Abstract: An exemplary embodiment provides an intruder detection method capable of accurately detecting an intruder and estimating an abnormal behavior of the intruder even when viewpoints of acquired images are different from each other. An intruder detection method is suitable for being performed by an intruder detection device for detecting an intruder based on images and includes: receiving input images acquired by multiple cameras; extracting feature maps associated with a plurality of viewpoints by applying the input images to a plurality of convolutional neural networks provided separately for the plurality of viewpoints of the images; and detecting the intruder based on the feature maps associated with the plurality of viewpoints.

Abstract: The present disclosure discloses a positive electrode active material for a lithium secondary battery comprising a secondary particle having an average particle size (D50) of 1 to 15 ?m, formed by agglomeration of at least two primary macro particles having an average particle size (D50) of 0.1 to 3 ?m; and a coating layer of a lithium-metal oxide on a surface of the secondary particle, wherein the primary macro particles are represented by LiaNi1-x-yCoxM1yM2wO2 (1.0?a?1.5, 0?x?0.2, 0?y?0.2, 0?w?0.1, 0?x+y?0.2, M1 includes at least one metal of Mn or Al, and M2 includes at least one metal selected from the group consisting of Ba, Ca, Zr, Ti, Mg, Ta, Nb and Mo), and wherein the lithium-metal oxide is a low-temperature phase LixCoO2(0<x?1) having at least one of a spinel structure (Fd-3m) or a disordered rock-salt structure (Fm-3m).

Abstract: Provided is a method and apparatus for detecting an unmanned aerial vehicle (UAV). The method of detecting a UAV includes receiving images from one or more cameras positioned in a UAV protected area; analyzing the received image to extract a UAV from the received image; and providing data related to the extracted UAV to train a UAV detection model, wherein photographing times or photographing positions of the one or more cameras are variably controlled.

Abstract: In order to detect an unmanned aerial vehicle (UAV) in a monitoring area using a rotatable camera, an integrated sensor configures a first detection area in the monitoring area by rotating the camera after fixing the focal length of the camera to an initial value, and configures a second detection area in the monitoring area by rotating the camera after changing at least one of the focal length of the camera and a rotation path representing a distance from the origin of the monitoring area.

Abstract: An intruder location-based dynamic virtual fence configuration and a multiple image sensor device operation method are provided. A method of detecting an intruder based on a virtual fence includes detecting an intruder invading a virtual fence set in a protection area by using a first image sensor device, and tracking the intruder through cooperation with second image sensor devices adjacent to the first image sensor device, based on information of the intruder.

Abstract: Provided is an apparatus for recognizing an object that includes an object inference module configured to process an original image captured by a camera module and generate an image of a size to be input to a machine learning inference model, wherein the object inference module includes the machine learning inference model, and outputs a result of recognition and classification of an object inferred through the machine learning inference model, and the machine learning inference model processes an input image to infer an object included in the input image.

Abstract: Provided are a method and system for detecting a UAV using a plurality of image sensors. A method of detecting a UAV includes detecting, by each of a plurality of detection image sensors, a UAV in a UAV detection area, transmitting, when the UAV is detected, position information of the detection image sensor detecting the UAV and distance information of the UAV to a classification image sensor, acquiring, by the classification image sensor, a magnified image of the UAV by setting a parameter of a camera of the classification image sensor according to the position information and the distance information, and classifying a type of the UAV by analyzing the magnified image.

Abstract: Provided is an artificial intelligence (AI)-based peer-to-peer (P2P) power trading method and apparatus that encourages a household with relatively great power consumption or a household with relatively small power consumption according to a power load pattern for each time period to participate in power trading by optimizing power consumption through AI-based P2P power trading in a cluster including a nanogrid.

Abstract: A method and device for detecting an illegal unmanned aerial vehicle (UAV) using a radio wave wall are provided. The method includes generating a radio wave wall between a plurality of reconnaissance UAVs that include a first reconnaissance UAV and second reconnaissance UAVs, using one or more wireless signals transmitted and received between the plurality of reconnaissance UAVs, and determining whether an illegal UAV enters the radio wave wall based on radio signal strengths of wireless signals received from the second reconnaissance UAVs.

Abstract: A reconnaissance UAV and a surveillance flight method of the reconnaissance UAV which can reduce a number of required reconnaissance UAVs are provided. The surveillance flight method for detecting an unauthorized UAV includes: defining multiple flight passages in a UAV no-fly zone for a plurality of reconnaissance UAVs; controlling each of the plurality of reconnaissance UAVs to fly across adjacent flight passages at an interval by crossing a border of the adjacent flight passages; and controlling at least one of an image sensor and a noise sensor mounted on the reconnaissance UAV to dynamically change a direction detection thereof.

Abstract: In order to detect an unmanned aerial vehicle (UAV) in a monitoring area using a rotatable camera, an integrated sensor configures a first detection area in the monitoring area by rotating the camera after fixing the focal length of the camera to an initial value, and configures a second detection area in the monitoring area by rotating the camera after changing at least one of the focal length of the camera and a rotation path representing a distance from the origin of the monitoring area.

Abstract: Provided is a method and apparatus for detecting an unmanned aerial vehicle (UAV). The method of detecting a UAV includes receiving images from one or more cameras positioned in a UAV protected area; analyzing the received image to extract a UAV from the received image; and providing data related to the extracted UAV to train a UAV detection model, wherein photographing times or photographing positions of the one or more cameras are variably controlled.

Abstract: An energy management method for a residential area, the method comprising, in response to an energy management request from a user who lives in the residential area, collecting environment information and energy information on a household of the user, selecting an imitation target among households which have environment information and energy information similar to the environment information and the energy information on the household of the user from among all households in the residential area, setting a method for applying the energy information on the imitation target, and controlling energy management of the household of the user according to the set method.

Abstract: According to a sensor network system and a method for processing sensor data, a sink node collects sensor data from sensor nodes, performs a prediction algorithm which is the same as a prediction algorithm performed in the sensor node to obtain a prediction value, when a sensing value is received from the sensor node, updates the prediction value to the sensing value, stores the sensing value as sensor data, determines whether the sensor node is a representative node when the sensing value is not received from the sensor node, and stores the prediction value as sensor data if the sensor node is the representative node, and corrects the prediction value based on the sensing value of a representative node and stores the corrected prediction value as sensor data if the sensor node is not the representative node.

Abstract: Provided herein is an apparatus for data processing, the apparatus including a receiving module configured to receive first data; a data processing module configured to process the first data received from the receiving module, to generate second data including meta data of the first data, and third data including the first data with its data type changed to an array type; and a mapping module configured to change a system of the first data and second data received from the data processing module to generate fourth data and fifth data, and to generate sixth data including mapping data of between the first data and fourth data and between the second data and fifth data.

Abstract: A nickel-doped Mg nanocrystals encapsulated by molecular-sieving reduced graphene oxide (rGO) layers is disclosed. Dual-channel doping, which combines external (rGO strain) and internal (Ni doping) mechanisms, efficiently promotes both hydriding and dehydriding processes of Mg nanocrystals, simultaneously improving both the kinetic and thermodynamic properties of the material. The composite achieves both high hydrogen storage capacity and excellent kinetics while maintaining robustness. The realization of three complementary functional components in one material-environmentally friendly and earth-abundant Mg for storage, Ni dopants for catalysis, and rGO layers for encapsulation-breaks new ground in metal hydrides and makes solid-state materials viable candidates for hydrogen-fueled applications.

Abstract: A high energy efficiency sensor node and an operating method thereof are provided. The high energy efficiency sensor node may include a sensing unit to generate sensed information and to store the sensed information in a database when a set period commences, and a control unit to obtain n pieces of sensed information corresponding to n periods from the database, and to transmit the n pieces of sensed information obtained to a first neighbor node when the n periods elapse.

Abstract: Provided herein is a sensing management system capable of minimizing the amount of data transmission from sensors, the system including a plurality of sensors configured to sense information of surrounding environment, and, in response to a sensing value transmitted to a controller having a slope of a same directivity for twice or more times, to obtain a predicted slope using a current slope of the transmitted sensing values; and the controller configured to obtain the predicted slope using a same algorithm as the sensors, wherein, in response to a next sensing value being within an error range of the predicted slope, the sensors do not transmit the next sensing value to the controller.

Abstract: A device and method for controlling lighting based on an illuminance model. The method for controlling lighting based on an illuminance model includes modeling actual illuminance measured by one or more illuminance measurement devices; generating a lighting profile based on the illuminance model; and controlling one or more lighting devices for each lighting scene according to the generated lighting profile.

Abstract: A device and method for controlling lighting based on an illuminance model. The method for controlling lighting based on an illuminance model includes modeling actual illuminance measured by one or more illuminance measurement devices; generating a lighting profile based on the illuminance model; and controlling one or more lighting devices for each lighting scene according to the generated lighting profile.