Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

23results about How to "Improve conversion performance" patented technology

Image modal conversion method and system based on multi-scale cross-modal alignment network

PendingCN118982735AImprove conversion performanceAccurately capture and retainBiological modelsImaging processingImaging Feature
The invention discloses an image modal conversion method and system based on a multi-scale cross-modal alignment network, and relates to the technical field of medical image processing, and the method comprises the steps: obtaining a medical image, and carrying out the preprocessing of the medical image; extracting preprocessed medical image features by using a deep learning network, and introducing a multi-attention mechanism to perform feature fusion; and reconstructing the medical image based on the fused features, and carrying out optimization processing on the reconstructed medical image. According to the method, by fusing multi-scale feature extraction, cross-modal alignment, image decoding path information and a loss function, the conversion efficiency of the medical image is remarkably enhanced, interference of artifacts and noise is effectively reduced, and then detail features and global structure information in the medical image are more accurately captured and reserved.
Owner:THE FIRST MEDICAL CENT CHINESE PLA GENERAL HOSPITAL

Heat pipe heat transfer type nuclear battery and application thereof

PendingCN119008067ASolve the problem of layout limitationsFlexible design structureRadiation electrical energyRadio isotopesNuclear engineering
The invention discloses a heat pipe heat transfer type nuclear battery and application thereof, and relates to the technical field of radioactive isotope batteries, the heat pipe heat transfer type nuclear battery comprises a radioactive heat source, a heat preservation layer and a shielding layer are sequentially arranged outside the radioactive heat source, a high-temperature heat pipe is connected to the radioactive heat source, and the high-temperature heat pipe penetrates through the heat preservation layer and the shielding layer to be connected with a thermoelectric device. The thermoelectric device is connected with a low-temperature heat pipe which is connected with a radiator. According to the heat pipe heat transfer type nuclear battery, the thermoelectric device is arranged outside the shielding layer, so that the device is prevented from being damaged by radioactive irradiation; meanwhile, the heat of the radioactive heat source is directly conducted to the thermoelectric device by using the excellent heat-conducting property of the heat pipe, so that the heat utilization rate of the heat source is improved, and the overall thermoelectric conversion efficiency of the nuclear battery is improved by establishing a relatively large temperature difference between two ends of the thermoelectric device; besides, by utilizing the characteristics that the length and the end part size of the heat pipe are adjustable in a certain range, the shape is bendable and the like, the problem of limited arrangement of nuclear battery components is solved, so that the position arrangement and the structural design of the components such as a radioactive heat source, a thermoelectric device and a radiator are more flexible.
Owner:NUCLEAR POWER INSTITUTE OF CHINA

Photovoltaic panel frame structure facilitating angle adjustment of photovoltaic panel

The utility model relates to the technical field of photovoltaic panel frames, and discloses a photovoltaic panel frame structure convenient for adjusting the angle of a photovoltaic panel, which comprises a base and a main frame body, one side of the top surface of the base is fixedly connected with two first support rods, the other side of the top surface of the base is fixedly connected with two second support rods, the second support rods are internally provided with empty grooves, and the empty grooves are arranged in the main frame body. A lifting screw rod is arranged in the second supporting rod, a lifting assembly used for adjusting the angle of the main frame body is arranged at the bottom of the lifting screw rod, hinge pieces are movably connected to the top ends of the first supporting rod and the lifting screw rod, the hinge pieces are fixedly connected with the main frame body, and clamping assemblies for fixing the photovoltaic panel are symmetrically fixed to the two sides of the main frame body. Therefore, the angle of the photovoltaic panel can be adjusted, so that the photovoltaic panel can better receive sunlight, the loss of sunlight energy is reduced, the energy conversion effect is further improved, and the power generation efficiency is also improved.
Owner:QINGDAO SHUANGCHENG ENERGY TECH CO LTD

Voice style conversion method and device, equipment and storage medium

PendingCN119559955AImprove conversion performanceachieve conversionSpeech analysisFeature extractionEngineering
The invention discloses a voice style conversion method, and the method comprises the steps: obtaining a voice data set which comprises a source voice and a target voice; performing feature extraction on the source voice and the target voice to obtain a first Mel spectrum and a second Mel spectrum; performing content feature extraction on the first Mel spectrum through a preset content encoder to obtain content features; performing down-sampling processing on the phoneme sequence aligned with the content features to obtain a hidden state sequence; performing style feature extraction on the second Mel spectrum through a preset style encoder to obtain style features; performing alignment embedding on the style features and the hidden state sequence to obtain target style features; and determining a voice conversion result according to the hidden state sequence and the target style feature. According to the method, the phoneme sequence aligned with the content features is subjected to down-sampling, so that the problem that noise frames exist is solved. By performing alignment embedding on the style characteristics and the hidden state sequence, other styles except timbre are effectively transferred.
Owner:PING AN TECH (SHENZHEN) CO LTD

Image processing method and device, nonvolatile storage medium and computer equipment

The invention discloses an image processing method and device, a nonvolatile storage medium and computer equipment. The method comprises the following steps: acquiring a to-be-processed image and a target image style; determining an original image style of the to-be-processed image according to the pixel color in the to-be-processed image; determining a first conversion strategy corresponding to the original image style and a second conversion strategy corresponding to the target image style; based on the first conversion strategy, processing the to-be-processed image to obtain a restored image of which the original image style is removed; and processing the restored image based on the second conversion strategy to obtain a target image matched with the target image in style. The technical problem that the effect is poor when style conversion is carried out by adopting a pixel color fusion method in the prior art is solved.
Owner:CHINA TELECOM BESTPAY CO LTD

Musical instrument timbre conversion model construction method and musical instrument timbre conversion method

ActiveCN114550735BImprove conversion performanceImprove training efficiencySpeech analysisEngineeringTimbre
An embodiment of the present invention provides a method for constructing a musical instrument timbre conversion model, comprising: converting a first sample audio vector sequence into a second sample audio vector sequence, and then converting the second sample audio vector sequence into an input sample audio vector sequence; and updating the model parameters of the model to be trained through the calculated first loss value and the first score, so as to train the musical instrument timbre conversion model; the embodiment of the present invention trains the model through the first loss value and the first branch, and the training efficiency is high; and the accuracy of the musical instrument timbre conversion model trained by the above method is improved.
Owner:PING AN TECH (SHENZHEN) CO LTD

A method and device for converting SAR images into optical images

ActiveCN115100091BImprove conversion performanceImage enhancementImage analysisImage resolutionComputational physics
The present application relates to a method and device for converting SAR images into optical images, the method comprising: obtaining a SAR image and a corresponding optical image to form a training set; inputting the SAR image and the optical image in the training set into a CFR-GAN model, and outputting a discrimination result, wherein the CFR-GAN model comprises a generation network and a discrimination network, and the generation network comprises an encoder, a cross-fusion inference structure, and a decoder; inputting the optical image and the discrimination result into a pre-constructed total loss function, training the model, and obtaining a CFR-GAN optimization model; inputting the SAR image to be processed into the CFR-GAN optimization model for conversion, and obtaining an optical conversion image. The present method can utilize the cross-fusion inference structure in the CFR-GAN optimization model to realize the conversion of SAR images of different scales to optical images at different resolutions, retain the feature information of each scale of the SAR image, and thus improve the conversion performance of the SAR image to the optical image.
Owner:NAT UNIV OF DEFENSE TECH

VR user identity authentication method based on multi-modal perception of vocalization vibration signals

ActiveCN121260163BImprove authentication capabilitiesimprove securityFrequency spectrumFeature extraction
The application relates to a VR user identity authentication method based on multi-modal perception of sound vibration signals, which comprises the following steps: S1, obtaining original acceleration signals and original speech signals based on time axis alignment; S2, identifying sound segments and taking the sound segments as first speech signals; S3, taking first acceleration signals corresponding to the first speech signals from the original acceleration signals; S4, generating acceleration amplitude spectrum diagrams of the first acceleration signals as first acceleration spectrum diagrams; S5, converting the first acceleration spectrum diagrams into speech mel spectrum diagrams as second mel spectrum diagrams, and converting the second mel spectrum diagrams into acceleration amplitude spectrum diagrams as second acceleration spectrum diagrams; S6, performing cyclic consistency feature extraction on the first acceleration spectrum diagrams and the second acceleration spectrum diagrams to obtain first features; and S7, inputting the first features into a trained first detection model to obtain a detection result. Compared with the prior art, the application has the advantages of improving the anti-interference ability and the accuracy, etc.
Owner:SOUTHEAST UNIV

High temperature geothermal conversion heat energy release device for middle-deep geothermal energy

This utility model relates to the field of geothermal energy utilization technology, and in particular to a high-temperature geothermal energy conversion and heat release device for medium-deep geothermal energy. It includes a heat-conducting sleeve, heat conduction pipes, and a confluence pipe. Multiple sets of heat conduction pipes are arranged in an array inside the heat-conducting sleeve. Multiple sets of ribs are fixedly installed around the heat conduction pipes and are fixedly connected to the heat-conducting sleeve. The heat-conducting sleeve is filled with thermally conductive gel. One end of each set of heat conduction pipes is connected to the confluence pipe. An evaporation chamber is provided inside the heat-conducting sleeve, and the confluence pipe communicates with the evaporation chamber. This utility model increases the contact area through the ribs, and in conjunction with the thermally conductive gel inside the heat-conducting sleeve, it can absorb geothermal energy and transfer it to the heat conduction pipes, heating the circulating fresh water inside the heat conduction pipes to extract geothermal energy. The heated hot water from the multiple sets of heat conduction pipes is then discharged into the evaporation chamber through the confluence pipe, evaporated into high-temperature steam, and then delivered to the energy conversion component for geothermal utilization.
Owner:西安市安居新能源发展有限公司

A Char type condition judgment method and system based on heterogeneous architecture

The present invention provides a Char type conditional judgment method and system based on heterogeneous architecture, including: the CPU of the server sends original data and parameters to the acceleration card, stores the original data and parameters in the acceleration card storage unit, and starts the acceleration core; the acceleration core reads and parses the parameters and performs initialization operations; the acceleration core reads the original data, and performs conditional judgment on the original data according to the parameters to obtain result data; the result data is stored in the storage unit and sent to the server CPU to realize the conditional judgment function of the char type under the heterogeneous architecture. The Char type conditional judgment method based on heterogeneous architecture provided by the present invention designs the conditional judgment function of the Char type in the heterogeneous system, greatly improves the data type conversion capability, offloads the task to the heterogeneous system for processing, reduces the CPU occupancy rate, and improves the CPU performance.
Owner:YUSUR TECH CO LTD

Anonymous voice conversion method and device, computer equipment and storage medium

ActiveCN119446162Aimprove securityImprove decoupling abilitySpeech synthesisData setEngineering
The invention belongs to the field of artificial intelligence and finance, and relates to an anonymous voice conversion method, which comprises the following steps: preprocessing a voice data set to obtain a voice enhancement data set; performing first-stage training on a pre-constructed initial speech synthesis model by using the speech enhancement data set to obtain a trained speech synthesis model; performing second-stage training on the pre-constructed initial anonymous identity generator to obtain a final anonymous identity generator; replacing a speaker encoder in the speech synthesis model with the final anonymous identity generator to obtain an anonymous speech conversion model; and performing voice conversion on the to-be-converted voice based on the anonymous voice conversion model to generate anonymous voice. The invention further provides an anonymous voice conversion device, computer equipment and a storage medium. In addition, the invention also relates to a block chain technology, and the to-be-converted voice can be stored in a block chain. According to the invention, the diversity of anonymous voice can be improved, and privacy security is protected.
Owner:PING AN TECH (SHENZHEN) CO LTD

A flexible thermoelectric device based on negative Poisson's ratio structure and its application

ActiveCN116887656BImprove conversion performancegood flexibilityMedical equipmentInsulation layer
The present invention relates to the technical field of electronic devices, and in particular to a flexible thermoelectric device based on a negative Poisson's ratio structure and its application. A flexible thermoelectric device based on a negative Poisson's ratio structure includes a heat-conducting layer for fitting with human skin; a hot end installed at one end of the top of the heat-conducting layer; a heat-insulating layer installed on the top of the heat-conducting layer and abutting the hot end at one end; a cold end installed on the top of the heat-insulating layer and located away from the hot end; a thermoelectric conversion layer installed on the top of the heat-insulating layer, one end connected to the hot end and the other end connected to the cold end; and a heat-insulating packaging layer covering the top of the thermoelectric conversion layer; the thermoelectric conversion layer includes a plurality of thermoelectric units arranged in an array and interconnected; the thermoelectric unit is a negative Poisson's ratio structure. The flexible thermoelectric device based on a negative Poisson's ratio structure provided by the present invention has good energy conversion performance, good flexibility, good bending strength, and long service life, and is suitable for use in wearable devices, especially wearable medical devices.
Owner:DONGGUAN UNIV OF TECH

Face feature conversion method based on feature information mapping and attention mechanism

The invention discloses a face feature conversion method based on feature information mapping and an attention mechanism. The method comprises the following steps: 1) constructing a face feature editing model; the face feature editing model comprises a generator, a discriminator and a loss network; 2) taking a to-be-converted target domain label and an original image as input of a model, and outputting a converted generated image; firstly, a target domain label is converted into data of depth feature information by a feature information mapping network in a generator, then the data is added into an image generation network to realize conversion in the process of converting an original image into a generated image, and finally the generated image enters a discriminator to be evaluated, and the true degree of the generated image is evaluated. According to the method, the structure of the generator and the structure of the discriminator are modified, and a loop structure in the model is constructed, so that correct generation of the image is facilitated; and by adding an attention mask generation network and context loss, the performance of retaining the original information of the image is improved.
Owner:NANCHANG UNIV

Double-path traveling wave ion migration tube device for ion mobility spectrometry instrument

The invention belongs to the technical field of mass spectrum and ion mobility spectrum analysis, and particularly relates to a double-path traveling wave ion mobility tube device for an ion mobility spectrometer. The main component is a printed circuit board, and the principle is that an electric field for pushing ions to move is formed by using electrodes drawn on the assembled circuit board after a traveling wave signal is accessed, and the ions are separated according to the collision sectional area of the ions and the difference of the charge quantity carried by the ions. Compared with the traditional design, the traveling wave ion migration tube disclosed by the invention adopts a double-path traveling wave array design, and two groups of traveling wave electric signals can be simultaneously input into a single group of channels to adjust an electric field. Therefore, the power supply is more flexible, and the performance requirement of an external power supply is lower.
Owner:HUNAN UNIV

Error correction method and device for converting medical voice into text, equipment and medium

The invention relates to the technical field of artificial intelligence, and discloses a medical voice-to-text error correction method, device and equipment and a medium, and the method comprises the steps: constructing a multi-modal medical voice model, initializing the multi-modal medical voice model according to the parameters of a visual encoder and a text encoder of a visual language pre-training model, and carrying out the pre-training of the multi-modal medical voice model; receiving the paired medical image and the text report after error simulation to train a multi-modal medical voice model; constructing a medical voice-to-text error correction model according to the voice-to-text model and the trained multi-mode medical voice model, and receiving to-be-corrected voice and a medical image corresponding to the voice to train the medical voice-to-text error correction model; and performing medical voice-to-text error correction on the input medical voice by adopting the trained medical voice-to-text error correction model. And the medical voice-to-text conversion effect is improved.
Owner:PING AN TECH (SHENZHEN) CO LTD

Folding solar cell support

The utility model discloses a foldable solar cell support, which relates to the technical field of solar cells and comprises a base, a first motor is fixedly mounted in the middle of the bottom of the base, the output end of the first motor penetrates through the base and is fixedly provided with a top seat, and the top of the top seat is fixedly provided with an adjusting and moving mechanism. And a folding mechanism is fixedly mounted at the top of the adjusting and moving mechanism. By adopting the structure, the second motor is used for driving the screw rod to drive the sliding block to slide mutually, the sliding block can be used for assisting in driving the telescopic support to move left and right, and the photovoltaic panel is hinged to the upper end of the telescopic support through the hinge seat and the inner side of the photovoltaic panel is hinged to the inner hinge seat through the left-right displacement of the telescopic support. At the moment, the hinge seats can be mutually closed, so that the photovoltaic panel is folded, the device does not need to be manually adjusted in the use period, the whole device has an automatic folding function, and the use convenience of the whole device can be improved.
Owner:NANTONG TONGXUAN MASCH CO LTD

Data weight averaging method based on double-capacitor array and related device

PendingCN119232169Asimple methodImprove conversion performanceAnalogue conversionCapacitanceControl theory
The invention discloses a data weight averaging method based on a double-capacitor array and a related device, and belongs to the technical field of modulators, and the method comprises the steps: obtaining an input digital signal of a digital-to-analog converter; splitting an input digital signal into two split digital signals, wherein the absolute value of the difference value of the two split digital signals is not more than 1; and respectively carrying out first-order dynamic element matching on the two split digital signals on the two independent capacitor arrays to obtain two matched capacitor arrays, and completing data weight averaging based on the double capacitor arrays. The method is simple in implementation logic, and meanwhile, the random depth of dynamic element matching can be effectively improved.
Owner:CHINA ELECTRIC POWER RESEARCH INSTITUTE CO LTD +3

An AI-driven low-code platform application automatic generation method and system

PendingCN122261560ASufficient domain knowledge supportprevent overflowBiological modelsKnowledge representationCode generationAlgorithm
The application belongs to the technical field of low-code development, and discloses an AI-driven low-code platform application automatic generation method and system, which comprises the following steps: constructing a style mapping table; loading an entry skill of an AI Agent, obtaining a low-code platform data model hierarchical structure and a skill navigation table, querying the navigation table according to a task type to load an analyzer skill, and obtaining an input source analysis rule; extracting an HTML prototype structure feature vector and performing similarity matching with a UI mode template library, generating a preliminary configuration within a threshold value, and generating a basic configuration structure outside the threshold value according to the analysis rule, range determination, form entity and page differentiation, and four-step component conversion; matching and loading atomic skills, generating scattered configuration files and analysis summary files according to a JSON configuration specification; displaying a summary for confirmation, assembling scattered configurations to generate a seed data file; and importing the seed data file through an API protocol bridge layer. The application improves the code generation efficiency of the low-code development platform.
Owner:ZHIWEI XINGYI (SHANGHAI) INTELLIGENT TECH CO LTD

Water accumulation prevention lap joint type solar photovoltaic module

The invention, which relates to the technical field of solar energy, discloses an anti-ponding lap joint type solar photovoltaic module comprising a composite mechanism. According to the water accumulation prevention lap joint type solar photovoltaic module, the components are fixed through the composite mechanism design and the fixing mechanism, so that the effect of keeping the structure stable is achieved, shaking of the internal structure caused by external force influence is avoided, component abrasion is reduced, the service life of equipment is prolonged, the photovoltaic module is protected by the composite shell, external force damage is avoided, and the service life of the equipment is prolonged. The transparent cover plate is arranged on the light receiving side of the composite shell, so that rain wash or external impact is prevented, a light source is prevented from being shielded, and the conversion efficiency is prevented from being reduced; the sliding block is controlled to slide through the linear module, so that the friction plate rubs on the surface of the transparent cover plate, and on one hand, water flow is conveniently drained from one side of the drainage mechanism; and on the other hand, water flow rubs the friction plate, so that the cleaning effect on the surface of the component is achieved, impurities are prevented from precipitating on the surface of the structure, and shielding of a light source is reduced.
Owner:润达光伏盐城有限公司

Text conversion method and device, storage medium and electronic equipment

The invention provides a text conversion method and device, a storage medium and electronic equipment. The method comprises the following steps: acquiring a target text feature corresponding to a first text sequence; obtaining the (i-1) th reference decoding feature output by the (i-1) th feature decoding layer, and obtaining the ith residual decoding feature according to the (i-1) th reference decoding feature; obtaining a first weight parameter corresponding to the (i-1) th reference decoding feature and a second weight parameter corresponding to the ith residual decoding feature, wherein the ratio of the first weight parameter to the second weight parameter is greater than or equal to a target threshold value; determining the ith reference decoding feature according to the first weight parameter, the second weight parameter, the (i-1) th reference decoding feature and the ith residual decoding feature; and under the condition that i is equal to N, determining a second text sequence matched with the first text sequence based on the ith reference decoding feature. The technical problem that the effect of a related text conversion method is poor is solved.
Owner:TENCENT TECHNOLOGY (SHENZHEN) CO LTD

Method, device and equipment for transferring picture to document and storage medium

The invention discloses a method, device and equipment for transferring a picture to a document and a storage medium, the method comprises the steps that a first picture is input to a target detection model to obtain at least one object, the target detection model is used for detecting objects of different object types in the picture, and the object types comprise an image type, a formula type and a table type; for each object in the at least one object, constructing a first corresponding relationship between the object and an object identifier corresponding to the object to obtain at least one first corresponding relationship; processing the first picture based on the at least one object identifier corresponding to the at least one object to obtain a second picture; identifying the text information in the second picture to obtain a first document; and determining a second document corresponding to the first picture based on the first document and the at least one first corresponding relationship. According to the method, the objects of the table type and the image type can be identified and converted into the document, omission of the objects of the table type and the image type is avoided, and the conversion effect is improved.
Owner:ANHUI ZHUOZHI EDUCATION TECH CO LTD

A facial gesture enhanced interactive method based on facial feature transfer

The present invention discloses a facial gesture enhancement interactive method for facial feature migration. First, the original data is processed to obtain frames and facial identification sketches. After resizing, the two types of images are merged to obtain preprocessed data. The processed data is submitted to a conditional generative adversarial network model for image prediction, and finally a migrated result image is generated. Pix2Pix applies the mapping function of the conditional generative adversarial network modeling conversion to convert an image from one domain to another domain to achieve feature migration. More precisely, Pix2Pix learns from facial markers and converts them into facial features, and detects a face from a network camera or video and places it on a trained face model in real time. The present invention is not only highly accurate, but also can improve image-to-image conversion performance.
Owner:NANJING UNIV OF FINANCE & ECONOMICS

Electromagnetic power generation and friction nanometer power generation composite wave energy conversion device

InactiveCN119543695AExpand the available wave rangeImprove conversion performanceMachines/enginesEngine componentsCopper electrodeMetal electrodes
The invention provides an electromagnetic power generation and friction nanometer power generation composite wave energy conversion device, and relates to the field of wave energy utilization, the electromagnetic power generation and friction nanometer power generation composite wave energy conversion device comprises an integrated power generation unit, the integrated power generation unit is formed by combining a plurality of power generation units, and each power generation unit comprises a square shell, a friction nanometer power generation part and an electromagnetic power generation part; the metal electrode clings to the inner wall of the square shell; a magnet rotor is arranged in the square shell, and the magnet rotor and a copper electrode in the square shell form a friction nanometer power generation part; a plurality of layers of coils are wound on the outer wall of the square shell, and the magnet rotor and the coils form an electromagnetic power generation part. Wave energy is collected through friction nanometer power generation and electromagnetic power generation, the two power generation modes are matched with each other, the available wave range can be effectively expanded, the conversion efficiency can be improved, the starting condition can be reduced, and therefore the device adapts to actual sea conditions with high randomness and time-varying characteristics.
Owner:DALIAN MARITIME UNIVERSITY +1