84results about How to "Improve compression performance" patented technology

The invention relates to the technical field of hydrogel, in particular to attapulgite-based hydrogel and a preparation method thereof. The hydrogel is prepared by polymerizing an alkenyl monomer and modified attapulgite, and the modified attapulgite is obtained by modifying attapulgite with branched acrylate quaternary ammonium salt. The hydrogel shows excellent water absorption, water retention and mechanical properties. The water absorption rate reaches up to 292%, the water retention capacity is excellent in the low-humidity environment, water can be kept for a long time in the drought or low-humidity environment, the continuous water supply capacity is remarkably improved, and the water retention agent is suitable for the fields of agriculture, soil improvement and ecological restoration.

The invention belongs to the technical field of natural language information processing, and discloses a text compression method based on natural language features and self-information, and the specific technical scheme is as follows: step 1, carrying out data preprocessing, useless information deletion and multi-document retrieval simulation on a Google question and answer data set in an open source corpus to obtain a data set oriented to long context compression, obtaining a multi-document long text; 2, performing self-information evaluation on the multi-document long text to obtain phrase granularity and text self-information distribution, extracting natural language features of part-of-speech analysis, named entity analysis and dependency relationship analysis, and optimizing distribution based on the natural language features to obtain a compressed text after distribution optimization; and step 3, compressed text prompt setting, target question and answer model construction, question and answer performance index obtaining and application to downstream tasks, the method is very excellent in long text compression accuracy, and the text compression performance is improved.

The invention discloses a model compression method and device, and the method comprises the steps: obtaining a to-be-compressed model; pruning the to-be-compressed model based on a preset pruning strategy to obtain a pruning model; and performing knowledge distillation on the pruning model based on the to-be-compressed model to obtain a target compression model corresponding to the pruning model. After the to-be-compressed model is obtained, the to-be-compressed model is pruned based on the preset pruning strategy, and then the knowledge of the to-be-compressed model is transmitted to the pruning model through knowledge distillation, so that pruning and knowledge distillation can be organically combined, and the compressed model simultaneously has the flexibility of pruning and the relatively high compression performance of knowledge distillation; the model compression effect can be improved.

The invention provides a cellulose composite carbon aerogel and a preparation method thereof, and the cellulose composite carbon aerogel is prepared by taking cellulose nanofibers and carbon nanotubes as raw materials, mixing and dispersing the raw materials through dispersoid, and then preparing cellulose nanofiber / carbon nanotube composite carbon aerogel through bidirectional freezing, freeze drying and high-temperature carbonization. The invention also provides an application of the cellulose composite carbon aerogel in the aspect of a battery electrode, and the cellulose composite carbon aerogel has an ultra-large specific surface area and an excellent compression ratio under ultra-low density, and shows super-hydrophilicity, high electrochemical activity, good compression resilience, good structural stability and certain mechanical strength; and the method can be used for various electrochemistry-based water-based flow batteries. The invention solves the technical problem that the existing cellulose carbon aerogel is poor in compressibility and mechanical strength, and belongs to the technical field of batteries.

The invention belongs to the technical field of intravascular stents, and particularly discloses a degradable peripheral artery stent, the cross section of the degradable peripheral artery stent is of an annular structure, the degradable peripheral artery stent is applied to the peripheral artery of the human body and comprises a plurality of supporting units and connecting units in the axial direction of the stent, the connecting units are located between the supporting units, and the adjacent supporting units are connected through the connecting units; the supporting unit comprises an end supporting unit and a middle supporting unit, and the end supporting unit and the middle supporting unit are each of a closed-loop structure formed by intersecting a sine-like curve and a reverse sine-like curve. The stent is integrally formed through a 3D printing technology, and is high in structural stability and good in flexibility. According to the degradable peripheral artery stent of the structure, degradable polylactic acid is adopted as a stent material, inflammatory reaction caused by the fact that the stent exists in blood vessels for a long time is greatly avoided, and the stent has good flexibility and adherence performance under the condition that the stent has enough radial supporting force; and the risk of restenosis after the stent operation is reduced.

The invention belongs to the technical field of artificial intelligence and research and development, and relates to a video compression method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining a target video frame in a to-be-compressed video stream, and generating a thermodynamic diagram according to the target video frame; dividing the thermodynamic diagram into a plurality of thermodynamic diagram macro blocks, and calculating a thermodynamic value of each thermodynamic diagram macro block; correspondingly determining a compression strategy of each thermodynamic diagram macro block according to the thermodynamic value of each thermodynamic diagram macro block; and compressing the target video frame according to the compression strategy of each thermodynamic diagram macro block. In addition, the invention also relates to a block chain technology, and data such as a target video frame and a thermodynamic diagram can be stored in a block chain. According to the invention, the problem of quality reduction of video key content in a dynamic video processing process can be solved.

The invention discloses a red clay improver and a red clay improvement method. The red clay modifier comprises rice hull ash and carbide slag. The improvement method of the red clay comprises the step of mixing the rice hull ash and the carbide slag with different mixing amounts with the red clay according to the optimal proportion. Obtaining the maximum dry density and the optimal water content of the mixed improved soil of each mixing amount; according to the maximum dry density and the optimal water content, the improved soil is prepared into multiple test pieces, and curing is conducted for 28 days; carrying out related mechanical property tests on the test pieces; and determining the optimal mixing amount of the rice hull ash and carbide slag improved red clay according to the mechanical property test result. According to the method, the unconfined compressive strength, the shear strength and the compression performance of the red clay can be remarkably improved, the improved materials are all industrial solid waste materials, the materials are convenient to obtain and low in cost, environmental pollution can be reduced, and the method is particularly suitable for solving the practical engineering problem in the red clay area.

The invention belongs to the technical field of terahertz, relates to terahertz communication, and particularly discloses a quasi-optical resonant cavity and a terahertz pulse compression system. According to the invention, the pulse generation module and the coupling compensation module are connected in parallel and are connected to the common reflector through the common coupler, so that function decoupling is realized. In a pulse compression stage, the common coupler introduces an incident terahertz wave beam into two optical paths, and destructive interference occurs between two paths of leakage signals by adjusting the coupling compensation module, so that zero coupling loss is realized, and the problem of large energy loss is solved. Meanwhile, the pulse generation module can independently adjust the optical path of the pulse generation module to control the pulse width, and the adjustment process does not affect the optimized interference condition, so that the high robustness of adjustment is ensured. In the output stage, energy is released by switching the state of the common coupler. Compared with the prior art, low-loss accumulation and robust pulse width continuous adjustment are realized at the same time, so that the terahertz pulse compression effect is remarkably improved.

The application provides a generative image compression method based on vector quantization, and belongs to the technical field of image and video compression. The method comprises the following steps: obtaining a continuous latent representation of an input image through an analysis transformation module; then performing vector quantization processing to obtain discrete indexes of the continuous latent representation and corresponding quantized features; constructing a continuous index probability distribution; predicting a conditional probability distribution of the discrete indexes through a conditional autoregressive entropy model; calculating a coding rate based on the continuous index probability distribution and the conditional probability distribution; reconstructing an image based on the quantized features and constructing a distortion loss; constructing a rate-distortion loss function and jointly training an image compression model comprising the analysis transformation module, the vector quantization module, the conditional autoregressive entropy model and a synthesis transformation module to obtain a trained image compression model; and compressing the input image by using the trained image compression model. The application can realize rate-distortion joint optimization and collaborative improvement of compression efficiency and reconstruction quality.

The present invention discloses an end-to-end video compression method and system based on a self-attention mechanism, which relates to the field of video coding, and the method includes: extracting shallow features of a current frame, a motion reference frame and a reconstructed frame at the previous moment; extracting motion information features between the current frame and the motion reference frame; compressing the motion information features to obtain compressed motion information features; compensating the compressed motion information features on the shallow features of the reconstructed frame at the previous moment to obtain context information features at the current moment; compressing to obtain compressed context information features; encoding the compressed context information features and the shallow features of the current frame to fuse features to obtain the reconstructed frame at the current moment; upsampling the reconstructed frame at the current moment to obtain a compressed viewpoint image; correcting the motion information accumulated error in the compressed viewpoint image to obtain the motion reference frame of the next frame. The present invention can improve the accuracy of motion estimation while improving the efficiency of coding.

The present invention discloses an image encoding and decoding method and device for eliminating spatial redundancy. An adjacent block attention model is set in the original encoding and decoding transformation network. When processing image features, information of adjacent (up, down, left, and right) positions can be introduced, thereby expanding the receptive field, better capturing and eliminating the redundancy of the image in space, and improving the image compression effect.

The present invention discloses a binocular video compression method, device and readable medium based on deep learning, which relates to the field of video processing, including: constructing and training a binocular video compression model to obtain a trained binocular video compression model, respectively obtaining the original frame of the left viewpoint at the current moment and the original frame of the previous moment and the original frame of the right viewpoint at the current moment and the original frame of the previous moment in the binocular video to be compressed and inputting them into the trained binocular video compression model, reconstructing the reconstructed frame sequence of the left viewpoint and the reconstructed frame sequence of the right viewpoint in an iterative manner, and obtaining a compressed binocular video. The present invention solves the problems of low redundancy utilization and low compression performance between viewpoints of the existing binocular video.

The utility model relates to the technical field of air compressors, in particular to a waterproof air compressor which comprises a waterproof shell, an air outlet pipe, an air collecting cover, a partition plate and an air compressor shell, the upper portion of the left portion of the waterproof shell is communicated with the air outlet pipe, the funnel-shaped air collecting cover is fixed to the middle of the front portion of the waterproof shell, and an air inlet of the air collecting cover faces forwards and is located in the center. Partition plates are arranged on all the inner wall faces of the waterproof shell, all the partition plates form a closed cavity, the partition plate on the left portion is communicated with the air outlet pipe, the partition plate on the front portion is communicated with the air collecting cover, an air compressor shell is installed in the closed cavity, and the rear portion of the air compressor shell is designed to be in a cross-shaped hollowed-out mode. According to the utility model, the waterproof housing is combined with the separator plate, so that internal parts are not affected by external moisture intrusion; through the combined design of the heat dissipation fins and the embedded electric fan, heat generated by the first motor is effectively conducted and dissipated out, and faults caused by overheating are avoided; the air compressor shell designed in a cross-shaped hollow-out mode is beneficial for evenly distributing pressure, and the air compression effect is improved.

The present application relates to the field of data compression technology, and in particular to a method and system for optimizing storage of control terminal data, the method comprising: performing polynomial fitting on the data to be stored to obtain a fitting error; in response to the fitting error being less than an error threshold, storing the fitting polynomial and completing the storage of the data to be stored, otherwise, calculating the concentration of abnormal residual values; in response to the concentration being less than the concentration threshold, storing the fitting polynomial, abnormal residual values ​​and their timestamps, and completing the storage of the data to be stored; in response to the concentration being not less than the concentration threshold, storing the fitting polynomial and the residual value sequence within the abnormal time period. Through the technical solution of the present application, according to the distribution of the fitting error and abnormal residual values ​​in the polynomial fitting process, the storage strategy can be flexibly selected to effectively optimize the use of storage space, thereby improving storage efficiency.

Provided is a vacuum pump having excellent back pressure performance even if a suction gas is hydrogen or the like. A vacuum pump (100) is provided with an exhaust section (114) configured from a turbine pump section (115) having a plurality of layers of rotating blades (102) and fixed blades (123), and a traction pump section (116) positioned further to the exhaust downstream side than the turbine pump section (115), and suction gas sucked from an intake port (101) is discharged from an exhaust port (133) via a gas flow path (FP1) by means of the exhaust section (114). The present invention is provided with: an introduction flow path (FP3) into which an intermediate introduction gas having a higher viscosity than the intake gas is introduced; the introduction flow path (FP3) is connected to the gas flow path (FP1) on the exhaust downstream side of the rotating blade (102) closest to the intake port (101) in the exhaust section (114) among the plurality of layers of rotating blades (102).

An encoding and decoding method, an encoder, a decoder, and a storage medium. The decoder decodes a bitstream, determines high throughput mode identification information of a current block, and determines, on the basis of the high throughput mode identification information, a coefficient of the current block using at least one of a context mode, a bypass mode, and an alignment process. The decoder determines decoded information, determines the high throughput mode identification information of the current block according to the decoded information, and determines, on the basis of the high throughput mode identification information, the coefficient of the current block using at least one of the context mode, the bypass mode, and the alignment process. The encoder determines the high throughput mode identification information of the current block, encodes, on the basis of the high throughput mode identification information, the coefficient of the current block using at least one of the context mode, the bypass mode, and the alignment process, and generates the bitstream. The encoder determines encoded information, determines the high throughput mode identification information of the current block according to the encoded information, encodes, on the basis of the high throughput mode identification information, the coefficient of the current block using at least one of the context mode, the bypass mode, and the alignment process, and generates the bitstream.

Disclosed is an image data encoding / decoding method and apparatus. A method for decoding a 360-degree image comprises the steps of: receiving a bitstream obtained by encoding a 360-degree image; generating a prediction image by making reference to syntax information obtained from the received bitstream; combining the generated prediction image with a residual image obtained by dequantizing and inverse-transforming the bitstream, so as to obtain a decoded image; and reconstructing the decoded image into a 360-degree image according to a projection format. Here, the step of generating the prediction image comprises the steps of: obtaining, from motion information included in the syntax information, a motion vector candidate group including a motion vector of a block adjacent to a current block to be decoded; deriving a prediction motion vector from the motion vector candidate group, on the basis of selection information extracted from the motion information; and determining a prediction block for the current block to be decoded, using a fmal motion vector derived by adding the prediction motion vector to a differential motion vector extracted from the motion information.

The invention relates to the technical field of building drainage, and discloses an anti-blocking hidden building drainage pipeline which comprises an embedded mounting box and an elastic pipeline arranged in the embedded mounting box, the top end of the elastic pipeline communicates with a water inlet connector, and the water inlet connector penetrates through the top wall of the embedded mounting box; a plurality of adjusting plates are arranged on the outer side of the elastic pipeline; when impurities block the interior of the elastic pipeline, the multiple adjusting plates can be driven to get close to one another through the adjusting assembly, so that the elastic pipeline is extruded, the inner diameter of the elastic pipeline is reduced, the impurities blocked in the elastic pipeline are compressed, and after compression is carried out for a period of time, the impurities in the elastic pipeline are removed. And then the multiple adjusting plates are driven by the adjusting assembly to be away from each other, so that the inner diameter of the elastic pipeline is increased, a gap is formed between the impurities and the inner wall of the elastic pipeline, then the impurities in the elastic pipeline can be discharged downwards under the action of water flow, and therefore the drainage pipeline is prevented from being blocked.

The utility model discloses a hexachlorodisilane pretreatment device, and relates to the technical field of semiconductors, through cooperation of a filtering device, a material receiving box, a material pushing mechanism, a compression box, an electric sealing door, a compression mechanism, an electric gate and other parts, hexachlorodisilane is pretreated by the filtering device, and physical impurities fall into the material receiving box below through a waste material outlet. The pushing mechanism pushes waste in the collecting box to the side, away from the waste outlet, of the collecting box. And when the waste materials need to be compressed, the electric sealing door is opened, the material collecting box communicates with the interior of the compression box, and the waste materials enter the compression box. And the compression mechanism compresses the waste materials entering the compression box, and after compression is completed, the electric gate is opened to discharge the compressed waste materials. Physical impurities generated in the hexachlorodisilane pretreatment process are collected, compressed and packaged, centralized treatment is facilitated, and environmental pollution is effectively avoided.

The invention discloses a push shovel device and transfer equipment, the push shovel device comprises a slide shovel mechanism and a turnover mechanism, the slide shovel mechanism is provided with a first material pushing surface, the first material pushing surface faces the back of the slide shovel mechanism, and the turnover mechanism is movably arranged on the slide shovel mechanism between a first position and a second position; and at the second position, the turnover mechanism protrudes out of the first material pushing surface, and the second position is located behind the first position. According to the push shovel device, the compression effect of the push shovel device is improved by moving the turnover mechanism.

The invention discloses a data compression method and device, a storage medium and electronic equipment, a server can perform multiple rounds of pre-compression on to-be-compressed texture data required by a client, so that compression parameters can be dynamically adjusted according to the effect of the multiple rounds of pre-compression to determine target compression parameters most suitable for a current scene, and the compression efficiency is improved. And the to-be-compressed data can be compressed according to the target compression parameter, so that the compression effect of the to-be-compressed data can be effectively improved.

The invention relates to the technical field of data compression, in particular to a data compression system based on shift processing, the system comprises a GPU, a database, a processor and a memory in which a computer program is stored, the database comprises to-be-compressed data, the GPU comprises a plurality of compression units, the to-be-compressed data is compressed through the compression units contained in the GPU, and the to-be-compressed data is compressed through the processor. According to the method, compression operation can be executed on the GPU side, then the method can be applied to intermediate data of the artificial intelligence model, the overall compression rate of the artificial intelligence model during training or reasoning can be effectively increased, and the compression effect is poor due to the fact that a compression mode based on lossless compression is affected by data distribution of data to be compressed. The compression effect of data compression can be improved by using a shift processing mode.

The invention relates to a preparation method and application of a biomass charcoal / PDMS composite material with piezoelectric characteristics, and the preparation method comprises the following steps: taking biomass as a raw material, carrying out gelation reaction at low temperature, and then carrying out bidirectional freezing treatment to obtain an ordered three-dimensional body structure of at least centimeter level; and carrying out freeze drying treatment on the obtained three-dimensional body structure to obtain the collagen aerogel. The collagen aerogel is subjected to high-temperature treatment in a protective atmosphere, biomass carbon aerogel is obtained, and the treatment temperature is 800-1600 DEG C; dissolving a PDMS prepolymer in an ethyl acetate solvent to form a PDMS prepolymer solution, wherein the volume ratio of the PDMS prepolymer to the ethyl acetate solvent is 1: (0.5-10); and completely immersing the obtained biomass charcoal aerogel in the PDMS prepolymer solution, dipping for 10-60 minutes, then putting into a drying oven, and carrying out curing reaction at the temperature of 50-80 DEG C, so as to finally obtain the biomass charcoal / PDMS composite material. According to the composite material, the piezoelectric sensitivity and the response speed of the composite material are improved through interface interaction and a stress transfer mechanism.

This utility model relates to the field of machining technology and discloses an environmentally friendly chip collection and processing device for valve stem machining. It includes a base, a central seat fixedly mounted on the upper end of the base, and a collection and compression mechanism mounted on the upper end of the base. By compressing the chips and detecting their volume, timely discharge is achieved to ensure efficient collection and discharge of the chips. This environmentally friendly chip collection and processing device for valve stem machining, through the inclusion of a collection and compression mechanism, and the alignment of the storage port and feed port on the rotating disc, can stably collect chips generated during valve stem machining. A support plate temporarily supports the chips in the storage port. A central motor drives the rotating disc to rotate, and the meshing of the central gear and transmission gear, in conjunction with the support wheel, causes the support plate to be continuously struck and vibrated, resulting in a more compact accumulation of waste chips above the support plate, ensuring the continuity of the collection process.

The utility model belongs to the motorcycle clutch technical field, concretely relates to a motorcycle clutch cover assembly body, include: cover, the top centre of cover is provided with the axle cylinder, the top array of cover is provided with a plurality of riveting hole no.

The utility model discloses environment-friendly garbage compression equipment, and relates to the technical field of garbage compression, the environment-friendly garbage compression equipment comprises a compression box and a guide seat, the inner wall of the compression box is fixedly sleeved with a feeding pipe, two crushing rollers are arranged below the feeding pipe, the guide seat is fixedly arranged on the inner wall of the compression box, and the guide seat is fixedly arranged on the inner wall of the compression box. And each crushing roller is arranged above a guide seat, a garbage can is arranged on one side of the guide seat, an electric push rod is fixedly arranged at the bottom of the fixing plate, and an output shaft of the electric push rod is fixedly connected with a pressing plate. Through the arrangement of the two crushing rollers and the guide seat, the driving mechanism can drive the two crushing rollers to reversely rotate, the two crushing rollers can crush garbage, and the garbage is smoothly conveyed into the garbage can through the guide seat, so that the size of the garbage can be reduced, and the garbage can be guided; and the crushed garbage is prevented from drifting away, so that the garbage compression effect can be improved.

The invention belongs to the technical field of gaskets, and particularly relates to a dip-formed flange gasket and a preparation method thereof.The preparation method comprises the following steps that firstly, a flexible graphite plate is taken, cut according to a flange gasket template, cleaned and dried; 2, putting the flexible graphite plate treated in the step 1 into a high-pressure impregnation kettle, heating, vacuumizing to negative pressure, adding impregnation liquid, and reacting until the vacuum degree in the high-pressure impregnation kettle is stable; 3, the vacuum state in the high-pressure impregnation kettle is removed, the pressure in the high-pressure impregnation kettle is increased to 0.85-1.25 MPa, and then heating reaction is conducted; and 4, repeating the step 2 and the step 3 for dipping after the reaction is finished, taking out the sample after dipping, and cleaning and drying to obtain the flange gasket. According to the method, more polytetrafluoroethylene can be immersed into the flexible graphite plate, so that the mechanical property and the leakage-proof property of the material are greatly improved.

The invention relates to the technical field of computers, in particular to an efficient lossless data compression method and system based on parallel computing, and the method comprises the following steps: calling a JNI mechanism of a bottom layer compression library function; integrating a parallel computing optimization algorithm; the memory data compression rate is improved; the API compatibility is ensured; testing and verifying; the efficient lossless data compression method and the efficient lossless data compression system based on parallel computing have the beneficial effects that the performance of a Java platform is improved by integrating a parallel computing optimization algorithm and transforming a compression library. Through actual testing, under the same compression task, the compression performance is improved by 5-10 times compared with the original Java class library, so that the data processing efficiency is greatly improved.

The invention belongs to the technical field of high polymer materials, and relates to a low-pressure-change, low-precipitation and high-weather-resistance bio-based extrusion grade TPV composite material as well as a preparation method and application thereof. According to the invention, the bio-based EPDM is used as a matrix material, the usage amount of a petroleum-based material is reduced, so that carbon emission is reduced, the realization of a carbon neutralization target is facilitated, and the environment-friendly material not only conforms to the concept of sustainable development, but also can reduce the influence on the environment on the premise of not influencing the performance. A cross-linked network structure formed by a phenolic aldehyde vulcanization system is more compact and uniform, the possibility of precipitation is reduced, phenolic aldehyde resin and a rubber phase have good compatibility, the whole system is more stable, and the prepared bio-based extrusion grade TPV material generates less die material slag accumulation in the extrusion process, so that the extrusion quality of the TPV material is improved. Therefore, the production efficiency and the appearance quality of the material are improved.

A rotary compressor with a low-pressure chamber and an air conditioner are disclosed. The rotary compressor with a low-pressure chamber includes a housing (100), a motor assembly and a pump assembly. The housing (100) is provided with a low-pressure air inlet component (120) and a high-pressure air outlet component (130), a low-pressure chamber (110) is arranged in the housing (100), the motor assembly is arranged in the low-pressure chamber (110). The motor assembly includes a stator (231) and a rotor (232). The pump assembly includes a crankshaft (210), a crankshaft shell (220), a cylinder (310), a piston (340), a sliding vane (330) and a bearing (320), the pump assembly is arranged in the low-pressure chamber (110). The piston (340), the sliding vane (330), the cylinder (310), the bearing (320) and the crankshaft shell (220) cooperatively form a compression chamber. A low-pressure refrigerant directly cools the rotor (232) and the stator (231), and the low-pressure refrigerant is heated to vaporize, so as to increase a temperature of a gaseous refrigerant before compression. The cylinder (310), the bearing (320) and the sliding vane (330) arranged in the low-pressure chamber (110) are fully cooled to minimize thermal expansion and deformation. The piston (340) and the crankshaft (210) are arranged in the cylinder (310), so that internal heat cannot be effectively dissipated in time, obtaining a large thermal expansion and deformation, and thus effectively strengthening the sealing performance between the cylinder (310) and the piston (340) and improving the compression effect on the refrigerant.