Abstract: This invention provides a structure for applying on a steam cleaner to control the amount of steam through PWM, including: a casing, one side of which is fixedly installed with a power line, the output end of which is electrically connected to a power switch, the output end of which is electrically connected to a control board, and the output end of which is electrically connected to a pump, steam generator, and display screen; a water tank, which is fixedly installed at the bottom of the inner side of the casing.

Abstract: A method of detecting and positioning an insulation failure of an aircraft generator is provided, comprising: S1, for each stator slot in the aircraft generator, wrapping one semiconductor layer on a surface of an external insulation layer of each copper conductor in the stator slot; S2, for each semiconductor layer, partitioning the semiconductor layer to obtain a plurality of zones, disposing a resistor-inductance branch network in each zone, and connecting the resistor-inductance branch networks to form an insulation failure detection circuit; S3, supplying power to a three-phase winding of the aircraft generator from an external power source; S4, determining whether an unearthed voltage is present in four end corners of the insulation failure detection circuit; if not, returning to S3 for further powering; if yes, outputting a detection result and collecting an insulation failure position with potential change and difference in the insulation failure detection circuit as a positioning result.

Abstract: An image processing system generates an image mask from an image. The image is processed by an object detector to identify a region having an object, and the region is classified based on an object type of the object. A masking pipeline is selected from a number of masking pipelines based on the classification of the region. The region is processed using the masking pipeline to generate a region mask. An image mask for the image is generated using the region mask.

Abstract: A roller embossing method for flexible graphite polar plates of fuel cells comprises: (1) adjusting a clearance of an embossing roller pair of a roller press to a target thickness value of a monopolar plate; (2) feeding a flexible graphite slab in front of the embossing roller pair, and entangling the flexible graphite slab by means of opposite rotation of the embossing roller pair; (3) after the slab is entangled, continuing to forward roll the slab for 10-100 mm, and then reversely rolling the slab for 5-90 mm by means of synchronous and opposite rotation of an upper and a lower embossing roller, wherein one time of forward rolling and one time of reverse rolling are referred to as reciprocal rolling; (4) forming a polar plate after several times of reciprocal rolling, and separating the polar plate from the embossing rollers; (5) performing subsequent treatment on the roll-formed flexible graphite polar plate.

Abstract: Various disclosed embodiments are directed to deriving an albedo output image from an input image based on deriving an inverse shading map. For example, an input image can be a photograph of a human face (i.e., the geometric features) with RGB values representing the color values of the face as well as pixels representing shadows (i.e., the shadow features) underneath the chin of the human face. The inverse shading map may be a black and white pixel value image that contains pixels representing the same human face without the RGB values and the shadows underneath the chin. The inverse shading map thus relies on the geometric space, rather than RGB space. Geometric space, for example, allows embodiments to capture the geometric features of a face, as opposed to those geometric features' RGB or shadow details.

Abstract: This disclosure describes one or more implementations of an alpha matting system that utilizes a deep learning model to generate alpha mattes for digital images utilizing an alpha-range classifier function. More specifically, in various implementations, the alpha matting system builds and utilizes an object mask neural network having a decoder that includes an alpha-range classifier to determine classification probabilities for pixels of a digital image with respect to multiple alpha-range classifications. In addition, the alpha matting system can utilize a refinement model to generate the alpha matte from the pixel classification probabilities with respect to the multiple alpha-range classifications.

Abstract: A method comprising: disposing a bottom hole assembly (BHA) comprising a Rotary Steerable System (RSS) into a borehole, obtaining bending of the drill bit (BOB) data with a measurement assembly disposed on the BHA, classifying the BOB data into a neutral RSS configuration or a geostationary RSS configuration, determining t1 based at least on a neutral RSS configuration, wherein t1 is the total time during a neutral RSS configuration, determining t2 based at least on a neutral RSS configuration, wherein t2 is the total time during a geostationary RSS configuration, and determining a real time Duty Cycle based at least on t1 and t2. Additionally, a bottom hole assembly (BHA) comprising a Rotary Steerable System (RSS) disposed into a borehole, and a steering control system a configured to perform the method described above.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods that implement a multi-branch harmonization neural network architecture to harmonize composite images. For example, in one or more implementations, the semantic-guided transformer-based harmonization system uses a convolutional branch, a transformer branch, and a semantic branch to generate a harmonized composite image based on an input composite image and a corresponding segmentation mask. More particularly, the convolutional branch comprises a series of convolutional neural network layers followed by a style normalization layer to extract localized information from the input composite image. Further, the transformer branch comprises a series of transformer neural network layers to extract global information based on different resolutions of the input composite image. The semantic branch includes a visual neural network that generates semantic features that inform the harmonization of the composite images.

Abstract: A flatness control structure for a drywall sander, comprising: a device body, wherein the bottom of the device body is equipped with a rotating device, a housing is arranged below the device body, a sanding component is arranged inside the housing, the sanding component comprises a sanding piece, the bottom of sanding piece is connected to a sanding disc, and the bottom of sanding disc is connected to sandpaper, a threaded bolt is arranged between the sanding piece and the housing, a fixing component is arranged inside the housing, and an installation component is arranged inside the housing. The sander replaces the bristle sections with a wear-resistant strip which is less likely to deform when in contact with the sanding surface. When force is applied, the sander can be firmly fixed on the sanding surface, ensuring stability of the entire sander, thereby improving the flatness of the sanding surface.

Abstract: Disclosed is a method for monitoring vehicle overload based on gravity anomaly, which includes the following steps of: setting a plurality of measuring positions on a single lane, arranging a gravimeter to acquire gravity anomaly values caused by a vehicle when the middle position along the length direction of the vehicle reaching each measuring position, using a monitoring camera to judge the category of the vehicle, acquire three geometric dimensions of the vehicle, and determine the position of the vehicle. The vehicle is simplified as a cuboid, and the mass density distribution is simplified as a piecewise constant function along the length direction of the vehicle. Calculating values of the piecewise constant function according to the gravity anomaly values, calculating a total weight of the vehicle according to the mass density distribution, comparing the total weight with the weight limit of the vehicle to judge whether the vehicle is overloaded.

Abstract: A mechanical lock password setting and reset structure, including: a face cover, the face cover with a skeleton, a positioning spring between the skeleton, the side of the skeleton, the reset plate with a side cover, a spring and a cam between the reset plate and the lock plate, the back of the bottom cover of the pressure plate, the rubber spring, a key cover, the key and the key, respectively. The present invention, the present invention provides a mechanical lock password setting and reset structure, this design can replace the now the traditional mechanical password lock internal complex structure, can improve the convenience of installation, and prevent the loss of separate parts during long-term use.

Abstract: A twist hand replacement lock core structure, including: a face cover, the face cover is provided with a lock core structure, the lock core structure is provided inside the rotating member, the turn member side is connected to a twist hand, the turn member and the twist hand are provided with a hand screw. The invention provides a twisted hand replacement lock core structure, The hand screw can lock the rotating parts and twist hand, When the lock core is damaged or the lock core needs to be replaced, The lock core structure can be removed and replaced by a wrench, This can realize the disassembly and replacement of the lock core structure from the handle part, Can adapt to the key to match the lock core structure, And only if entering the correct password, After turning the handle to reach the specified position.

Abstract: A control method for intelligent equipment includes: determining a positional relationship between a control gesture and various intelligent equipment according to a depth image containing the control gesture and positional information of various intelligent equipment, wherein the control gesture is a gesture made by a user to control various intelligent equipment; and starting an equipment control operation on target intelligent equipment based on the control gesture in response to determining that a positional relationship between the target intelligent equipment and the control gesture meets a preset condition.

Abstract: An acrylate-based single-chain nanoparticle is disclosed, which includes 10-20 parts of methacrylic acid, 1-2 parts of tetramethylguanidine and 1-2 parts of halohydrocarbon. The scheme provided by the disclosure establishes a “meta template” for synthesizing various single-chain nanoparticles, unifies the preparation method of the polyacrylate, simultaneously carries out high-concentration preparation in an aqueous phase only through ultraviolet irradiation, which reduces the production cost, reduces the carbon emission in production, and provides possibility for the industrialized production of the polyacrylate-based single-chain nanoparticle.

Abstract: Methods and apparatus for etching high aspect ratio features in substrates having mixed material stacks are provided herein. Methods involve using low plasma power, high chamber pressure, and/or low temperature while exposing the substrate to a metal-containing additive gas during etching using a fluorocarbon gas.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer-readable media that generate a height map for a digital object portrayed in a digital image and further utilizes the height map to generate a shadow for the digital object. Indeed, in one or more embodiments, the disclosed systems generate (e.g., utilizing a neural network) a height map that indicates the pixels heights for pixels of a digital object portrayed in a digital image. The disclosed systems utilize the pixel heights, along with lighting information for the digital image, to determine how the pixels of the digital image project to create a shadow for the digital object. Further, in some implementations, the disclosed systems utilize the determined shadow projections to generate (e.g., utilizing another neural network) a soft shadow for the digital object. Accordingly, in some cases, the disclosed systems modify the digital image to include the shadow.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods that implement a dual-branched neural network architecture to harmonize composite images. For example, in one or more implementations, the transformer-based harmonization system uses a convolutional branch and a transformer branch to generate a harmonized composite image based on an input composite image and a corresponding segmentation mask. More particularly, the convolutional branch comprises a series of convolutional neural network layers followed by a style normalization layer to extract localized information from the input composite image. Further, the transformer branch comprises a series of transformer neural network layers to extract global information based on different resolutions of the input composite image.

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for utilizing a plurality of neural networks in a multi-branch pipeline to generate image masks for digital images. Specifically, the disclosed system can classify a digital image as a portrait or a non-portrait image. Based on classifying a portrait image, the disclosed system can utilize separate neural networks to generate a first mask portion for a portion of the digital image including a defined boundary region and a second mask portion for a portion of the digital image including a blended boundary region. The disclosed system can generate the mask portion for the blended boundary region by utilizing a trimap generation neural network to automatically generate a trimap segmentation including the blended boundary region. The disclosed system can then merge the first mask portion and the second mask portion to generate an image mask for the digital image.

Abstract: In accordance with the described techniques, a portrait relighting system receives user input defining one or more markings drawn on a portrait image. Using one or more machine learning models, the portrait relighting system generates an albedo representation of the portrait image by removing lighting effects from the portrait image. Further, the portrait relighting system generates a shading map of the portrait image using the one or more machine learning models by designating the one or more markings as a lighting condition, and applying the lighting condition to a geometric representation of the portrait image. The one or more machine learning models are further employed to generate a relit portrait image based on the albedo representation and the shading map.

Abstract: In accordance with the described techniques, an image delighting system receives an input image depicting a human subject that includes lighting effects. The image delighting system further generates a segmentation mask and a skin tone mask. The segmentation mask includes multiple segments each representing a different portion of the human subject, and the skin tone mask identifies one or more color values for a skin region of the human subject. Using a machine learning lighting removal network, the image delighting system generates an unlit image by removing the lighting effects from the input image based on the segmentation mask and the skin tone mask.