sensors is used to achieve an average synchronization error of 13 milliseconds. The insight underlying the effectiveness of the proposed approach is that the described sensors capture overlapping aspects of vehicle vibrations and vehicle steering allowing the cross-correlation function to serve as a way to compute the delay shift in each sensor. Furthermore, we show the decrease in synchronization error as a function of the duration of the data stream.

cs.RO↗