Abstract: The present disclosure provides a method of training a video tag recommendation model, a method of determining a video tag, an electronic device, and a storage medium. The video tag recommendation model includes a video feature extraction network and a tag feature extraction network. The method of training the video tag recommendation model includes: obtaining a first video feature of a video sample by using the video feature extraction network; inputting a first tag as a positive sample to the tag feature extraction network, so as to obtain a first feature of the first tag; inputting a second tag as a negative sample to the tag feature extraction network, so as to obtain a second feature of the second tag; and training the video tag recommendation model according to the first video feature, the first feature, the second feature, and a predetermined loss function.

Abstract: Provided are an illegal building identification method and apparatus, a device, and a storage medium, which relate to the field of cloud computing. The specific implementation scheme is: acquiring a target image and a reference image associated with the target image; extracting a target building feature of the target image and a reference building feature of the reference image, respectively; and determining, according to the target building feature and the reference building feature, an illegal building identification result of the target image.

Abstract: The present application discloses a video type detection method, apparatus, electronic device and storage medium. A specific implementation solution is as follows: obtaining N key frames of a first video, where N is an integer greater than 1, and a type of the first video is to be detected; obtaining M confidence scores corresponding to each of the N key frames by inputting each of the N key frames into M algorithm models corresponding to the first video type respectively, where M is an integer greater than 1; determining a confidence score of the first video by a fusion strategy algorithm model according to N×M confidence scores of the N key frames; and comparing the confidence score of the first video with a confidence score threshold corresponding to a first video type, to determine whether the type of the first video is the first video type or not.

Abstract: The present disclosure provides a method for detecting illegal parking. The method includes: obtaining a detection frame having a category of vehicle from an image acquired by an acquisition module through a first neural network to determine coordinate information of the detection frame; determining whether an area corresponding to the detection frame in the image is an illegal parking area through a second neural network; and when determining that the area is the illegal parking area, determining whether the coordinate information of the detection frame is changed within a preset period of time to determine whether the vehicle is parked illegally.

Abstract: A method includes detecting a read error during a first read operation by a read sensor on a first track on a data storage medium. The method also includes, after the read error is detected, performing a second read operation by the read sensor on the first track, carrying out bit error value (BEV) measurements for the second read operation, and determining, from the BEV measurements, whether the read sensor produces a marginal BEV, which is between first and second predetermined BEV thresholds. In response to determining that the read sensor produces the marginal BEV, a third read operation by the read sensor on the first track is carried out, and BEV measurements for the third read operation are performed. When the BEV measurements from the second and third read operations meet at least one predetermined read sensor instability criterion, it is concluded that the read sensor is unstable.

Abstract: The present application discloses a video type detection method, apparatus, electronic device and storage medium. A specific implementation solution is as follows: obtaining N key frames of a first video, where N is an integer greater than 1, and a type of the first video is to be detected; obtaining M confidence scores corresponding to each of the N key frames by inputting each of the N key frames into M algorithm models corresponding to the first video type respectively, where M is an integer greater than 1; determining a confidence score of the first video by a fusion strategy algorithm model according to N×M confidence scores of the N key frames; and comparing the confidence score of the first video with a confidence score threshold corresponding to a first video type, to determine whether the type of the first video is the first video type or not.

Abstract: A method includes storing a superset of data on a data storage medium along with a corresponding superset superparity. The superset of data includes multiple sets of data, and the corresponding superset superparity is calculated based on all of the multiple sets of data. The method also includes updating at least one subset of the superset of data. The subset has a subset superparity. The superset superparity is updated with the subset superparity, and the subset superparity and a location of the subset within the superset are employed to carry out error correction operations.

Abstract: A method includes storing a superset of data on a data storage medium along with a corresponding superset superparity. The superset of data includes multiple sets of data, and the corresponding superset superparity is calculated based on all of the multiple sets of data. The method also includes updating at least one subset of the superset of data. The subset has a subset superparity. The superset superparity is updated with the subset superparity, and the subset superparity and a location of the subset within the superset are employed to carry out error correction operations.

Abstract: The present disclosure provides a method for detecting illegal parking. The method includes: obtaining a detection frame having a category of vehicle from an image acquired by an acquisition module through a first neural network to determine coordinate information of the detection frame; determining whether an area corresponding to the detection frame in the image is an illegal parking area through a second neural network; and when determining that the area is the illegal parking area, determining whether the coordinate information of the detection frame is changed within a preset period of time to determine whether the vehicle is parked illegally.

Abstract: A method is provided for processing a dual-energy CT scan image, which includes filtering the pixels in a dual-energy CT scan image to obtain pixels to be grouped; grouping the pixels to be grouped into a plurality of pixel groups based on the positions of the pixels to be grouped in the dual-energy CT scan image; performing material decomposition on the pixels in each pixel group; and determining the object corresponding to each pixel group based on the results of material decomposition. By using the method, the scan time and the X-ray dose radiated to a target (e.g., a user to be diagnosed) are reduced.

Abstract: A method is provided for processing a dual-energy CT scan image, which includes filtering the pixels in a dual-energy CT scan image to obtain pixels to be grouped; grouping the pixels to be grouped into a plurality of pixel groups based on the positions of the pixels to be grouped in the dual-energy CT scan image; performing material decomposition on the pixels in each pixel group; and determining the object corresponding to each pixel group based on the results of material decomposition. By using the method, the scan time and the X-ray dose radiated to a target (e.g., a user to be diagnosed) are reduced.

Abstract: A thin-film transistor with a fluorinated channel and fluorinated source and drain regions and methods of fabrication are provided. The thin-film transistor includes: a substrate; a semiconductor active layer of fluorine-doped metal-oxide formed on the substrate; fluorine-doped source and drain regions disposed adjacent to the semiconductor active layer; a gate electrode disposed over the semiconductor active layer, configured to induce a continuous conduction channel between the source and drain regions; and a gate dielectric material separating the gate electrode and the channel.

Abstract: Techniques for data recovery from a data block wherein portions of the data block are written at different off-track positions are described. For example, portions of a data block can be read separately for each on-track and off-track position in the data block; then, the separately read portions can be recombined into a recovered data block.

Abstract: A light-guide member includes a transparent light-guide body, a reflective layer and a shading layer. The light-guide body includes a light incident surface, a light-emitting surface and a peripheral surface connecting the light incident surface and the light-emitting surface. The peripheral surface includes at least one reflective surface for transmitting the light entering form the light incident surface out of the light-emitting surface. The reflective layer may be painted on the reflective surface. The shading layer may be painted on the reflective layer.

Abstract: A housing for an electronic device comprises a substrate, a base coat, and a transparent top coat. The substrate is a transparent or a translucent plastic substrate and has opposite surfaces. The base coat and the top coat are respectively formed on the opposite surfaces of the substrate.

Abstract: A thin-film transistor with a fluorinated channel and fluorinated source and drain regions and methods of fabrication are provided. The thin-film transistor includes: a substrate; a semiconductor active layer of fluorine-doped metal-oxide formed on the substrate; fluorine-doped source and drain regions disposed adjacent to the semiconductor active layer; a gate electrode disposed over the semiconductor active layer, configured to induce a continuous conduction channel between the source and drain regions; and a gate dielectric material separating the gate electrode and the channel.

Abstract: A housing for an electronic device comprises a substrate, a base coat, and a transparent top coat. The substrate is a transparent or a translucent plastic substrate and has opposite surfaces. The base coat and the top coat are respectively formed on the opposite surfaces of the substrate.

Abstract: Techniques for data recovery from a data block wherein portions of the data block are written at different off-track positions are described. For example, portions of a data block can be read separately for each on-track and off-track position in the data block; then, the separately read portions can be recombined into a recovered data block.

Abstract: A reactor containing a plurality of porous rotors is disclosed. A liquid sprayed in a direction from a rotating axis to the rotors is hit by the porous rotors, creating fine liquid particles, so that a mass transfer takes place between the surfaces of the liquid particles and a gas injected into the reactor, and thus a crystallization reaction is initiated. The crystallization reaction is confined by the amounts of the reactants contained in the liquid particles, and will stop as soon as one of the reactants is consumed, so that the resulting crystals can be controlled in nanometer level. The reactor of the present invention is also suitable for use in a liquid-liquid reaction, wherein two different liquids are separately sprayed in a direction from the rotating axis and hit by the porous rotors, and the resulting fine liquid particles of the two different liquids contact and react with each other, so that ultra fine particles are formed.

Abstract: A reactor containing a plurality of porous rotors is disclosed. A liquid sprayed in a direction from a rotating axis to the rotors is hit by the porous rotors, creating fine liquid particles, so that a mass transfer takes place between the surfaces of the liquid particles and a gas injected into the reactor, and thus a crystallization reaction is initiated. The crystallization reaction is confined by the amounts of the reactants contained in the liquid particles, and will stop as soon as one of the reactants is consumed, so that the resulting crystals can be controlled in nanometer level. The reactor of the present invention is also suitable for use in a liquid-liquid reaction, wherein two different liquids are separately sprayed in a direction from the rotating axis and hit by the porous rotors, and the resulting fine liquid particles of the two different liquids contact and react with each other, so that ultra fine particles are formed.