11results about How to "Deepen the depth" patented technology

The invention discloses a beam focusing optimization method based on dense connection expansion self-coding element learning, and relates to the technical field of super-large-scale IRS assistance, and the method comprises the following steps: S1, analyzing a super-large-scale intelligent reflector-assisted multi-user communication system, and obtaining a problem optimization formula; s2, constructing a beam focusing optimization algorithm network based on dense connection expansion self-coding element learning according to a problem optimization formula, and taking a matrix form of a phase shift matrix and a precoding matrix as the input of the beam focusing optimization algorithm network based on dense connection expansion self-coding element learning, and obtaining an optimized pre-coding matrix and an optimized phase shift matrix, and obtaining the maximum spectrum efficiency value of the super-large scale intelligent reflector assisted near field communication system according to the optimized pre-coding matrix and the optimized phase shift matrix. According to the invention, the spectrum efficiency of the system can be improved and the processing overhead of high-dimensional signals can be reduced.

The utility model discloses a fast tunneling device for broken roofs, which relates to the technical field of tunneling devices and comprises a draw-off pump, a draw-off pipe is fixedly mounted on one side of the draw-off pump, a storage tank is fixedly mounted on one side of the draw-off pipe, and a diverter valve is fixedly mounted at the top of the draw-off pump. Flow dividing pipes are fixedly mounted in the two sides of the flow dividing valve, cross valves are fixedly mounted at the tops of the flow dividing pipes, and conveying pipes are fixedly mounted in the two sides and the top of each cross valve. According to the rapid tunneling device for the broken roof, chemical liquid in the storage tank can be pumped out through the draw-off pump and the draw-off pipe and then fed into the flow dividing valve, the liquid can be fed into the cross valve through the two flow dividing pipes after being divided through the flow dividing valve, and the liquid is fed into the spray head through the three conveying pipes after being divided through the cross valve; and finally, liquid can be sprayed out through a spray head, so that grouting reinforcement can be conducted on the broken top plate, and the broken top plate can be supported through a supporting plate.

The present invention relates to a cleaning machine, comprising a plastic water tank body with a washing chamber, and also comprising a metal reinforcement, wherein the metal reinforcement is arranged on the inner wall surface of the upper part of the water tank body, a first reinforcement rib is arranged at least in the middle of the outer peripheral wall of the water tank body, and a plurality of spaced protrusions are arranged at least in the lower part of the inner wall surface of the water tank body. The present invention manufactures the water tank body with plastic, which has strong plasticity, and is convenient for deepening the depth of the tank body according to needs, and also greatly reduces the production cost; the metal reinforcement, the first reinforcement rib and the protrusion can all improve the strength of the tank body and prevent the water tank body from being deformed by heat or cold; the metal reinforcement can also accelerate the condensation of steam at the bottom of the washing chamber; the protrusion can also resonate with the ultrasonic sounding device at the bottom of the water tank cleaning machine to improve the ultrasonic cleaning effect; the outer wall surface of the water tank body can play a shockproof and soundproof effect.

The invention provides a mask feature progressive knowledge distillation completion method for a user portrait of a nasopharyngeal carcinoma patient, and relates to the technical field of image data processing, and the method comprises the following steps: a pre-distillation stage: carrying out the pre-distillation of a pre-training language model, so as to obtain a high-quality teacher model, generating feature distillation through a mask, and carrying out the feature distillation; applying a mask operation in the input data, and setting a mask proportion according to the number of student model parameters; and S2, mask generation feature distillation: the teacher model and the student model perform mask operation on input data including the triad and text description thereof in a data input stage. In the masking generation characteristic distillation, masking operation is applied to input marks, and a proper masking rate is set in a pre-distillation stage according to student model parameter counting, so that the performance of an initial model is improved. In the progressive distillation stage, a multi-grade student model is designed, and the parameter count and the mask rate are gradually reduced. Each student model is trained under the three kinds of supervision described above.

The invention discloses a laser shock marking method for a gear transmission component. The laser shock marking method comprises the steps that the to-be-marked position of the gear transmission component is determined; forming a light-transmitting restraint layer capable of allowing high-amplitude plasma shock waves to propagate on the surface of the position to be marked; adjusting the distance between the position to be marked and the laser to a preset distance; performing laser impact marking on the to-be-marked position based on a laser; a pit is introduced into the surface of the position to be marked based on laser spot impact induced plastic deformation; and forming a mark based on continuous lap joint of the pits. According to the impact marking method, an absorption layer is not used in the marking process, the marking efficiency is improved, the marking width is increased, the mechanical property of the material is greatly improved, and the fatigue life of the material is greatly prolonged.

The invention discloses a method for segmenting an aorta blood vessel image from an X-ray angiography image and a related device, and belongs to the technical field of medical instruments. The method comprises the following steps: extracting blood vessel features from an X-ray angiography image, constructing and optimizing an aorta blood vessel segmentation network according to the extracted blood vessel features, and segmenting an aorta blood vessel image from the X-ray angiography image by using the optimized aorta blood vessel segmentation network. According to the method, learning simulation is carried out on an X-ray angiography image through computer software, calibration and optimization are carried out on the X-ray angiography image in a DSA operation, registration of an actual operation image and a pre-stored image can be achieved, a high-pixel-level accurate segmentation result can be provided, and the segmentation accuracy is improved. In addition, the integrity of the topological structure of the blood vessel can be kept in the process, and more detailed image information is provided for surgeons.

The present invention relates to a cleaning machine, comprising a water tank body with a washing chamber, the water tank body is made of plastic, and also comprises a metal part, the metal part is arranged on the inner wall surface of the upper part of the water tank body, and a plurality of spaced protrusions are arranged on the inner wall surface of the water tank body at least at the lower part. The present invention makes the water tank body of plastic, which has strong plasticity, and is convenient for deepening the depth of the tank body as needed, and also greatly reduces the production cost; when the hot water cleaning is completed, a large amount of steam will be retained at the top of the washing chamber, and the metal part can accelerate the condensation of the steam, and the metal part also improves the strength of the upper part of the plastic tank body, preventing the water tank body from being deformed by heat, and can also resonate with the ultrasonic sounding device in the water tank cleaning machine to improve the ultrasonic cleaning effect; the protrusion at the bottom of the water tank body can prevent the water tank body from being deformed by heat or cold, and at the same time, it can also resonate with the ultrasonic sounding device at the bottom of the water tank cleaning machine to improve the ultrasonic cleaning effect.

The invention aims to provide an underwater image super-resolution reconstruction method based on deep residual learning, relates to the technical field of underwater image super-resolution reconstruction, and provides an underwater image super-resolution reconstruction method based on a deep residual learning network (VDSR), which introduces a residual block structure and combines an alternate connection strategy to deepen the depth of the network. Not only is the characterization capability of the network to a complex underwater scene enhanced, but also the common problems of gradient disappearance and model degradation of the deep network are effectively relieved by introducing a residual learning mechanism. The alternative configuration of the identical residual connection blocks and the convolution residual connection blocks not only retains the stability advantage of the identical connection, but also promotes the effective multiplexing and transmission of the features, and significantly reduces the phenomenon of low-level feature loss in the deep network. Experimental results show that the proposed method is generally superior to a comparison algorithm in subjective and objective evaluation, and the texture, edge and other information of the reconstructed underwater image are greatly reserved.

This utility model relates to the field of grinding technology, specifically to a grinding device for keyway grinding of workpiece inner holes. It includes a base plate and a moving component. The base plate has an accommodating opening, and a top block and a bottom block are respectively provided at the top and bottom of the accommodating opening. The top and bottom blocks extend horizontally in the same direction and are perpendicular to the base plate. Top grooves and bottom grooves are respectively provided on the sides of the top and bottom blocks that are close to each other. The moving component is located between the top and bottom blocks, and its top and bottom are slidably connected within the top and bottom grooves, respectively. In use, this grinding device for keyway grinding of workpiece inner holes uses a reciprocating drive mechanism to repeatedly move the grinding mechanism to grind the keyway of a workpiece fixed within a fixed mechanism. The reciprocating drive mechanism ensures grinding effect and efficiency. Furthermore, during the grinding process, the workpiece can be moved by a moving motor to deepen the keyway depth, meaning that keyways of different depths can be processed using the same honing rod, offering flexibility in use.

The utility model discloses an anchor structure of floating type photovoltaic power station, including photovoltaic platform and the anchor chain of setting in photovoltaic platform bottom four corners, every anchor chain bottom is provided with counterweight, every middle part of anchor chain is provided with elastic sleeve pole, the utility model can promote the stability of photovoltaic platform.