Abstract: The invention discloses a mobile flash butt welding method for 136RE+SS heat-treated rail, and particularly a mobile flash butt welding method for 136RE+SS heat-treated rail in the technical field of rail welding. The mobile flash butt welding method for 136RE+SS heat-treated rail in the invention includes a pre-flash stage, a flash stage, a boost stage, an upset stage and a forge stage, with a total heat input of 7.1 MJ-10.0 MJ, a total duration of 110 s-135 s and an upsetting distance of 12.8 mm-16.7 mm during the welding process. By adopting the method of the invention, mobile flash butt welding can be conducted for 136RE+SS heat-treated rail successfully, and the rail joint has less internal defects but stable welding quality, and can pass fatigue test, tensile test and slow bend test to meet the requirements. Besides, the rail joint can pass the drop weight test for 15 welds continuously, demonstrating better stability.

Abstract: Methods and systems are provided for generating harmonized images for input composite images. A neural network system can be trained, where the training includes training a neural network that generates harmonized images for input composite images. This training is performed based on a comparison of a training harmonized image and a reference image, where the reference image is modified to generate a training input composite image used to generate the training harmonized image. In addition, a mask of a region can be input to limit the area of the input image that is to be modified. Such a trained neural network system can be used to input a composite image and mask pair for which the trained system will output a harmonized image.

Abstract: A method for purifying a crude 2,5-furandicarboxylic acid composition including 2,5-furandicarboxylic acid and 5-formylfuran-2-carboxylic acid is disclosed. The method includes the steps of: (a) subjecting the crude 2,5-furandicarboxylic acid composition to complete dissolution in a solvent solution to obtain a mixture, the solvent solution including an organic solvent and water; (b) subjecting 5-formylfuran-2-carboxylic acid in the mixture to an addition reaction with sodium hydrogen sulfite to obtain an addition product; and (c) subjecting 2,5-furandicarboxylic acid to precipitation after step (b) to obtain purified 2,5-furandicarboxylic acid.

Abstract: A communications system which includes a first data center and a second data center, where the first data center includes a first message proxy, a first routing proxy, a service server, and a storage server and configured to receive a first message from an application running on the service server, add routing information to the first message to generate a second message when the first message is processed by the second data center, and store the second message in a public area of the storage server, where the public area is configured to store a message processed by the second data center. The first routing proxy is configured to obtain the second message from the public area, and send the second message to the second data center based on the routing information.

Abstract: The present disclosure relates to training and utilizing an image exposure transformation network to generate a long-exposure image from a single short-exposure image (e.g., still image). In various embodiments, the image exposure transformation network is trained using adversarial learning, long-exposure ground truth images, and a multi-term loss function. In some embodiments, the image exposure transformation network includes an optical flow prediction network and/or an appearance guided attention network. Trained embodiments of the image exposure transformation network generate realistic long-exposure images from single short- exposure images without additional information.

Abstract: Certain embodiments involve using an image completion neural network to perform user-guided image completion. For example, an image editing application accesses an input image having a completion region to be replaced with new image content. The image editing application also receives a guidance input that is applied to a portion of a completion region. The image editing application provides the input image and the guidance input to an image completion neural network that is trained to perform image-completion operations using guidance input. The image editing application produces a modified image by replacing the completion region of the input image with the new image content generated with the image completion network. The image editing application outputs the modified image having the new image content.

Abstract: A collaborative document editing system includes a server and at least two user terminals communicatively coupled to the server through a network. The server distributes a collaborative editing task to the two user terminals. The collaborative editing task includes a document to be edited. The two user terminals send local operation records to the server. The server records the local operation records of the user terminals. The server receives an edit operation command from the two user terminals. The server adds the edit operation commands of the user terminals to an operation queue and sends the edit operation commands to the user terminals according to the operation queue. The user terminals receive the edit operation commands according to the operation queue and update the document according to the local operation record corresponding to the edit operation commands.

Abstract: The invention provides a process of isolating or enriching a heavy metal present in a liquid medium. The process comprising the following steps: (a) incubating a suspension containing (i) particulate scrap metal, household waste and/or industrial waste containing a heavy metal in elemental form and (ii) biomass comprising a bacterium, or a combination of two or more bacteria, capable of binding the heavy metal; (b) separating the biomass having bound heavy metal from the suspension of step (a); and (c) isolating the heavy metal from the biomass separated in step (b).

Abstract: A method for transmitting an application programming interface API request includes receiving, by a first API gateway, a first API request; obtaining, by the first API gateway, a first forwarding label corresponding to the first API request, where the first forwarding label includes a first target security domain identifier, and a security domain identifier of the first API gateway is different from the first target security domain identifier. The method also includes determining an address of a second API gateway according to a mapping relationship between the first target security domain identifier and the address of the second API gateway. The method additionally includes sending the first API request to the second API gateway based on the address which is a next-hop API gateway of the first API gateway that sends the first API request to an API gateway corresponding to the first target security domain identifier.

Abstract: Digital image completion by learning generation and patch matching jointly is described. Initially, a digital image having at least one hole is received. This holey digital image is provided as input to an image completer formed with a dual-stage framework that combines a coarse image neural network and an image refinement network. The coarse image neural network generates a coarse prediction of imagery for filling the holes of the holey digital image. The image refinement network receives the coarse prediction as input, refines the coarse prediction, and outputs a filled digital image having refined imagery that fills these holes. The image refinement network generates refined imagery using a patch matching technique, which includes leveraging information corresponding to patches of known pixels for filtering patches generated based on the coarse prediction. Based on this, the image completer outputs the filled digital image with the refined imagery.

Abstract: A process includes forming one or more apertures on a component backside, creating a vacuum in a mold chase, and engaging the component backside with a mold compound in the mold chase. The one or more apertures form an aperture structure. The aperture structure may include multiple apertures parallel or orthogonal to each other. The apertures have an aperture width, aperture depth, and aperture pitch. These characteristics may be altered to minimize the likelihood of trapped air remaining after creating the vacuum in the mold chase.

Abstract: In some embodiments, an image manipulation application receives an incomplete image that includes a hole area lacking image content. The image manipulation application applies a contour detection operation to the incomplete image to detect an incomplete contour of a foreground object in the incomplete image. The hole area prevents the contour detection operation from detecting a completed contour of the foreground object. The image manipulation application further applies a contour completion model to the incomplete contour and the incomplete image to generate the completed contour for the foreground object. Based on the completed contour and the incomplete image, the image manipulation application generates image content for the hole area to generate a completed image.

Abstract: An offline mode user account authorization method includes obtaining an access privilege of a user account and setting an offline task access privilege of the user account according to the access privilege, retrieving user account information corresponding to a task to be executed from a preset task information comparison table when the task to be executed is implemented, determining whether the user account is online, executing the task to be executed with a user ID of the user account if the user account is online, and creating a super offline user account and assigning offline task access privilege of the user account to the super offline user account if the user account is not online. The implemented task to be executed is executed with the super offline user account.

Abstract: Techniques are disclosed for a facial expression classification. In an embodiment, a multi-class classifier is trained using labelled training images, each training image including a facial expression. The trained classifier is then used to predict expressions for unlabelled video frames, whereby each frame is effectively labelled with a predicted expression. In addition, each predicted expression can be associated with a confidence score. Anchor frames can then be identified in the labelled video frames, based on the confidence scores of those frames (anchor frames are frames having a confidence score above an established threshold). Then, for each labelled video frame between two anchor frames, the predicted expression is refined or otherwise updated using interpolation, thereby providing a set of video frames having calibrated expression labels.

Abstract: The present disclosure relates to training and utilizing an image exposure transformation network to generate a long-exposure image from a single short-exposure image (e.g., still image). In various embodiments, the image exposure transformation network is trained using adversarial learning, long-exposure ground truth images, and a multi-term loss function. In some embodiments, the image exposure transformation network includes an optical flow prediction network and/or an appearance guided attention network. Trained embodiments of the image exposure transformation network generate realistic long-exposure images from single short-exposure images without additional information.

Abstract: Digital image completion by learning generation and patch matching jointly is described. Initially, a digital image having at least one hole is received. This holey digital image is provided as input to an image completer formed with a dual-stage framework that combines a coarse image neural network and an image refinement network. The coarse image neural network generates a coarse prediction of imagery for filling the holes of the holey digital image. The image refinement network receives the coarse prediction as input, refines the coarse prediction, and outputs a filled digital image having refined imagery that fills these holes. The image refinement network generates refined imagery using a patch matching technique, which includes leveraging information corresponding to patches of known pixels for filtering patches generated based on the coarse prediction. Based on this, the image completer outputs the filled digital image with the refined imagery.

Abstract: The invention relates to a dispersion containing at least one rhamnolipid derivative and also to the use of the rhamnolipid derivatives as emulsifier or dispersing aid.

Abstract: Predicting patch displacement maps using a neural network is described. Initially, a digital image on which an image editing operation is to be performed is provided as input to a patch matcher having an offset prediction neural network. From this image and based on the image editing operation for which this network is trained, the offset prediction neural network generates an offset prediction formed as a displacement map, which has offset vectors that represent a displacement of pixels of the digital image to different locations for performing the image editing operation. Pixel values of the digital image are copied to the image pixels affected by the operation.

Abstract: Digital image completion using deep learning is described. Initially, a digital image having at least one hole is received. This holey digital image is provided as input to an image completer formed with a framework that combines generative and discriminative neural networks based on learning architecture of the generative adversarial networks. From the holey digital image, the generative neural network generates a filled digital image having hole-filling content in place of holes. The discriminative neural networks detect whether the filled digital image and the hole-filling digital content correspond to or include computer-generated content or are photo-realistic. The generating and detecting are iteratively continued until the discriminative neural networks fail to detect computer-generated content for the filled digital image and hole-filling content or until detection surpasses a threshold difficulty.

Abstract: A range finder includes an image capturing module, a transceiver module, a switching module, an output device and a processing unit. The image capturing module is configured to receive a first light beam emitted by an object. The transceiver module is configured to emit a second light beam to the object and receive the second light beam reflected by the object for obtaining a distance between the range finder and the object. The switching module is configured to output an electrical signal corresponding to a selected operating mode selected from a plurality of operating modes. The output device is configured to show the selected operating mode corresponding to the electrical signal. The processing unit is connected to the image capturing module, the transceiver module and the output device for receiving the electrical signal and controlling the image capturing module, the transceiver module and the output device.