Conjunctival goblet cells (CGCs) are specialized mucin-secreting epithelial cells, playing key roles for ocular surface homeostasis. Their examination is important for diagnosing various ocular surface disorders. However, existing imaging modalities have limitations in examining CGCs over large conjunctival regions. In this study, we present an oblique light sheet fluorescence microscopy (OLSFM) system integrated with real-time surface tracking for high-speed, large-area imaging of CGCs. The system incorporates an obliquely oriented imaging arm, providing high tolerance to variations in the sample surface. Continuous lateral scanning with pulsed light sheet illumination enables rapid large-area imaging. Additional real-time surface tracking maintains focus during the lateral scanning over the curved conjunctival surface. OLSFM successfully imaged an 8 mm² conjunctival area within 0.5 seconds in animal models, highlighting its potential for rapid CGC examination and the precise diagnosis of ocular surface diseases.

Previous Deepfake detection methods perform well within their training domains, but their effectiveness diminishes significantly with new synthesis techniques. Recent studies have revealed that detection models make decision boundaries based on facial identity instead of synthetic artifacts, leading to poor cross-domain performance. To address this issue, we propose FRIDAY, a novel training method that attenuates facial identity utilizing a face recognizer. To be specific, we first train a face recognizer using the same backbone as the Deepfake detector. We then freeze the recognizer and use it during the detector’s training to mitigate facial identity information. This is achieved by feeding input images into both the recognizer and the detector, then minimizing the similarity of their feature embeddings using our Facial Identity Attenuating loss. This process encourages the detector to produce embeddings distinct from the recognizer, effectively attenuating facial identity. Comprehensive experiments demonstrate that our approach significantly improves detection performance on both in-domain and cross-domain datasets.

This paper describes the Dubuque Electricity Portal, a city-scale system aimed at supporting voluntary reductions of electricity consumption. The Portal provided each household with fine-grained feedback on its electricity use, as well as using incentives, comparisons, and goal setting to encourage conservation. Logs, a survey and interviews were used to evaluate the user experience of the Portal during a 20-week pilot with 765 volunteer households. Although the volunteers had already made a wide range of changes to conserve electricity prior to the pilot, those who used the Portal decreased their electricity use by about 3.7%. They also reported increased understanding of their usage, and reported taking an array of actions - both changing their behavior and their electricity infrastructure. The paper discusses the experience of the system's users, and describes challenges for the design of ECF systems, including balancing accessibility and security, a preference for time-based visualizations, and the advisability of multiple modes of feedback, incentives and information presentation.

Recent advances in smart metering technology enable utility companies to have access to tremendous amount of smart meter data, from which the utility companies are eager to gain more insight about their customers. In this paper, we aim to detect electric heat pumps from coarse grained smart meter data for a heat pump marketing campaign. However, appliance detection is a challenging task, especially given a very low granularity and partial labeled even unlabeled data. Traditional methods install either a high granularity smart meter or sensors at every appliance, which is either too expensive or requires technical expertise. We propose a novel approach to detect heat pumps that utilizes low granularity smart meter data, prior sales data and weather data. In particular, motivated by the characteristics of heat pump consumption pattern, we extract novel features that are highly relevant to heat pump usage from smart meter data and weather data. Under the constraint that only a subset of heat pump users are available, we formalize the problem into a positive and unlabeled data classification and apply biased Support Vector Machine (BSVM) to our extracted features. Our empirical study on a real-world data set demonstrates the effectiveness of our method. Furthermore, our method has been deployed in a real-life setting where the partner electric company runs a targeted campaign for 292,496 customers. Based on the initial feedback, our detection algorithm can successfully detect substantial number of non-heat pump users who were identified heat pump users with the prior algorithm the company had used.

We present a long-term and cross-sectional study of a vibration-based water flow rate monitoring system in practical environments and scenarios. In our earlier research, we proved that a water flow monitoring system with vibration sensors is feasible by deploying and evaluating it in a small-scale laboratory setting. To validate the proposed system, the system was deployed in existing environments—two houses and a public restroom—and in two different laboratory test settings. With the collected data, we first demonstrate various aspects of the system's performance, including sensing stability, sensor node lifetime, the stability of autonomous sensor calibration, time to adaptation, and deployment complexity. We then discuss the practical challenges and lessons from the full-scale deployments. The evaluation results show that our water monitoring solution is a practical, quick-to-deploy system with a less than 5% average flow estimation error.

Data loss in wireless sensing applications is inevitable and while there have been many attempts at coping with this issue, recent developments in the area of Compressive Sensing (CS) provide a new and attractive perspective. Since many physical signals of interest are known to be sparse or compressible, employing CS, not only compresses the data and reduces effective transmission rate, but also improves the robustness of the system to channel erasures. This is possible because reconstruction algorithms for compressively sampled signals are not hampered by the stochastic nature of wireless link disturbances, which has traditionally plagued attempts at proactively handling the effects of these errors. In this paper, we propose that if CS is employed for source compression, then CS can further be exploited as an application layer erasure coding strategy for recovering missing data. We show that CS erasure encoding (CSEC) with random sampling is efficient for handling missing data in erasure channels, paralleling the performance of BCH codes, with the added benefit of graceful degradation of the reconstruction error even when the amount of missing data far exceeds the designed redundancy. Further, since CSEC is equivalent to nominal oversampling in the incoherent measurement basis, it is computationally cheaper than conventional erasure coding. We support our proposal through extensive performance studies.

Continuously understanding a user's location context in colloquial terms and the paths that connect the locations unlocks many opportunities for emerging applications. While extensive research effort has been made on efficiently tracking a user's raw coordinates, few attempts have been made to efficiently provide everyday contextual information about these locations as places and paths. We introduce SensLoc, a practical location service to provide such contextual information, abstracting location as place visits and path travels from sensor signals. SensLoc comprises of a robust place detection algorithm, a sensitive movement detector, and an on-demand path tracker. Based on a user's mobility, SensLoc proactively controls active cycle of a GPS receiver, a WiFi scanner, and an accelerometer. Pilot studies show that SensLoc can correctly detect 94% of the place visits, track 95% of the total travel distance, and still only consume 13% of energy than algorithms that periodically collect coordinates to provide the same information.

A key prerequisite for residential energy conservation is knowing when and where energy is being spent. Unfortunately, the current generation of energy reporting devices only provide partial and coarse grained information or require expensive professional installation. This limitation stems from the presumption that calculating per-appliance consumption requires per-appliance current measurements. However, since appliances typically emit measurable signals when they are consuming energy, we can estimate their consumption using indirect sensing. This paper presents ViridiScope, a fine-grained power monitoring system that furnishes users with an economical, self-calibrating tool that provides power consumption of virtually every appliance in the home. ViridiScope uses ambient signals from inexpensive sensors placed near appliances to estimate power consumption, thus no in-line sensor is necessary. We use a model-based machine learning algorithm that automates the sensor calibration process. Through experiments in a real house, we show that ViridiScope can estimate the end-point power consumption within 10% error.

Compressive Sensing (CS) is a recently developed mechanism that allows signal acquisition and compression to be performed in one inexpensive step so that the sampling process itself produces a compressed version of the signal. This significantly improves systemic energy efficiency because the average sampling rate can be considerably reduced and explicit compression eliminated.

Water is nature's most precious resource and growing demand is pushing fresh water supplies to the brink of non-renewability. New technological and social initiatives that enhance conservation and reduce waste are needed. Providing consumers with fine-grained real-time information has yielded benefits in conservation of power and gasoline. Extending this philosophy to water conservation, we introduce a novel water monitoring system, NAWMS, that similarly empowers users.