Abstract: In various examples, sensor data—such as masked sensor data—may be used as input to a machine learning model to determine a confidence for object to person associations. The masked sensor data may focus the machine learning model on particular regions of the image that correspond to persons, objects, or some combination thereof. In some embodiments, coordinates corresponding to persons, objects, or combinations thereof, in addition to area ratios between various regions of the image corresponding to the persons, objects, or combinations thereof, may be used to further aid the machine learning model in focusing on important regions of the image for determining the object to person associations.

Abstract: In various examples, sensor data—such as masked sensor data—may be used as input to a machine learning model to determine a confidence for object to person associations. The masked sensor data may focus the machine learning model on particular regions of the image that correspond to persons, objects, or some combination thereof. In some embodiments, coordinates corresponding to persons, objects, or combinations thereof, in addition to area ratios between various regions of the image corresponding to the persons, objects, or combinations thereof, may be used to further aid the machine learning model in focusing on important regions of the image for determining the object to person associations.

Abstract: The present disclosure provides various approaches for smart area monitoring suitable for parking garages or other areas. These approaches may include ROI-based occupancy detection to determine whether particular parking spots are occupied by leveraging image data from image sensors, such as cameras. These approaches may also include multi-sensor object tracking using multiple sensors that are distributed across an area that leverage both image data and spatial information regarding the area, to provide precise object tracking across the sensors. Further approaches relate to various architectures and configurations for smart area monitoring systems, as well as visualization and processing techniques. For example, as opposed to presenting video of an area captured by cameras, 3D renderings may be generated and played from metadata extracted from sensors around the area.

Abstract: Methods and apparatuses are described for system for identifying and mitigating high-risk database queries through ranked variance analysis. A server identifies database queries executed against databases in a production computing environment within a predetermined time period, each database query associated with execution plans and each execution plan having corresponding plan data elements. For each database query: the server generates execution variance data for the execution plans for a database query based upon the corresponding plan data elements, comprising: determining an execution time variance between the execution plans; and determining a buffer gets variance between the execution plans. The server ranks the database queries according to (i) the execution time variance, and (ii) the buffer gets variance.

Abstract: An integrated dual-modality microfluidic sensor chip and methods for using the same. In one form, the sensor comprises a patterned periodic array of nanoposts coated with a noble metal and graphene oxide (GO) to detect target biomarker molecules in a limited sample volume. The device generates both electrochemical and surface plasmon resonance (SPR) signals from a single sensing area of the metal-GO nanoposts. The metal-GO nanoposts are functionalized with specific receptor molecules, serving as a spatially well-defined nanostructured working electrode for electrochemical sensing, as well as a nanostructured plasmonic crystal for SPR-based sensing via the excitation of surface plasmon polaritons.

Abstract: Transfer learning can be used to enable a user to obtain a machine learning model that is fully trained for an intended inferencing task without having to train the model from scratch. A pre-trained model can be obtained that is relevant for that inferencing task. Additional training data, as may correspond to at least one additional class of data, can be used to further train this model. This model can then be pruned and retrained in order to obtain a smaller model that retains high accuracy for the intended inferencing task.

Abstract: The present disclosure provides various approaches for smart area monitoring suitable for parking garages or other areas. These approaches may include ROI-based occupancy detection to determine whether particular parking spots are occupied by leveraging image data from image sensors, such as cameras. These approaches may also include multi-sensor object tracking using multiple sensors that are distributed across an area that leverage both image data and spatial information regarding the area, to provide precise object tracking across the sensors. Further approaches relate to various architectures and configurations for smart area monitoring systems, as well as visualization and processing techniques. For example, as opposed to presenting video of an area captured by cameras, 3D renderings may be generated and played from metadata extracted from sensors around the area.

Abstract: A technique, and its applications, for high resolution, rapid, and simple nanopatterning. The general method has been demonstrated in several forms and applications. One is patterning nanophotonic structures at an optical fiber tip for refractive index sensing. Another is patterning nanoresonator structures on a sensor substrate for plasmonic effect related detection of VOCs. In the latter example, a graphene oxide coated plasmonic crystal as a gas sensor capable of identifying different gas species using an array of such structures. By coating the surface of multiple identical plasmonic crystals with different thicknesses of Graphene-Oxide (GO) layer, the effective refractive index of the GO layer on each plasmonic crystal is differently modulated when exposed to a specific gas. Identification of various gas species is accomplished using pattern recognition algorithm.

Abstract: A tracking system comprising: a radio frequency (RF) signal circulator; an iridium modem coupled to first port of the circulator; an antenna coupled to second port of the circulator, wherein the circulator passes a signal transmitted by the modem to the antenna when the switch is switched to a first mode; a low noise amplifier (LNA) coupled to third port of the circulator, wherein the circulator passes RF signals received from the antenna to the LNA; a diplexer coupled to an output of the LNA; a GNSS receiver coupled to a first output of the diplexer through a GNSS filter; an iridium filter coupled to a second output of the diplexer; wherein the switch couples the iridium filter to the modem when the iridium modem is in a receiving mode, and wherein the switch couples the modem to the first port when the modem is in a transmitting mode.

Abstract: In various examples, sensor data—such as masked sensor data—may be used as input to a machine learning model to determine a confidence for object to person associations. The masked sensor data may focus the machine learning model on particular regions of the image that correspond to persons, objects, or some combination thereof. In some embodiments, coordinates corresponding to persons, objects, or combinations thereof, in addition to area ratios between various regions of the image corresponding to the persons, objects, or combinations thereof, may be used to further aid the machine learning model in focusing on important regions of the image for determining the object to person associations.

Abstract: In various examples, a neural network may be trained for use in vehicle re-identification tasks—e.g., matching appearances and classifications of vehicles across frames—in a camera network. The neural network may be trained to learn an embedding space such that embeddings corresponding to vehicles of the same identify are projected closer to one another within the embedding space, as compared to vehicles representing different identities. To accurately and efficiently learn the embedding space, the neural network may be trained using a contrastive loss function or a triplet loss function. In addition, to further improve accuracy and efficiency, a sampling technique—referred to herein as batch sample—may be used to identify embeddings, during training, that are most meaningful for updating parameters of the neural network.

Abstract: An electrophoresis chip-based setup for detection of different ions in buffer solution. The device is able to differentiate ions in, e.g., real soil or other solutions, and to detect concentration of a specific ion in the solution. Fabrication of the electrophoresis chip can use a soft lithography based molding process. The chip can be made out of PDMS on a glass substrate where on-chip valves were used to control timing of injecting sample and buffer solutions. Detection electrodes are used to detect the presence of ions over a period of several minutes. A controllable high voltage power supply system and related signal acquisition, processing and detection setup can be implemented with the sensor in a system. A microfluidic system for automated collection of soil sample through a porous ceramic and using vacuum suction can be used.

Abstract: The present disclosure provides various approaches for smart area monitoring suitable for parking garages or other areas. These approaches may include ROI-based occupancy detection to determine whether particular parking spots are occupied by leveraging image data from image sensors, such as cameras. These approaches may also include multi-sensor object tracking using multiple sensors that are distributed across an area that leverage both image data and spatial information regarding the area, to provide precise object tracking across the sensors. Further approaches relate to various architectures and configurations for smart area monitoring systems, as well as visualization and processing techniques. For example, as opposed to presenting video of an area captured by cameras, 3D renderings may be generated and played from metadata extracted from sensors around the area.

Abstract: Methods and apparatuses are described for system for identifying and mitigating high-risk database queries through ranked variance analysis. A server identifies database queries executed against databases in a production computing environment within a predetermined time period, each database query associated with execution plans and each execution plan having corresponding plan data elements. For each database query: the server generates execution variance data for the execution plans for a database query based upon the corresponding plan data elements, comprising: determining an execution time variance between the execution plans; and determining a buffer gets variance between the execution plans. The server ranks the database queries according to (i) the execution time variance, and (ii) the buffer gets variance.

Abstract: A tracking system comprising: a radio frequency (RF) signal circulator; an iridium modem coupled to first port of the circulator; an antenna coupled to second port of the circulator, wherein the circulator passes a signal transmitted by the modem to the antenna when the switch is switched to a first mode; a low noise amplifier (LNA) coupled to third port of the circulator, wherein the circulator passes RF signals received from the antenna to the LNA; a diplexer coupled to an output of the LNA; a GNSS receiver coupled to a first output of the diplexer through a GNSS filter; an iridium filter coupled to a second output of the diplexer; wherein the switch couples the iridium filter to the modem when the iridium modem is in a receiving mode, and wherein the switch couples the modem to the first port when the modem is in a transmitting mode.

Abstract: The invention relates to a bistable piezoelectric cantilever-based vibration energy harvester that increases the energy harvested over a broad frequency of vibrations by introducing bistability to the cantilever through use of two repelling magnets, one mounted on the cantilever and one at a location facing the first. This increases the amplitude and velocity of cantilever vibration, hence harvested power, while nonlinearity makes the system efficient over almost the entire range of frequencies, lower than the natural resonant frequency of the linear version without the magnets. Such improved performance is seen while the cantilever remains in the bistable mode.

Abstract: A soil sensor includes a housing, an antenna disposed within the housing, a measurement circuit disposed within the housing and operatively connected to the antenna, the circuit configured to measure impedance of soil at a plurality of different frequencies using the antenna as a sensor electrode, and a wireless interface disposed within the housing and operatively connected to the antenna and configured for wireless communications over the antenna at its communications frequency.

Abstract: According to one example of the present invention, a cooling system is provided. The cooling system comprises a water-cooled air conditioning module for cooling air from a chamber. The chamber is arranged for receiving air to be cooled and hot humid air. The cooling system also comprises a control system for controlling the flow of hot humid air to the chamber such that the air input to the air conditioning module has a predetermined relative humidity level within a predetermined range.

Abstract: Resource management for data centers is disclosed. In an exemplary embodiment, a method includes determining electrical power usage for the data center, and determining cooling fluid usage for the data center. The method also includes processing a resource utilization cap for the data center, and adjust ng at least one of the electrical power and the cooling fluid for the data center based on the resource utilization cap.

Abstract: A method for determining regions of influence of a plurality of fluid moving devices in an infrastructure is provided. In the method, a plurality of clusters are generated, with each of the plurality of clusters containing a representation of fluid supply data pertaining to a fluid moving device and one or more of the plurality of sensors based upon collected sensor and fluid supply data. In addition, regions of influence of each of the plurality of fluid moving devices are identified from an analysis of the generated clusters.