Abstract: A method for collecting contents performed by a content collection device is provided. The method may comprise acquiring a captured image, recognizing a first object included in the acquired captured image by inputting the captured image into an object recognition model, the first object being an advertisement type object, generating content classification information for the recognized first object, determining whether the first object has already been reported by using the content classification information and transmitting metadata for the first object to a server when it is determined that the first object is a new object that has not already been reported as a result of performing the determining of whether the first object has already been reported.

Abstract: A semiconductor device may include a backside wiring line in a first backside interlayer insulating film, a fin-type pattern on the backside wiring line, a second backside interlayer insulating film between the fin-type pattern and the first backside interlayer insulating film, a gate electrode on the fin-type pattern, a first source/drain pattern on a side of the gate electrode, and a backside source/drain contact in a backside contact hole defined by the fin-type pattern and the second backside interlayer insulating film. The backside source/drain contact may connect the backside wiring line and the first source/drain pattern. The backside source/drain contact may include an upper pattern and a lower pattern. The upper pattern may be between the lower pattern and the first source/drain pattern, and may fill at least a portion of the first backside contact hole. The upper pattern may have a single conductive film structure.

Abstract: A pixel of a display device may pre-charge a light emitting element in non-emission periods immediately before emission periods by varying a voltage level of an initialization voltage. As a result, a luminance non-uniformity phenomenon that may occur as the result of a deterioration deviation of the light emitting element may be removed or reduced.

Abstract: A neural network processing unit may be configured to perform an approximate multiplication operation and a system on chip may include the neural network processing unit. The neural network processing unit may include a plurality of neural processing units and may perform a computation based on one or more instances of input data and a plurality of weights. At least one neural processing unit is configured to receive a first value and a second value and perform an approximate multiplication operation based on the first value and the second value and is further configured to perform a stochastic rounding operation based on an output value of the approximate multiplication operation.

Abstract: The present disclosure relates to a method and a device generating a navigational route using an integrated electronic chart, which includes receiving a plurality of electronic charts with different scales; extracting at least one overlapping area of the plurality of electronic charts; integrating the plurality of electronic charts into the integrated electronic chart; dividing the integrated electronic chart into a configurable grid; generating an initial route using first object information in the configurable grid; mapping second object information detected by a sensing device onto the integrated electronic chart; and resetting the initial route using the second object information. According to the method and device of the present disclosure, the integrated electronic chart allows for efficient use of computing resources. Overall route calculation time and memory usage is reduced, and each area is accurately expressed.

Abstract: A method of producing a lube base oil mixture includes providing a waste lubricant-derived refined oil fraction derived from a waste lubricant containing a lube base oil of API Group I or II. The waste lubricant-derived refined oil fraction contains an ionic refined oil, a first regenerated base oil, or a combination thereof, dewaxing the waste lubricant-derived refined oil fraction to produce a second regenerated base oil, and blending the second regenerated base oil with a separate lube base oil to produce a lube base oil mixture of Group III or higher. The method has an economical advantage because low-quality waste lubricant is used as a feedstock for the production of higher quality lube base oils. In addition, the method is advantageous in an environmentally friendly aspect because waste lubricant is reused rather than being disposed of.

Abstract: A semiconductor device includes a first substrate having a first surface and a second opposite surface, a first lower interlayer insulating layer on the second surface, a first active pattern including a first lower pattern contacting the first surface, a plurality of first sheet patterns spaced apart from the first lower pattern in a second direction, a first gate structure on the first lower pattern, a first source/drain pattern on a side of the first gate structure, a second lower interlayer insulating layer including a third surface and a fourth opposite surface, a second active pattern including a second lower pattern contacting the third surface, a plurality of second sheet patterns spaced apart from the second lower pattern in the second direction, a second gate structure on the second lower pattern, wherein the first lower pattern has a first height, and the second lower pattern has a second different height.

Abstract: A method of recycling waste lubricant. The method includes providing a waste lubricant-derived oil fraction, pretreating the waste lubricant-derived oil fraction, and hydrocracking the pretreated waste lubricant-derived oil fraction. The method has economic advantages in that waste lubricant can be recycled into fuel oil and high-grade lube base oil with a relatively high yield compared to a lube base oil manufacturing process utilizing only unconverted oil as a feedstock.

Abstract: A method of recycling waste lubricant, the method including providing a waste lubricant-derived oil fraction, pretreating the waste lubricant-derived oil fraction, and hydrocracking the pretreated waste lubricant-derived oil fraction. The method recycles waste lubricant into fuel oil and high-grade lube base oil. According to the above method, high-quality lubricating base oil can be produced from waste lubricating oil, which is advantageous in economic terms.

Abstract: A sealing structure may include a lid including a first lid face, a second lid face opposite to the first lid face, and a fragile area between the first lid face and the second lid face, a cover including a first cover face facing the second lid face and covering the fragile area and a second cover face opposite to the first cover face, wherein a first distance between the second lid face and the first cover face is substantially equal to or less than a second distance between the first cover face and the second cover face, and a connector configured to connect the lid and the cover.

Abstract: An integrated circuit included in a device for performing a neural network operation includes a buffer configured to store feature map data in units of cells each including at least one feature, wherein the feature map data is for use in the neural network operation; and a multiplexing circuit configured to receive the feature map data from the buffer, and output extracted data by extracting feature data of one of features that are included within a plurality of cells in the received feature map data, the features each corresponding to an identical coordinate value.

Abstract: A redundancy dynamic braking system may include: a main braking force adjusting device configured to control a hydraulic braking device of a vehicle; a receiving unit configured to receive driving information of the vehicle; an electronic control braking device configured to be operated electrically to generate braking force; and an auxiliary braking force adjusting device configured to control the hydraulic braking device and the electronic control braking device, in an event of a failure in the main braking force adjusting device, wherein the auxiliary braking force adjusting device is configured to control the electronic control braking device by adjusting a magnitude and an application time of a current applied to the electronic control braking device based on an estimated target pressure using the driving information.

Abstract: A frequency multiplier is provided. A harmonic generator of the frequency multiplier comprises: a harmonic generating core unit; a first resonant tank which is connected to a first output terminal and a second output terminal of the harmonic generating core unit; and a first feedback circuit which is connected to the first output terminal and the second output terminal of the harmonic generating core unit to change the effective resistance of the first resonant tank.

Abstract: Provided is a hot-dip galvanized steel sheet used for an automobile outer panel and so forth and, more specifically, to a method for producing a hot-dip galvanized steel sheet having excellent press formability and clarity after painting, and to a hot-dip galvanized steel sheet produced by the method.

Abstract: An artificial neural network (ANN) system includes a processor, a virtual overflow detection circuit and a data format controller. The processor performs node operations with respect to a plurality of nodes included in each layer of an ANN to obtain a plurality of result values of the node operations and performs a quantization operation on the obtained plurality of result values based on a k-th fixed-point format for a current quantization of the each layer to obtain a plurality of quantization values. The virtual overflow detection circuit generates a virtual overflow information indicating a distribution of valid bit numbers of the obtained plurality of quantization values. The data format controller determines a (k+1)-th fixed-point format for a next quantization of the each layer based on the generated virtual overflow information. An overflow and/or an underflow are prevented efficiently by controlling the fixed-point format using the virtual overflow.

Abstract: An integrated circuit included in a device for performing a neural network operation includes a buffer configured to store feature map data in units of cells each including at least one feature, wherein the feature map data is for use in the neural network operation; and a multiplexing circuit configured to receive the feature map data from the buffer, and output extracted data by extracting feature data of one of features that are included within a plurality of cells in the received feature map data, the features each corresponding to an identical coordinate value.

Abstract: An integrated circuit included in a device for performing a neural network operation includes a buffer configured to store feature map data in units of cells each including at least one feature, wherein the feature map data is for use in the neural network operation; and a multiplexing circuit configured to receive the feature map data from the buffer, and output extracted data by extracting feature data of one of features that are included within a plurality of cells in the received feature map data, the features each corresponding to an identical coordinate value.

Abstract: An electrostatic chuck of the present invention includes a top block and a bottom block bonded by a bonding layer. The top block has a first plate on which a chucking electrode and a heater are installed, and the bottom block is provided with a cooling member. A second plate made of a material having lower heat transfer rate than the first plate is disposed between the first plate and the bottom block. Accordingly, when the heater is heated at a high temperature, it is possible to prevent the bonding layer from being damaged by thermal impact.

Abstract: Proposed is a substrate support unit, a method of manufacturing the same, and a substrate processing apparatus including the same. The substrate support unit includes a base component, a chucking component mounted on the base component, and an intermediate layer interposed between the base component and the chucking component, wherein the intermediate layer includes a joining part formed in a partial region to couple the base component and the chucking component.

Abstract: A neural processing system includes a first frontend module, a second frontend module, a first backend module, and a second backend module. The first frontend module executes a feature extraction operation using a first feature map and a first weight, and outputs a first operation result and a second operation result. The second frontend module executes the feature extraction operation using a second feature map and a second weight, and outputs a third operation result and a fourth operation result. The first backend module receives an input of the first operation result provided from the first frontend module and the fourth operation result provided from the second frontend module via a second bridge to sum up the first operation result and the fourth operation result.