Abstract: A system and method for unsupervised superpixel-driven instance segmentation of a remote sensing image are provided. The remote sensing image is divided into one or more image patches. The one or more image patches are processed to generate one or more superpixel aggregation patches based on a graph-based aggregation model, respectively. The graph-based aggregation model is configured to learn at least one of a spatial affinity or a feature affinity of a plurality of superpixels from each image patch and aggregate the plurality of superpixels based on the at least one of the spatial affinity or the feature affinity of the plurality of superpixels. The one or more superpixel aggregation patches are combined into an instance segmentation image.

Abstract: A composite circuit board includes a flexible board, rigid boards, adhesive layers, and protection glue; the adhesive layers are sandwiched between the rigid boards and the flexible board and used for bonding the rigid boards and the flexible board; the rigid boards are provided with step slots passing through the rigid boards; the adhesive layers are provided with through slots passing through the adhesive layers; the step slots and the through slots are communicated with each other to form a thinning recess; the thinning recess exposes the flexible board; and the protection glue covers steps of the thinning recess and at least a portion of the exposed area of the flexible board.

Abstract: The present invention discloses a gas based testing method for a pore volume compressibility of a low-permeability rock. The method comprises the following steps: obtaining a compressibility and an isothermal adsorption-desorption curve of experimental gas under a reservoir condition; performing a physical simulation experiment by using a steel core drilled with a through hole, and obtaining a gas volume collected due to the deformation of an experimental instrument; then repeating the experiment by using an actual core with the same size, and obtaining pore volumes of the reservoir core under the experimental conditions by deducting deformation of the instrument; and finally, obtaining a pore compressibility of the reservoir core under different reservoir conditions according to the pore volumes of the reservoir core under different reservoir conditions. The present invention considers gas adsorption, corrects errors caused by deformation of a physical simulation device.

Abstract: A computer device extracts an image feature of an image that includes one or more characters to be recognized. The image feature includes a plurality of image feature vectors. The device uses an attention mechanism to compute and output attention weight values corresponding to the target number of characters, based on the image feature vectors, through parallel computing. Each of the attention weight values corresponds to one or more respective characters and represents an importance of the plurality of image feature vectors for the respective characters. The device obtains at least one character according to the plurality of image feature vectors and the target number of attention weight values. Therefore, in a character recognition process, with recognition based on the foregoing attention mechanism, a character in any shape can be effectively recognized by using a simple procedure, thereby avoiding a cyclic operation process and greatly improving operation efficiency.

Abstract: The present invention provides a hybrid-power-driven underwater robot, including a robot body and a releasing and recovering device, wherein a first transmission cavity is formed in the releasing and recovering device, a seawater removing mechanism and an electric take-up reel are arranged inside the first transmission cavity, a cable is arranged on the electric take-up reel, the cable extends out of the first transmission cavity, and the robot body is fixedly connected to one end of the cable; and a tail fin is movably connected to the robot body, an acceleration mechanism is further arranged on the robot body, a second transmission cavity is formed in the robot body, a driving mechanism for driving the tail fin and the acceleration mechanism is arranged inside the second transmission cavity, and the driving mechanism can drive the tail fin separately or drive the acceleration mechanism separately.

Abstract: A circuit board, a preparation method thereof, and an electronic device are provided. The circuit board includes: a substrate, defining a first through-hole; a metal block, embedded in the first through-hole and fixedly connected to the substrate; a conductive line layer, arranged on at least one side surface of the substrate; wherein the conductive line layer partially covers an opening of the first through-hole on a corresponding side surface of the substrate; and a conductive channel, penetrating the conductive line layer and the metal block in turn. The conductive channel comprises a second through-hole and a conductive medium plated on a wall of the second through-hole; an end of the conductive medium is connected to the conductive line layer, and another end of the conductive medium is connected to the metal block.

Abstract: A media access control (MAC) address allocation method and device are disclosed. The method includes: a wireless access point device sending a notification message on a wireless medium. The notification message includes at least one candidate virtual MAC address. The wireless access point device receiving, on the wireless medium, a request message from a terminal device. A source address of the request message is a first virtual MAC address. The first virtual MAC address is one of the at least one candidate virtual MAC address.

Abstract: A system and method for unsupervised superpixel-driven instance segmentation of a remote sensing image are provided. The remote sensing image is divided into one or more image patches. The one or more image patches are processed to generate one or more superpixel aggregation patches based on a graph-based aggregation model, respectively. The graph-based aggregation model is configured to learn at least one of a spatial affinity or a feature affinity of a plurality of superpixels from each image patch and aggregate the plurality of superpixels based on the at least one of the spatial affinity or the feature affinity of the plurality of superpixels. The one or more superpixel aggregation patches are combined into an instance segmentation image.

Abstract: A method, applied to an apparatus for mammographic multi-view mass identification, includes receiving a main image, a first auxiliary image, and a second auxiliary image. The main image and the first auxiliary image are images of a breast of a person, and the second auxiliary image is an image of another breast of the person. The method further includes detecting the nipple location based on the main image and the first auxiliary image; generating a first probability map of the main image based on the main image, the first auxiliary image, and the nipple location; generating a second probability map of the main image based on the main image, the second auxiliary image, and the nipple location; and generating and outputting a fused probability map based on the first probability map and the second probability map.

Abstract: The present disclosure relates to a steering wheel, a steering wheel vibration prompt system and a vehicle incorporating such a steering wheel. The steering wheel includes a steering wheel body, a flexible support member, a vibration member and a vibrator. The steering wheel body is provided with an accommodation space. The vibration member is connected to the flexible support member and is suspended and supported on the steering wheel body by the flexible support member. The vibrator is connected to the vibration member and drives the vibration member to vibrate. The vibrator of the steering wheel and the sensor device cooperate with each other, and the vibrator performs a vibration action according to the output signal of the sensor device, and then drives the vibration member to perform corresponding vibration actions.

Abstract: The present disclosure provides a printed circuit board and a manufacturing method of the printed circuit board. The manufacturing method may include: at least two core plates may be provided; a composite anti-glue film assembly may be arranged at a preset position of one of the at least two core plates, the composite anti-glue film assembly may include a first anti-glue film layer, a second anti-glue film layer and a bonding layer. The first anti-glue film layer may contact the preset position. The first anti-glue film layer may be a polyimide layer. The bonding layer may be configured to bond the first anti-glue film layer and the second anti-glue film layer together to produce the composite anti-glue film assembly. Two adjacent core plates may be connected through a medium layer. The core plates may be cut-out and form the printed circuit board.

Abstract: An energy harvesting device based on wave energy comprises: a box body; a swing rod hinged with the box body, wherein one end of the swing rod extends into the box body, and a float is fixed at the other end of the swing rod; a base layer disposed inside the box body and provided with a plurality of piezoelectric patches in a length direction; and an energy transmission assembly fixed inside the box body and located between the swing rod and the base layer, one side of the energy transmission assembly is connected with one end of the swing rod extending into the box body, and the other side of the energy transmission assembly is in transmission connection with the base layer, wherein the energy transmission assembly converts the swinging of the swing rod into squeezing to the base layer.

Abstract: A composite circuit board includes a flexible board, rigid boards, adhesive layers, and protection glue; the adhesive layers are sandwiched between the rigid boards and the flexible board and used for bonding the rigid boards and the flexible board; the rigid boards are provided with step slots passing through the rigid boards; the adhesive layers are provided with through slots passing through the adhesive layers; the step slots and the through slots are communicated with each other to form a thinning recess; the thinning recess exposes the flexible board; and the protection glue covers steps of the thinning recess and at least a portion of the exposed area of the flexible board.

Abstract: A global and local binary pattern image crack segmentation method based on robot vision comprises the following steps: enhancing a contrast of an acquired original image to obtain an enhanced map; using an improved local binary pattern detection algorithm to process the enhanced map and construct a saliency map; using the enhanced map and the saliency map to segment cracks and obtaining a global and local binary pattern automatic crack segmentation method; and evaluating performance of the obtained global and local binary pattern automatic crack segmentation method. The present application uses logarithmic transformation to enhance the contrast of a crack image, so that information of dark parts of the cracks is richer. Texture features of a rotation invariant local binary pattern are improved. Global information of four directions is integrated, and the law of universal gravitation and gray and roundness features are introduced to correct crack segmentation results, thereby improving segmentation accuracy.

Abstract: The present invention discloses a device carrying platform based on an underwater robot. The device carrying platform comprises a mounting rack and more than two mounting seats. The mounting rack comprises a mounting plate and more than two mounting columns fixed to the bottom of the mounting plate, the mounting plate is provided with a plurality of mounting holes, the mounting plate is hinged with a protective screen, the protective screen is provided with an opening, and the opening is hinged with a screen door. The plurality of mounting seats arranged on the mounting rack can be used for mounting devices and instruments of various models, and a first butting part and a second butting part in a mounting groove are matched with each other, so that devices and instruments needed to be mounted are fixed.

Abstract: A circuit board, a preparation method thereof, and an electronic device are provided. The circuit board includes: a substrate, defining a first through-hole; a metal block, embedded in the first through-hole and fixedly connected to the substrate; a conductive line layer, arranged on at least one side surface of the substrate; wherein the conductive line layer partially covers an opening of the first through-hole on a corresponding side surface of the substrate; and a conductive channel, penetrating the conductive line layer and the metal block in turn. The conductive channel comprises a second through-hole and a conductive medium plated on a wall of the second through-hole; an end of the conductive medium is connected to the conductive line layer, and another end of the conductive medium is connected to the metal block.

Abstract: The present invention discloses a device carrying platform based on an underwater robot. The device carrying platform comprises a mounting rack and more than two mounting seats. The mounting rack comprises a mounting plate and more than two mounting columns fixed to the bottom of the mounting plate, the mounting plate is provided with a plurality of mounting holes, the mounting plate is hinged with a protective screen, the protective screen is provided with an opening, and the opening is hinged with a screen door. The plurality of mounting seats arranged on the mounting rack can be used for mounting devices and instruments of various models, and a first butting part and a second butting part in a mounting groove are matched with each other, so that devices and instruments needed to be mounted are fixed.

Abstract: The present disclosure relates to a steering wheel, a steering wheel vibration prompt system and a vehicle incorporating such a steering wheel. The steering wheel includes a steering wheel body, a flexible support member, a vibration member and a vibrator. The steering wheel body is provided with an accommodation space. The vibration member is connected to the flexible support member and is suspended and supported on the steering wheel body by the flexible support member. The vibrator is connected to the vibration member and drives the vibration member to vibrate. The vibrator of the steering wheel and the sensor device cooperate with each other, and the vibrator performs a vibration action according to the output signal of the sensor device, and then drives the vibration member to perform corresponding vibration actions.

Abstract: A method, applied to an apparatus for mammographic multi-view mass identification, includes receiving a main image, a first auxiliary image, and a second auxiliary image. The main image and the first auxiliary image are images of a breast of a person, and the second auxiliary image is an image of another breast of the person. The method further includes detecting the nipple location based on the main image and the first auxiliary image; generating a first probability map of the main image based on the main image, the first auxiliary image, and the nipple location; generating a second probability map of the main image based on the main image, the second auxiliary image, and the nipple location; and generating and outputting a fused probability map based on the first probability map and the second probability map.

Abstract: A computer device extracts an image feature of an image that includes one or more characters to be recognized. The image feature includes a plurality of image feature vectors. The device uses an attention mechanism to compute and output attention weight values corresponding to the target number of characters, based on the image feature vectors, through parallel computing. Each of the attention weight values corresponds to one or more respective characters and represents an importance of the plurality of image feature vectors for the respective characters. The device obtains at least one character according to the plurality of image feature vectors and the target number of attention weight values. Therefore, in a character recognition process, with recognition based on the foregoing attention mechanism, a character in any shape can be effectively recognized by using a simple procedure, thereby avoiding a cyclic operation process and greatly improving operation efficiency.