Abstract: The present invention provides a method and a device for image rectification, which is applied to the field of image processing. The method includes: receiving two images, wherein the two images are images of a target object captured at different viewpoints, and an image capturing device configured to capture the two images, includes a first image capturing device and a second image capturing device; determining positions of corresponding feature point pairs in the two images; and aligning the two images according to the positions of the feature point pairs and parameters of the first image capturing device and the second image capturing device. The above-mentioned method can rectify the images captured at different viewpoints, thereby ensuring the alignment of pixels, and avoiding visual discomfort of the viewer.

Abstract: The present invention provides a monocular unsupervised depth estimation method based on contextual attention mechanism, belonging to the technical field of image processing and computer vision. The invention adopts a depth estimation method based on a hybrid geometric enhancement loss function and a context attention mechanism, and adopts a depth estimation sub-network, an edge sub-network and a camera pose estimation sub-network based on convolutional neural network to obtain high-quality depth maps. The present invention uses convolutional neural network to obtain the corresponding high-quality depth map from the monocular image sequences in an end-to-end manner. The system is easy to construct, the program framework is easy to implement, and the algorithm runs fast; the method uses an unsupervised method to solve the depth information, avoiding the problem that ground-truth data is difficult to obtain in the supervised method.

Abstract: The invention discloses a depth estimation and color correction method for monocular underwater images based on deep neural network, which belongs to the field of image processing and computer vision. The framework consists of two parts: style transfer subnetwork and task subnetwork. The style transfer subnetwork is constructed based on generative adversarial network, which is used to transfer the apparent information of underwater images to land images and obtain abundant and effective synthetic labeled data, while the task subnetwork combines the underwater depth estimation and color correction tasks with the stack network structure, carries out collaborative learning to improve their respective accuracies, and reduces the gap between the synthetic underwater image and the real underwater image through the domain adaptation strategy, so as to improve the network's ability to process the real underwater image.

Abstract: Embodiments of this application provide a text recommendation method performed at an electronic device. The method includes: extracting feature content of from the a target text; processing the feature content by using at least two text analysis models to obtain at least two semantic vectors; integrating the at least two semantic vectors into an integrated semantic vector of the target text; selecting, according to the integrated semantic vector and an integrated semantic vector of at least one to-be-recommended text, a recommended text corresponding to the target text from the at least one to-be-recommended text. Because the integrated semantic vector of the target text is obtained based on the at least two text analysis models, the integrated semantic vector has a stronger representing capability. When text recommendation is subsequently performed, an association degree between the recommended text and the target text can be increased, thereby improving recommendation accuracy.

Abstract: Provided is a method based on deep neural network to extract appearance and geometry features for pulmonary textures classification, which belongs to the technical fields of medical image processing and computer vision. Taking 217 pulmonary computed tomography images as original data, several groups of datasets are generated through a preprocessing procedure. Each group includes a CT image patch, a corresponding image patch containing geometry information and a ground-truth label. A dual-branch residual network is constructed, including two branches separately takes CT image patches and corresponding image patches containing geometry information as input. Appearance and geometry information of pulmonary textures are learnt by the dual-branch residual network, and then they are fused to achieve high accuracy for pulmonary texture classification. Besides, the proposed network architecture is clear, easy to be constructed and implemented.

Abstract: The present invention discloses a grid map obstacle detection method fusing probability and height information, and belongs to the field of image processing and computer vision. A high-performance computing platform is constructed by using a GPU, and a high-performance solving algorithm is constructed to obtain obstacle information in a map. The system is easy to construct, the program is simple, and is easy to implement. The positions of obstacles are acquired in a multi-layer grid map by fusing probability and height information, so the robustness is high and the precision is high.

Abstract: Certain aspects of the present disclosure are generally directed to circuitry and techniques for current sensing. For example, certain aspects provide a circuit for signal amplification including a first amplifier, a second amplifier, and a third amplifier. The circuit also includes a first capacitive element coupled between a first output of the first amplifier and a first input of the third amplifier, a second capacitive element coupled between a second output of the first amplifier and a second input of the third amplifier, a third capacitive element coupled between a first output of the second amplifier and the first input of the third amplifier, and a fourth capacitive element coupled between a second output of the second amplifier and the second input of the third amplifier.

Abstract: The column-base joint includes a prefabricated-reinforced-concrete column and a reinforced-concrete foundation, and further includes a coating-steel-plate sleeve that is integrally prefabricated with the concrete column and encloses an exterior of a column base. A gap is between a bottom end of the coating-steel-plate sleeve and an upper surface of the foundation. An outside of the gap is provided with a plugging material. The joint further includes an annular rib plate all of whose edges are horizontally and seamlessly connected to an inner wall of the coating-steel-plate sleeve and whose middle portion is provided with a rib-plate opening. The annular rib plate is provided with a sleeve opening that matches with the grouting sleeve device.

Abstract: Various aspects of the present disclosure generally relate to control of a user device under a wet condition. In some aspects, a user device may determine whether the user device is operating under a wet condition; select, based at least in part on whether the user device is operating under the wet condition, a set of input components to control the user device, wherein the set of input components is selected from a plurality of different sets of input components; and configure a user interface of the user device according to the set of input components. Numerous other aspects are provided.

Abstract: A submarine optical positioning beacon system with self-generating capability, which has an array of underwater beacons. When the underwater rover moves to the vicinity of an certain underwater beacon, the underwater beacon's COMS sensor detects the underwater rover's light and then turns on the LED lamp group. The COMS sensor of the underwater rover analyzes the light species of the LED light group and converts it into digital information. The underwater rover analyzes the digital signal to obtain its location. Each underwater beacon has an independent power generation component, which generates power by utilizing ocean current, greatly increasing the working time of the beacon. The LED lamp group gives positional information feedback through the light, which can reduce the system power consumption and increase the system working duration.

Abstract: The present disclosure relates to the technical field of assembled concrete frame structures, and discloses a prestressed assembled concrete frame-joint connecting structure and constructing method thereof. The present disclosure includes a prefabricated concrete column, a prefabricated concrete superposed beam, a composite slab, a post-tensioned prestressing tendon that connects the prefabricated concrete column and the prefabricated concrete superposed beam, a bending-resistant energy consuming steel bar and a shearing-resistant high-strength steel bar.

Abstract: Provided is a double-station cleaning system comprising a cleaning machine, wherein workpieces are placed into the feeding frames, the feeding frames are placed into trays, the bracket assembly is pushed to drive the trays to move along the guide rail assembly, the feeding frame in one of the trays is conveyed to a feeding inlet of the cleaning machine, the feeding frame is pushed to move along two rows of nylon wheels and linear guide rails inside the cleaning machine, and the feeding frame and the workpieces in the feeding frame are pushed into the cleaning machine so as to be cleaned. The double stations work alternately, so that the work time is saved, and the work efficiency is increased.

Abstract: The present invention provides a method of dense 3D scene reconstruction based on monocular camera and belongs to the technical field of image processing and computer vision, which builds the reconstruction strategy with fusion of traditional geometry-based depth computation and convolutional neural network (CNN) based depth prediction, and formulates depth reconstruction model solved by efficient algorithm to obtain high-quality dense depth map. The system is easy to construct because of its low requirement for hardware resources and achieves dense reconstruction only depending on ubiquitous monocular cameras.

Abstract: The present invention provides a method for estimating high-quality depth map based on depth prediction and enhancement sub-networks, belonging to the technical field of image processing and computer vision. This method constructs depth prediction subnetwork to predict depth information from color image and uses depth enhancement subnetwork to obtain high-quality depth map by recovering the low-resolution depth map. It is easy to construct the system, and can obtain the high-quality depth map from the corresponding color image directly by the well-trained end to end network. The algorithm is easy to be implemented. It uses high-frequency component of color image to help to recover the lost depth boundaries information caused by down-sampling operators in depth prediction sub-network, and finally obtains high-quality and high-resolution depth maps. It uses spatial pyramid pooling structure to increase the accuracy of depth map prediction for multi-scale objects in the scene.

Abstract: Provided is a method based on deep neural network to extract appearance and geometry features for pulmonary textures classification, which belongs to the technical fields of medical image processing and computer vision. Taking 217 pulmonary computed tomography images as original data, several groups of datasets are generated through a preprocessing procedure. Each group includes a CT image patch, a corresponding image patch containing geometry information and a ground-truth label. A dual-branch residual network is constructed, including two branches separately takes CT image patches and corresponding image patches containing geometry information as input. Appearance and geometry information of pulmonary textures are learnt by the dual-branch residual network, and then they are fused to achieve high accuracy for pulmonary texture classification. Besides, the proposed network architecture is clear, easy to be constructed and implemented.

Abstract: Embodiments of this application provide a text recommendation method performed at an electronic device. The method includes: extracting feature content of from the a target text; processing the feature content by using at least two text analysis models to obtain at least two semantic vectors; integrating the at least two semantic vectors into an integrated semantic vector of the target text; selecting, according to the integrated semantic vector and an integrated semantic vector of at least one to-be-recommended text, a recommended text corresponding to the target text from the at least one to-be-recommended text. Because the integrated semantic vector of the target text is obtained based on the at least two text analysis models, the integrated semantic vector has a stronger representing capability. When text recommendation is subsequently performed, an association degree between the recommended text and the target text can be increased, thereby improving recommendation accuracy.

Abstract: A submarine optical positioning beacon system with self-generating capability, which has an array of underwater beacons. When the underwater rover moves to the vicinity of an certain underwater beacon, the underwater beacon's COMS sensor detects the underwater rover's light and then turns on the LED lamp group. The COMS sensor of the underwater rover analyzes the light species of the LED light group and converts it into digital information. The underwater rover analyzes the digital signal to obtain its location. Each underwater beacon has an independent power generation component, which generates power by utilizing ocean current, greatly increasing the working time of the beacon. The LED lamp group gives positional information feedback through the light, which can reduce the system power consumption and increase the system working duration.

Abstract: The present disclosure relates to the technical field of assembled concrete frame structures, and discloses a prestressed assembled concrete frame-joint connecting structure and constructing method thereof. The present disclosure includes a prefabricated concrete column, a prefabricated concrete superposed beam, a composite slab, a post-tensioned prestressing tendon that connects the prefabricated concrete column and the prefabricated concrete superposed beam, a bending-resistant energy consuming steel bar and a shearing-resistant high-strength steel bar.

Abstract: The column-base joint includes a prefabricated-reinforced-concrete column and a reinforced-concrete foundation, and further includes a coating-steel-plate sleeve that is integrally prefabricated with the concrete column and encloses an exterior of a column base. A gap is between a bottom end of the coating-steel-plate sleeve and an upper surface of the foundation. An outside of the gap is provided with a plugging material. The joint further includes an annular rib plate all of whose edges are horizontally and seamlessly connected to an inner wall of the coating-steel-plate sleeve and whose middle portion is provided with a rib-plate opening. The annular rib plate is provided with a sleeve opening that matches with the grouting sleeve device.

Abstract: The joint has a prefabricated-reinforced-concrete column, a reinforced-concrete foundation, a column anchoring longitudinal bar, a grouting sleeve and a foundation anchoring steel bar. The foundation anchoring steel bar and the column anchoring longitudinal bar are connected by a seam filling material filling the grouting sleeve. A splicing seam between the reinforced-concrete foundation and the prefabricated-reinforced-concrete column is filled with the seam filling material. The foundation anchoring steel bar includes a vertical portion and a horizontal portion. The vertical portion includes an upper-portion anchoring section protruding out of an upper surface of the reinforced-concrete foundation, a middle-portion non-adhesive section buried within the foundation and a lower-portion anchoring section. An exterior of the middle-portion non-adhesive section is provided with an isolating sheath for isolating the middle-portion non-adhesive section and the concrete adhesion.