Search

Air pollution poses significant public health risks, necessitating accurate and efficient monitoring of particulate matter (PM). These organic compounds may be released from natural sources like trees and vegetation, as well as from anthropogenic, or human-made sources including industrial activities and motor vehicle emissions. Therefore, measuring PM concentrations is paramount to understanding people's exposure levels to pollutants. This paper introduces a novel image processing technique utilizing photographs/pictures of Do-it-Yourself (DiY) sensors for the detection and quantification of P M 10 particles, enhancing community involvement and data collection accuracy in Citizen Science (CS) projects. A synthetic data generation algorithm was developed to overcome the challenge of data scarcity commonly associated with citizen-based data collection to validate the image processing technique. This algorithm generates images by precisely defining parameters such as image resolution, image dimension, and PM airborne particle density. To ensure these synthetic images mimic real-world conditions, variations like Gaussian noise, focus blur, and white balance adjustments and combinations were introduced, simulating the environmental and technical factors affecting image quality in typical smartphone digital cameras. The detection algorithm for P M 10 particles demonstrates robust performance across varying levels of noise, maintaining effectiveness in realistic mobile imaging conditions. Therefore, the methodology retains sufficient accuracy, suggesting its practical applicability for environmental monitoring in diverse real-world conditions using mobile devices. [ABSTRACT FROM AUTHOR]

Marker-less hand–eye calibration permits the acquisition of an accurate transformation between an optical sensor and a robot in unstructured environments. Single monocular cameras, despite their low cost and modest computation requirements, present difficulties for this purpose due to their incomplete correspondence of projected coordinates. In this work, we introduce a hand–eye calibration procedure based on the rotation representations inferred by an augmented autoencoder neural network. Learning-based models that attempt to directly regress the spatial transform of objects such as the links of robotic manipulators perform poorly in the orientation domain, but this can be overcome through the analysis of the latent space vectors constructed in the autoencoding process. This technique is computationally inexpensive and can be run in real time in markedly varied lighting and occlusion conditions. To evaluate the procedure, we use a color-depth camera and perform a registration step between the predicted and the captured point clouds to measure translation and orientation errors and compare the results to a baseline based on traditional checkerboard markers. [ABSTRACT FROM AUTHOR]

This paper explores a data augmentation approach for images of rigid bodies, particularly focusing on electrical equipment and analogous industrial objects. By leveraging manufacturer-provided datasheets containing precise equipment dimensions, we employed straightforward algorithms to generate synthetic images, permitting the expansion of the training dataset from a potentially unlimited viewpoint. In scenarios lacking genuine target images, we conducted a case study using two well-known detectors, representing two machine-learning paradigms: the Viola–Jones (VJ) and You Only Look Once (YOLO) detectors, trained exclusively on datasets featuring synthetic images as the positive examples of the target equipment, namely lightning rods and potential transformers. Performances of both detectors were assessed using real images in both visible and infrared spectra. YOLO consistently demonstrates F 1 scores below 26% in both spectra, while VJ's scores lie in the interval from 38% to 61%. This performance discrepancy is discussed in view of paradigms' strengths and weaknesses, whereas the relatively high scores of at least one detector are taken as empirical evidence in favor of the proposed data augmentation approach. [ABSTRACT FROM AUTHOR]

This article describes a novel fusion of a generative formal model for three-dimensional (3D) shapes with deep learning (DL) methods to understand the geometric structure of 3D objects and the relationships between their components, given a collection of unorganized point cloud measurements. Formal 3D shape models are implemented as shape grammar programs written in Procedural Shape Modeling Language (PSML). Users write PSML programs to describe complex objects, and DL networks estimate the configured free parameters of the program to generate 3D shapes. Users write PSML programs to enforce fundamental rules that define an object class and encode object attributes, including shapes, components, size, position, etc., into a parametric representation of objects. This fusion of the generative model with DL offers artificial intelligence (AI) models an opportunity to better understand the geometric organization of objects in terms of their components and their relationships to other objects. This approach allows human-in-the-loop control over DL estimates by specifying lists of candidate objects, the shape variations that each object can exhibit, and the level of detail or, equivalently, dimension of the latent representation of the shape. The results demonstrate the advantages of the proposed method over competing approaches. [ABSTRACT FROM AUTHOR]

Smartwatch health sensor data are increasingly utilized in smart health applications and patient monitoring, including stress detection. However, such medical data often comprise sensitive personal information and are resource-intensive to acquire for research purposes. In response to this challenge, we introduce the privacy-aware synthetization of multi-sensor smartwatch health readings related to moments of stress, employing Generative Adversarial Networks (GANs) and Differential Privacy (DP) safeguards. Our method not only protects patient information but also enhances data availability for research. To ensure its usefulness, we test synthetic data from multiple GANs and employ different data enhancement strategies on an actual stress detection task. Our GAN-based augmentation methods demonstrate significant improvements in model performance, with private DP training scenarios observing an 11.90–15.48% increase in F1-score, while non-private training scenarios still see a 0.45% boost. These results underline the potential of differentially private synthetic data in optimizing utility–privacy trade-offs, especially with the limited availability of real training samples. Through rigorous quality assessments, we confirm the integrity and plausibility of our synthetic data, which, however, are significantly impacted when increasing privacy requirements. [ABSTRACT FROM AUTHOR]

Industrial-quality inspections, particularly those leveraging AI, require significant amounts of training data. In fields like injection molding, producing a multitude of defective parts for such data poses environmental and financial challenges. Synthetic training data emerge as a potential solution to address these concerns. Although the creation of realistic synthetic 2D images from 3D models of injection-molded parts involves numerous rendering parameters, the current literature on the generation and application of synthetic data in industrial-quality inspection scarcely addresses the impact of these parameters on AI efficacy. In this study, we delve into some of these key parameters, such as camera position, lighting, and computational noise, to gauge their effect on AI performance. By utilizing Blender software, we procedurally introduced the "flash" defect on a 3D model sourced from a CAD file of an injection-molded part. Subsequently, with Blender's Cycles rendering engine, we produced datasets for each parameter variation. These datasets were then used to train a pre-trained EfficientNet-V2 for the binary classification of the "flash" defect. Our results indicate that while noise is less critical, using a range of noise levels in training can benefit model adaptability and efficiency. Variability in camera positioning and lighting conditions was found to be more significant, enhancing model performance even when real-world conditions mirror the controlled synthetic environment. These findings suggest that incorporating diverse lighting and camera dynamics is beneficial for AI applications, regardless of the consistency in real-world operational settings. [ABSTRACT FROM AUTHOR]

Perception algorithms for autonomous vehicles demand large, labeled datasets. Real-world data acquisition and annotation costs are high, making synthetic data from simulation a cost-effective option. However, training on one source domain and testing on a target domain can cause a domain shift attributed to local structure differences, resulting in a decrease in the model's performance. We propose a novel domain adaptation approach to address this challenge and to minimize the domain shift between simulated and real-world LiDAR data. Our approach adapts 3D point clouds on the object level by learning the local characteristics of the target domain. A key feature involves downsampling to ensure domain invariance of the input data. The network comprises a state-of-the-art point completion network combined with a discriminator to guide training in an adversarial manner. We quantify the reduction in domain shift by training object detectors with the source, target, and adapted datasets. Our method successfully reduces the sim-to-real domain shift in a distribution-aligned dataset by almost 50%, from 8.63% to 4.36% 3D average precision. It is trained exclusively using target data, making it scalable and applicable to adapt point clouds from any source domain. [ABSTRACT FROM AUTHOR]

An increasing number of special-use and high-rise buildings have presented challenges for efficient evacuations, particularly in fire emergencies. At the same time, however, the use of autonomous vehicles within indoor environments has received only limited attention for emergency scenarios. To address these issues, we developed a method that classifies emergency symbols and determines their location on emergency floor plans. The method incorporates color filtering, clustering and object detection techniques to extract walls, which were used in combination to generate clean, digitized plans. By integrating the geometric and semantic data digitized with our method, existing building information modeling (BIM) based evacuation tools can be enhanced, improving their capabilities for path planning and decision making. We collected a dataset of 403 German emergency floor plans and created a synthetic dataset comprising 5000 plans. Both datasets were used to train two distinct faster region-based convolutional neural networks (Faster R-CNNs). The models were evaluated and compared using 83 floor plan images. The results show that the synthetic model outperformed the standard model for rare symbols, correctly identifying symbol classes that were not detected by the standard model. The presented framework offers a valuable tool for digitizing emergency floor plans and enhancing digital evacuation applications. [ABSTRACT FROM AUTHOR]

The rapid growth in dataset sizes in modern deep learning has significantly increased data storage costs. Furthermore, the training and time costs for deep neural networks are generally proportional to the dataset size. Therefore, reducing the dataset size while maintaining model performance is an urgent research problem that needs to be addressed. Dataset condensation is a technique that aims to distill the original dataset into a much smaller synthetic dataset while maintaining downstream training performance on any agnostic neural network. Previous work has demonstrated that matching the training trajectory between the synthetic dataset and the original dataset is more effective than matching the instantaneous gradient, as it incorporates long-range information. Despite the effectiveness of trajectory matching, it suffers from complex gradient unrolling across iterations, which leads to significant memory and computation overhead. To address this issue, this paper proposes a novel approach called Expert Subspace Projection (ESP), which leverages long-range information while avoiding gradient unrolling. Instead of strictly enforcing the synthetic dataset's training trajectory to mimic that of the real dataset, ESP only constrains it to lie within the subspace spanned by the training trajectory of the real dataset. The memory-saving advantage offered by our method facilitates unbiased training on the complete set of synthetic images and seamless integration with other dataset condensation techniques. Through extensive experiments, we have demonstrated the effectiveness of our approach. Our method outperforms the trajectory matching method on CIFAR10 by 16.7% in the setting of 1 Image/Class, surpassing the previous state-of-the-art method by 3.2%. [ABSTRACT FROM AUTHOR]

To better control eutrophication, reliable and accurate information on phosphorus and nitrogen loading is desired. However, the high-frequency monitoring of these variables is economically impractical. This necessitates using virtual sensing to predict them by utilizing easily measurable variables as inputs. While the predictive performance of these data-driven, virtual-sensor models depends on the use of adequate training samples (in quality and quantity), the procurement and operational cost of nitrogen and phosphorus sensors make it impractical to acquire sufficient samples. For this reason, the variational autoencoder, which is one of the most prominent methods in generative models, was utilized in the present work for generating synthetic data. The generation capacity of the model was verified using water-quality data from two tributaries of the River Thames in the United Kingdom. Compared to the current state of the art, our novel data augmentation—including proper experimental settings or hyperparameter optimization—improved the root mean squared errors by 23–63%, with the most significant improvements observed when up to three predictors were used. In comparing the predictive algorithms' performances (in terms of the predictive accuracy and computational cost), k-nearest neighbors and extremely randomized trees were the best-performing algorithms on average. [ABSTRACT FROM AUTHOR]

A key challenge in further improving infrared (IR) sensor capabilities is the development of efficient data pre-processing algorithms. This paper addresses this challenge by providing a mathematical model and synthetic data generation framework for an uncooled IR sensor. The developed model is capable of generating synthetic data for the design of data pre-processing algorithms of uncooled IR sensors. The mathematical model accounts for the physical characteristics of the focal plane array, bolometer readout, optics and the environment. The framework permits the sensor simulation with a range of sensor configurations, pixel defectiveness, non-uniformity and noise parameters. [ABSTRACT FROM AUTHOR]

Companies that own water systems to provide water storage and distribution services always strive to enhance and efficiently distribute water to different places for various purposes. However, these water systems are likely to face problems ranging from leakage to destruction of infrastructures, leading to economic and life losses. Thus, apprehending the nature of abnormalities that may interrupt or aggravate the service or cause the destruction is at the core of their business model. Normally, companies use sensor networks to monitor these systems and record operational data including any fluctuations in water levels considered abnormalities. Detecting abnormalities allows water companies to enhance the service's sustainability, quality, and affordability. This study investigates a 2D-CNN-based method for detecting water-level abnormalities as time-series anomaly pattern detection in the One-Class Classification (OCC) problem. Moreover, since abnormal data are usually scarce or unavailable, we explored a cheap method to generate synthetic temporal data and use them as a target class in addition to the normal data to train the CNN model for feature extraction and classification. These settings allow us to train a model to learn relevant pattern representations of the given classes in a binary classification fashion using cross-entropy loss. The ultimate goal of these investigations is to determine if any 2D-CNN-based model can be trained from scratch or if transfer learning of any pre-trained CNN model can be partially trained and used as the base network for one-class classification. The evaluation of the proposed One-Class CNN and previous approaches have shown that our approach has outperformed several state-of-the-art approaches by a significant margin. Additionally, in this paper, we mention two interesting findings: using synthetic data as the pseudo-class is a promising direction, and transfer learning should be dealt with considering that underfitting can happen because the transferred model is too complicated for training data. [ABSTRACT FROM AUTHOR]

Weather prediction from real-world images can be termed a complex task when targeting classification using neural networks. Moreover, the number of images throughout the available datasets can contain a huge amount of variance when comparing locations with the weather those images are representing. In this article, the capabilities of a custom built driver simulator are explored specifically to simulate a wide range of weather conditions. Moreover, the performance of a new synthetic dataset generated by the above simulator is also assessed. The results indicate that the use of synthetic datasets in conjunction with real-world datasets can increase the training efficiency of the CNNs by as much as 74%. The article paves a way forward to tackle the persistent problem of bias in vision-based datasets. [ABSTRACT FROM AUTHOR]

Semantic segmentation of an incoming visual stream from cameras is an essential part of the perception system of self-driving cars. State-of-the-art results in semantic segmentation have been achieved with deep neural networks (DNNs), yet training them requires large datasets, which are difficult and costly to acquire and time-consuming to label. A viable alternative to training DNNs solely on real-world datasets is to augment them with synthetic images, which can be easily modified and generated in large numbers. In the present study, we aim at improving the accuracy of semantic segmentation of urban scenes by augmenting the Cityscapes real-world dataset with synthetic images generated with the open-source driving simulator CARLA (Car Learning to Act). Augmentation with synthetic images with a low degree of photorealism from the MICC-SRI (Media Integration and Communication Center–Semantic Road Inpainting) dataset does not result in the improvement of the accuracy of semantic segmentation, yet both MobileNetV2 and Xception DNNs used in the present study demonstrate a better accuracy after training on the custom-made CCM (Cityscapes-CARLA Mixed) dataset, which contains both real-world Cityscapes images and high-resolution synthetic images generated with CARLA, than after training only on the real-world Cityscapes images. However, the accuracy of semantic segmentation does not improve proportionally to the amount of the synthetic data used for augmentation, which indicates that augmentation with a larger amount of synthetic data is not always better. [ABSTRACT FROM AUTHOR]

Submarines are considered extremely strategic for any naval army due to their stealth capability. Periscopes are crucial sensors for these vessels, and emerging to the surface or periscope depth is required to identify visual contacts through this device. This maneuver has many procedures and usually has to be fast and agile to avoid exposure. This paper presents and implements a novel architecture for real submarine periscopes developed for future Brazilian naval fleet operations. Our system consists of a probe that is connected to the craft and carries a 360 camera. We project and take the images inside the vessel using traditional VR/XR devices. We also propose and implement an efficient computer vision-based MR technique to estimate and display detected vessels effectively and precisely. The vessel detection model is trained using synthetic images. So, we built and made available a dataset composed of 99,000 images. Finally, we also estimate distances of the classified elements, showing all the information in an AR-based interface. Although the probe is wired-connected, it allows for the vessel to stand in deep positions, reducing its exposure and introducing a new way for submarine maneuvers and operations. We validate our proposal through a user experience experiment using 19 experts in periscope operations. [ABSTRACT FROM AUTHOR]

The aging population has resulted in interest in remote monitoring of elderly individuals' health and well being. This paper describes a simple unsupervised monitoring system that can automatically detect if an elderly individual's pattern of presence deviates substantially from the recent past. The proposed system uses a small set of low-cost motion sensors and analyzes the produced data to establish an individual's typical presence pattern. Then, the algorithm uses a distance function to determine whether the individual's observed presence for each day significantly deviates from their typical pattern. Empirically, the algorithm is validated on both synthetic data and data collected by installing our system in the residences of three older individuals. In the real-world setting, the system detected, respectively, five, four, and one deviating days in the three locations. The deviating days detected by the system could result from a health issue that requires attention. The information from the system can aid caregivers in assessing the subject's health status and allows for a targeted intervention. Although the system can be refined, we show that otherwise hidden but relevant events (e.g., fall incident and irregular sleep patterns) are detected and reported to the caregiver. [ABSTRACT FROM AUTHOR]

Gait analysis based on inertial sensors has become an effective method of quantifying movement mechanics, such as joint kinematics and kinetics. Machine learning techniques are used to reliably predict joint mechanics directly from streams of IMU signals for various activities. These data-driven models require comprehensive and representative training datasets to be generalizable across the movement variability seen in the population at large. Bottlenecks in model development frequently occur due to the lack of sufficient training data and the significant time and resources necessary to acquire these datasets. Reliable methods to generate synthetic biomechanical training data could streamline model development and potentially improve model performance. In this study, we developed a methodology to generate synthetic kinematics and the associated predicted IMU signals using open source musculoskeletal modeling software. These synthetic data were used to train neural networks to predict three degree-of-freedom joint rotations at the hip and knee during gait either in lieu of or along with previously measured experimental gait data. The accuracy of the models' kinematic predictions was assessed using experimentally measured IMU signals and gait kinematics. Models trained using the synthetic data out-performed models using only the experimental data in five of the six rotational degrees of freedom at the hip and knee. On average, root mean square errors in joint angle predictions were improved by 38% at the hip (synthetic data RMSE: 2.3°, measured data RMSE: 4.5°) and 11% at the knee (synthetic data RMSE: 2.9°, measured data RMSE: 3.3°), when models trained solely on synthetic data were compared to measured data. When models were trained on both measured and synthetic data, root mean square errors were reduced by 54% at the hip (measured + synthetic data RMSE: 1.9°) and 45% at the knee (measured + synthetic data RMSE: 1.7°), compared to measured data alone. These findings enable future model development for different activities of clinical significance without the burden of generating large quantities of gait lab data for model training, streamlining model development, and ultimately improving model performance. [ABSTRACT FROM AUTHOR]

In contrast to previous works on data augmentation using LIDAR (Light Detection and Ranging), which mostly consider point clouds under good weather conditions, this paper uses point clouds which are affected by spray. Spray water can be a cause of phantom braking and understanding how to handle the extra detections caused by it is an important step in the development of ADAS (Advanced Driver Assistance Systems)/AV (Autonomous Vehicles) functions. The extra detections caused by spray cannot be safely removed without considering cases in which real solid objects may be present in the same region in which the detections caused by spray take place. As collecting real examples would be extremely difficult, the use of synthetic data is proposed. Real scenes are reconstructed virtually with an added extra object in the spray region, in a way that the detections caused by this obstacle match the characteristics a real object in the same position would have regarding intensity, echo number and occlusion. The detections generated by the obstacle are then used to augment the real data, obtaining, after occlusion effects are added, a good approximation of the desired training data. This data is used to train a classifier achieving an average F-Score of 92. The performance of the classifier is analyzed in detail based on the characteristics of the synthetic object: size, position, reflection, duration. The proposed method can be easily expanded to different kinds of obstacles and classifier types. [ABSTRACT FROM AUTHOR]

Creation of realistic synthetic behavior-based sensor data is an important aspect of testing machine learning techniques for healthcare applications. Many of the existing approaches for generating synthetic data are often limited in terms of complexity and realism. We introduce SynSys, a machine learning-based synthetic data generation method, to improve upon these limitations. We use this method to generate synthetic time series data that is composed of nested sequences using hidden Markov models and regression models which are initially trained on real datasets. We test our synthetic data generation technique on a real annotated smart home dataset. We use time series distance measures as a baseline to determine how realistic the generated data is compared to real data and demonstrate that SynSys produces more realistic data in terms of distance compared to random data generation, data from another home, and data from another time period. Finally, we apply our synthetic data generation technique to the problem of generating data when only a small amount of ground truth data is available. Using semi-supervised learning we demonstrate that SynSys is able to improve activity recognition accuracy compared to using the small amount of real data alone. [ABSTRACT FROM AUTHOR]