Recently, intelligent settings have begun to demand real-time indoor positioning, particularly where GPS signals are not accessible or distrusted. Although Received Signal Strength Indicator (RSSI) readings are very susceptible to multipath propagation, signal attenuation, and indoor interference, Bluetooth Low Energy (BLE) can be used as an inexpensive option To address these challenges, this study proposes a machine learning-driven indoor localization framework that emphasizes signal stability-oriented preprocessing, robust feature engineering, and a dynamically weighted hybrid ensemble strategy to improve resilience against environmental noise and device heterogeneity. The proposed approach evaluates multiple regression and ensemble learning models using experimentally collected BLE RSSI data. Results show that tree-based and ensemble methods outperform conventional approaches by effectively modeling nonlinear and noisy RSSI-distance relationships. Random Forest achieved a high average $R^{\mathrm{2}}$ of 0.993 with low MAE (0.113) and RMSE (0.352), while XGBoost produced the lowest mean Euclidean positioning error (0.177). Stacking ensembles based on ElasticNet and Ridge further improved prediction stability, surpassing deep learning baselines such as LSTM in overall localization accuracy. The findings demonstrate that combining stability-focused preprocessing with adaptive ensemble learning provides a reliable and accurate BLE-based indoor localization solution, offering improved robustness compared to traditional machine learning and deep learning methods.

Timely fire detection in industrial environments is crucial to safeguarding people and property. Deep neural networks have shown effectiveness in fire detection over traditional methods. However, they require high-quality datasets, which are costly and time-intensive to gather. To overcome this issue, we created the SYN-FIRE dataset, consisting of 2000 labeled images of simulated indoor industrial fires using NVIDIA Omniverse. By using U-Net++ as the baseline, this study explores the impact of the new SYN-FIRE dataset on models' performance when combined with four publicly available datasets. Two ablation studies were conducted: one replacing portions of real data from publicly available datasets with synthetic data and the other adding various amounts of synthetic data. With over 200 models trained across three resolutions, the results indicate that incorporating additional synthetic data improved DiceScore by [Formula: see text] to [Formula: see text] (FireBot and BowFire datasets, respectively) while substituting real data with synthetic data generally enhanced performance but with exceptions. Furthermore, tests on challenging real-life fire images confirmed that synthetic data boosts model generalization, supported by GRAD-CAM saliency maps. Finally, we provide key takeaways that point out the main findings of our research. The SYN-FIRE dataset is publicly available to encourage further research in fire detection and prevention.

Early fire detection and timely notification are paramount for preventing human and material casualties caused by fire. As a result, scientists have developed various fire monitoring systems based on sensors and images. Image-based systems have proven more advantageous than sensor-based ones as they provide additional details about the fire, such as its location, intensity, and progression. However, accurately determining the shape and boundaries of flames in image-based systems is difficult due to various factors, including backgrounds, different fire sizes, and interference from objects that resemble flames. In this study, we first evaluate convolutional neural networks used in related research and benchmark them on our challenging indoor fire dataset. Secondly, we propose a novel Feature Merging Model (F2M), which combines the output fire segmentation masks of the top five segmentation models obtained through the performed evaluation. The F2M relies on a novel mechanism that introduces bias into the mask estimation and achieves better results than all the tested methods, along with Monte-Carlo dropout for uncertainty estimation. The F2M ensemble-based uncertainty estimation model achieved improvements in comparison to the best performance convolutional neural network U-Net++, on three different image resolutions: 256 × 256, 640 × 640 and 800 × 800, according to the following metrics: Total error, Dice coefficient, and IoU score.

Fire presents a dangerous occurrence in inhabited places or places occupied by humans. Apart from very high material damage, it can result in human fatalities. In the past, numerous sensor-based methods were developed in order to reduce the reaction time and improve fire detection accuracy. With the advent of modern convolutional neural networks (CNN), object detection models quickly emerged, together with novel fire detection methods utilizing said models. In this paper, we aim to demonstrate the performance of YOLOv5 on our custom dataset, which contains indoor fire occurrences. Furthermore, we are investigating the influence of the input image resolution on the models' performances. This is important due to the hardware limitations of the models that would be used in real-life applications. The research shows that different metrics (F1-Score, inference time, mAP50) can yield different models as the best-performing ones. However, since the models have to be relatively small, quick, and accurate, we have proposed the ranking-based evaluation of the models from the aspect of the input image resolution. The evaluation showed that the models obtained the best overall score when the input image resolution was set to $512 \times 512$ pixels.

Sliding-window-based low-rank matrix approximation (LRMA) is a technique widely used in hyperspectral images (HSIs) denoising or completion. However, the uncertainty quantification of the restored HSI has not been addressed to date. Accurate uncertainty quantification of the denoised HSI facilitates applications such as multisource or multiscale data fusion, data assimilation, and product uncertainty quantification since these applications require an accurate approach to describe the statistical distributions of the input data. Therefore, we propose a prior-free closed-form element-wise uncertainty quantification method for LRMA-based HSI restoration. Our closed-form algorithm overcomes the difficulty of handling uncertainty in HSI patch mixing caused by the sliding-window strategy used in the conventional LRMA process. The proposed approach only requires the uncertainty of the observed HSI and provides the uncertainty result relatively rapidly and with similar computational complexity as the LRMA technique. We conduct extensive experiments to validate the estimation accuracy of the proposed closed-form uncertainty approach. The method is robust to at least 10% random impulse noise at the cost of 10%–20% of additional processing time compared to the LRMA. The experiments indicate that the proposed closed-form uncertainty quantification method is more applicable to real-world applications than the baseline Monte Carlo test, which is computationally expensive.

Semi-Definite Programming (SDP) with low-rank prior has been widely applied in Non-Rigid Structure from Motion (NRSfM). Based on a low-rank constraint, it avoids the inherent ambiguity of basis number selection in conventional base-shape or base-trajectory methods. Despite the efficiency in deformable shape reconstruction, it remains unclear how to assess the uncertainty of the recovered shape from the SDP process. In this paper, we present a statistical inference on the element-wise uncertainty quantification of the estimated deforming 3D shape points in the case of the exact low-rank SDP problem. A closed-form uncertainty quantification method is proposed and tested. Moreover, we extend the exact low-rank uncertainty quantification to the approximate low-rank scenario with a numerical optimal rank selection method, which enables solving practical application in SDP based NRSfM scenario. The proposed method provides an independent module to the SDP method and only requires the statistic information of the input 2D tracked points. Extensive experiments prove that the output 3D points have identical normal distribution to the 2D trackings, the proposed method and quantify the uncertainty accurately, and supports that it has desirable effects on routinely SDP low-rank based NRSfM solver.

This letter proposes a new image-based localization framework that explicitly localizes the camera/robot by fusing Convolutional Neural Network (CNN) and sequential images' geometric constraints. The camera is localized using a single or few observed images and training images with 6-degree-of-freedom pose labels. A Siamese network structure is adopted to train an image descriptor network, and the visually similar candidate image in the training set is retrieved to localize the testing image geometrically. Meanwhile, a probabilistic motion model predicts the pose based on a constant velocity assumption. The two estimated poses are finally fused using their uncertainties to yield an accurate pose prediction. This method leverages the geometric uncertainty and is applicable in indoor scenarios predominated by diffuse illumination. Experiments on simulation and real data sets demonstrate the efficiency of our proposed method. The results further show that combining the CNN-based framework with geometric constraint achieves better accuracy when compared with CNN-only methods, especially when the training data size is small.

Tracking monocular colonoscope in the GastroIntestinal (GI) tract is challenging as the obtained images suffer from deformation, blurred textures, and significant changes in appearance. These drawbacks greatly restrict the tracking ability of conventional geometry-based methods, which are heavily dependent on the performance of corner points extraction from the image. Even though end-to-end Deep Learning (DL) can overcome these issues, limited labeling data is a roadblock to the state-of-the-art DL-based method. To handle these drawbacks, we propose a novel approach to combine the DL-based method with the traditional geometry-based approach to achieve better localization with small training data. In this work, a DL network is trained with the images of the pre-operative endoscopy/colonoscopy. Siamese architecture is introduced to perform the zone labeling of the image based on the anatomical segmentation with expert knowledge. Then, using the image in the therapeutic intervention, our method predicts the 6 degrees of freedom scope pose and recover geometric reference to the images from the pre-operative endoscopy/colonoscopy. The DL network predicts the zone of the testing image, and the pre-generated triangulated map points within the zone in the training set are registered with the bundle adjustment algorithm. The proposed hybrid method is tested on the synthetic data sets and the real-world in-vivo data sets. Further, the results achieved through various experiments validate that the proposed method outperforms traditional geometry-based only or DL-based only localization techniques.

Learning a mapping between two unrelated domainssuch as image and audio, without any supervision is a challenging task. In this work, we propose to use a distancepreserving generative adversarial model to translate images of human faces into an audio domain. The audio domain is defined by a collection of musical note sounds recorded by 10 different instrument families (NSynth [5]) and a distance metric where the instrument family class information is incorporated together with a mel-frequency cepstral coefficients (MFCCs) feature. To enforce distancepreservation, a loss term that penalizes the difference between pairwise distances of the faces and the translated audio samples is used. Further, we discover that the distance preservation constraint in the generative adversarial model leads to reduced diversity in the translated audio samples, and propose the use of an auxiliary discriminator to enhance the diversity of the translations while using the distance preservation constraint. We also provide a visual demonstration of the results (video demo) and numerical analysis of the fidelity of the translations.

Activity recognition is a core component of many intelligent and context-aware systems. We present a solution for discreetly and unobtrusively recognizing common work activities above a work surface without using cameras. We demonstrate our approach, which utilizes an RF-radar sensor mounted under the work surface, in three domains: recognizing work activities at a convenience-store counter, recognizing common office deskwork activities, and estimating the position of customers in a showroom environment. Our examples illustrate potential benefits for both post-hoc business analytics and for real-time applications. Our solution was able to classify seven clerk activities with 94.9% accuracy using data collected in a lab environment and able to recognize six common deskwork activities collected in real offices with 95.3% accuracy. Using two sensors simultaneously, we demonstrate coarse position estimation around a large surface with 95.4% accuracy. We show that using multiple projections of RF signal leads to improved recognition accuracy. Finally, we show how smartwatches worn by users can be used to attribute an activity, recognized with the RF sensor, to a particular user in multi-user scenarios. We believe our solution can mitigate some of users’ privacy concerns associated with cameras and is useful for a wide range of intelligent systems.

Semi-Definite Programming (SDP) with low-rank prior has been widely applied in Non-Rigid Structure from Motion (NRSfM). Based on a low-rank constraint, it avoids the inherent ambiguity of basis number selection in conventional base-shape or base-trajectory methods. Despite the efficiency in deformable shape reconstruction, it remains unclear how to assess the uncertainty of the recovered shape from the SDP process. In this paper, we present a statistical inference on the element-wise uncertainty quantification of the estimated deforming 3D shape points in the case of the exact low-rank SDP problem. A closed-form uncertainty quantification method is proposed and tested. Moreover, we extend the exact low-rank uncertainty quantification to the approximate low-rank scenario with a numerical optimal rank selection method, which enables solving practical application in SDP based NRSfM scenario. The proposed method provides an independent module to the SDP method and only requires the statistic information of the input 2D tracked points. Extensive experiments prove that the output 3D points have identical normal distribution to the 2D trackings, the proposed method and quantify the uncertainty accurately, and supports that it has desirable effects on routinely SDP low-rank based NRSfM solver.

As is expressed in the adage "a picture is worth a thousand words", when using spoken language to communicate visual information, brevity can be a challenge. This work describes a novel technique for leveraging machine-learned feature embeddings to sonify visual (and other types of) information into a perceptual audio domain, allowing users to perceive this information using only their aural faculty. The system uses a pretrained image embedding network to extract visual features and embed them in a compact subset of Euclidean space -- this converts the images into feature vectors whose $L^2$ distances can be used as a meaningful measure of similarity. A generative adversarial network (GAN) is then used to find a distance preserving map from this metric space of feature vectors into the metric space defined by a target audio dataset equipped with either the Euclidean metric or a mel-frequency cepstrum-based psychoacoustic distance metric. We demonstrate this technique by sonifying images of faces into human speech-like audio. For both target audio metrics, the GAN successfully found a metric preserving mapping, and in human subject tests, users were able to accurately classify audio sonifications of faces.

An important capability of most smart, Internet-of-Things-enabled spaces (e.g., office, home, hospital, factory) is the ability to leverage context of use. Location is a key context element, particularly indoor location. Recent advances in radio ranging technologies, such as Wi-Fi RTT, promise the availability of low-cost, near-ubiquitous time-of-flight-based ranging estimates. In this paper, we build on prior work to enhance this ranging technology's ability to provide useful location estimates. For further improvements, we model user motion behavior to estimate the user motion state by taking the temporal measurements available from time-of-flight ranging. We select the velocity parameter of a particle-filter-based on this motion state. We demonstrate meaningful improvements in coordinate-based estimation accuracy and substantial increases in room-level estimation accuracy. Furthermore, insights gained in our real-world deployment provide important implications for future Internet-of-Things applications and their supporting technology deployments such as social interaction, workflow management, inventory control, or healthcare information tools.

Learning a mapping between two unrelated domains-such as image and audio, without any supervision is a challenging task. In this work, we propose a distance-preserving generative adversarial model to translate images of human faces into an audio domain. The audio domain is defined by a collection of musical note sounds recorded by 10 different instrument families (NSynth ) and a distance metric where the instrument family class information is incorporated together with a mel-frequency cepstral coefficients (MFCCs) feature. To enforce distance-preservation, a loss term that penalizes difference between pairwise distances of the faces and the translated audio samples is used. Further, we discover that the distance preservation constraint in the generative adversarial model leads to reduced diversity in the translated audio samples, and propose the use of an auxiliary discriminator to enhance the diversity of the translations while using the distance preservation constraint. We also provide a visual demonstration of the results and numerical analysis of the fidelity of the translations. A video demo of our proposed model's learned translation is available in https://www.dropbox.com/s/the176w9obq8465/face_to_musical_note.mov?dl=0.

Localization in an indoor or Global Positioning System (GPS)-denied environment is paramount. It drives various applications that require locating humans or robots in an unknown environment. Various localization systems using different ubiquitous sensors such as camera, radio frequency, inertial measurement unit have been developed. Most of these systems cannot accommodate for scenarios which have substantial changes in the environment such as a large number of people (unpredictable) and sudden change in the environment floor plan (unstructured). In this paper, we propose a system, InFo that can leverage real-time visual information captured by surveillance cameras and augment that with images captured by the smart device user to deliver accurate discretized location information. Through our experiments, we demonstrate that our deep learning based InFo system provides an improvement of 10% as compared to a system that does not utilize this real-time information.

A motivating, core capability of most smart, Internet of Things enabled spaces (e.g., home, office, hospital, factory) is the ability to leverage context of use. Location is a key context element; particularly indoor location. Recent advances in radio ranging technologies, such as 802.11-2016 FTM, promise the availability of low-cost, near-ubiquitous time-of-flight-based ranging estimates. In this paper, we build on prior work to enhance the technology's ability to provide useful location estimates. We demonstrate meaningful improvements in coordinate-based estimation accuracy and substantial increases in room-level estimation accuracy. Furthermore, insights gained in our real-world deployment provides important implications for future Internet of Things context applications and their supporting technology deployments such as workflow management, inventory control, or healthcare information tools.

In this paper, we propose a novel solution to optimize the deployment of Radio Frequency (RF) beacons for the purpose of indoor localization. We propose a system that optimizes both the number of beacons and their placement in a given environment. We propose a novel cost-function, called CovBSM, that allows to simultaneously optimize the 3-coverage while maximizing the beacon spreading. Using this cost function, we propose a framework that maximize both the number of beacons and their placement in a given environment. The proposed solution accounts for the indoor infrastructure and its influence on the RF signal propagation by embedding a realistic simulator into the optimization process.

In this paper, we develop a system for the low-cost indoor localization and tracking problem using radio signal strength indicator, Inertial Measurement Unit (IMU), and magnetometer sensors. We develop a novel and simplified probabilistic IMU motion model as the proposal distribution of the sequential Monte-Carlo technique to track the robot trajectory. Our algorithm can globally localize and track a robot with a priori unknown location, given an informative prior map of the Bluetooth Low Energy (BLE) beacons. Also, we formulate the problem as an optimization problem that serves as the Back-end of the algorithm mentioned above (Front-end). Thus, by simultaneously solving for the robot trajectory and the map of BLE beacons, we recover a continuous and smooth trajectory of the robot, corrected locations of the BLE beacons, and the time-varying IMU bias. The evaluations achieved using hardware show that through the proposed closed-loop system the localization performance can be improved; furthermore, the system becomes robust to the error in the map of beacons by feeding back the optimized map to the Front-end.

In this paper, we propose a novel solution to optimize the deployment of Radio Frequency (RF) beacons for the purpose of indoor localization. We propose a system that optimizes both the number of beacons and their placement in a given environment. We propose a novel cost-function, called CovBSM, that allows to simultaneously optimize the 3coverage while maximizing the beacon spreading. Using this cost function, we propose a framework that maximize both the number of beacons and their placement in a given environment. The proposed solution accounts for the indoor infrastructure and its influence on the RF signal propagation by embedding a realistic simulator into the optimization process.

Accurate localization is a fundamental requirement for a variety of applications, ranging from industrial robot operations to location-powered applications on mobile devices. A key technical challenge in achieving this goal is providing a clean and reliable estimation of location from a variety of low-cost, uncalibrated sesnors. Many current techniques rely on Particle Filter (PF) based algorithms. They have proven successful at effectively fusing various sensors inputs to create meaningful location predictions. In this paper we build upon this large corpous of work. Like prior work, our technique fuses Received Signal Strength Indicator (RSSI) measurements from Bluetooth Low Energy (BLE) beacons with map information. A key contribution of our work is a new sensor model for BLE beacons that does not require the mapping from RSSI to distance. We further contribute a novel method of utilizing map information during the initialization of the system and during the resampling phase when new particles are generated. Using our proposed sensor model and map prior information the performance of the overall localization is improved by 1.20 m on comparing the 75th percentile of the cumulative distribution with traditional localization techniques.