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

14results about How to "Reliable generation" patented technology

Game generation method and device, storage medium and computer equipment

The invention discloses a game generation method and device, a storage medium and computer equipment. The game generation method comprises the steps of obtaining a game development demand based on natural language interaction; generating game content data based on the game development demand; and driving a game engine based on the game content data to obtain a target game. By means of the method, the game content data can be automatically generated according to the game development requirements, the target game capable of driving the game engine to run is finally generated, efficient and high-quality game generation is achieved, the game development efficiency is improved, the game development threshold is lowered, the game development requirements of more developers are met, and the game development efficiency is improved. The illusion problem occurring in the game code generation task is relieved, the game code capable of well running is generated, and the adverse effect of redundant codes existing in the generated game on the game performance is avoided.
Owner:SHENZHEN AOTUOGAIMU TECHNOLOGY CO LTD

Distributed self-increasing sequence synchronization method and device, equipment and storage medium

The embodiment of the invention discloses a distributed self-increasing sequence synchronization method and device, equipment and a storage medium. The method comprises the steps that in a distributed database cluster, a master node and a plurality of slave nodes are arranged, and the master node is used for maintaining a global self-increasing sequence value; the slave node requests to distribute a local self-increasing sequence to the master node according to a preset cache parameter, and the master node calculates an initial value of the local self-increasing sequence according to the global self-increasing sequence value and distributes a preset number of continuous self-increasing sequence values to the slave node; the slave node caches the local auto-increment sequence to the local; and when the local self-increasing sequence cached by the slave node is used up, sending a new allocation request to the master node, and reallocating a new local self-increasing sequence by the master node according to the current global self-increasing sequence value. According to the method, the master node and the slave nodes are arranged in the distributed database cluster environment, and the global self-increasing sequence value is uniformly managed by the master node, so that reliable generation of the self-increasing sequence in the distributed environment is realized.
Owner:TIANJIN SHENZHOU GENERAL DATA TECH CO LTD

Pathological image classification method and system based on online pseudo supervision and dynamic mutual learning

The invention discloses a pathological image classification method and system based on online pseudo supervision and dynamic mutual learning, and the method comprises the steps: designing a self-adaptive storage library AMB to collect a component part with the most information content in a current WSI, and formulating a self-progressive feature fusion SPFF module, so as to achieve the discrimination and compact representation; integrating label related historical information and instant semantic change in the AMB to enhance the current pseudo packet mark representation; designing a decision correction false label DRPL generation module to explore the internal relation between the false packet representation and the packet label prediction so as to generate a more reliable false label; in order to relieve redundant and fuzzy representation, category priori of pseudo tag prediction is borrowed to promote tag-related feature learning and update AMB, so that dynamic mutual learning and optimization between feature representation and pseudo tag generation are formed; a dynamic decision DDM module is developed to coordinate dominant and implicit representations of packet information, thereby achieving a more robust decision.
Owner:HARBIN INST OF TECH

Industry report generation method and system based on deep learning

The present application provides an industry report generation method and system based on deep learning, which uses natural language processing technology to analyze the text quality of industry texts to help identify better original data. In the adjustment phase of the natural language processing network, a collaborative training task of implicit representation reasoning is introduced. Based on this, in the process of text quality reasoning learning, the natural language processing network is made to obtain a richer implicit text representation of each text paragraph based on implicit representation reasoning learning, which helps alleviate the high variance problem and early stopping phenomenon caused by over-training of the quality segmentation network. In this way, the quality recognition ability and universality of the neural network for industry texts are increased, which helps to determine high-quality industry texts in massive industry texts, so as to generate accurate, reliable and high-quality industry reports.
Owner:JIERUAN TECH (GRP) CO LTD

Electroencephalogram head ring interaction control method and device based on concentration level

The invention discloses an electroencephalogram head ring interaction control method and device based on concentration level, and the method comprises the steps: collecting multi-band electroencephalogram signals and user interaction parameters, carrying out spectrum decoupling analysis to recognize a concentration feature baseline, exporting an attention stability coefficient, and building a cognitive state mapping relation; tracking a high-response frequency band core to implement frequency band fusion analysis to form a control channel, and collecting concentration fluctuation data to construct an interaction control network; executing multi-task response analysis to identify an intention switching point, extracting cognitive load parameters to divide a low-load area and a high-load area, and extracting a recovery rate to construct an attention toughness index in an identification conversion process; determining an optimal trigger position in a low-load area according to an attention toughness index, constructing a brain wave feature chain, aligning the brain wave feature chain with an attention stability coefficient, determining a low-interference response window, and setting a self-adaptive threshold group; and the coupling identification false triggering area extracts redundant brain wave components to execute false triggering suppression to generate an interaction control instruction, so that accurate matching between teaching intervention and the cognitive state of the student is realized.
Owner:SHANXI HANGYI BIOTECHNOLOGY CO LTD

Sliding component

PendingEP4394217A4positive pressureImprove the lubrication effectEngine sealsShafts
There are provided sliding parts in which sliding surfaces separate from each other for relative rotations in both directions, and which has excellent lubricity. In sliding parts which is disposed at a relative rotational location of a rotating machine, and in which a first dynamic pressure generation groove 14 extending so as to be inclined with respect to a relative rotation direction and a second dynamic pressure generation groove 15 extending so as to be inclined in a direction opposite to the inclination direction of the first dynamic pressure generation groove 14 with respect to the relative rotation direction in a plan view are provided on a sliding surface 11, bottom surfaces 14a and 15a of the first dynamic pressure generation groove 14 and the second dynamic pressure generation groove 15 are inclined in the same direction with respect to a radial direction in a cross-sectional view.
Owner:EAGLE INDS

Method and device for generating character interpolation image, computer-readable storage medium, and terminal

A method and device for generating a person interpolation image, a computer-readable storage medium, and a terminal, the method comprising: obtaining a pixel matrix of a first image and a pixel matrix of a second image, and splicing them to obtain a first splicing matrix, wherein the first image and the second image contain the same target person; predicting the human body prediction features of the target person according to the first splicing matrix, and performing feature extraction on the first splicing matrix to obtain image features, wherein the human body prediction features contain the predicted human body information of the target person at the time when the person interpolation image belongs; performing feature fusion on the human body prediction features of the target person and the image features to obtain fusion features; and decoding the fusion features to obtain the person interpolation image. The above scheme is helpful to obtain a person interpolation image with higher definition and better quality.
Owner:HUA DATA TECH (SHANGHAI) CO LTD

Battery pack, power equipment and battery pack state diagnosis method

PendingCN118970245Areliable generationSecondary cellsElectrical batteryElectric power equipment
The embodiment of the invention relates to the technical field of batteries, and discloses a battery pack, power equipment and a battery pack state diagnosis method. The battery pack comprises a battery cell module and a battery management system, the battery cell module comprises N battery cells which are sequentially connected in series, the N battery cells comprise at least one A-type battery cell, the A-type battery cell comprises a current interruption device, the A-type battery cell is the ith battery cell, and i is larger than or equal to 2 and smaller than or equal to N. The battery management system comprises a battery monitoring unit and a power circuit. The power supply circuit comprises a first power supply circuit and a second power supply circuit, and the power supply circuit is configured to be one of the first power supply circuit and the second power supply circuit to supply power to the battery monitoring unit.
Owner:XIAMEN AMPACK TECH LTD

Spacecraft module-level code automatic generation method based on artificial intelligence

The invention discloses a spacecraft module-level code automatic generation method based on artificial intelligence. The method comprises the following steps: constructing a multi-level thinking map; constructing a multi-source spaceflight domain knowledge base, and performing domain specialization enhancement on the large language model according to the multi-source spaceflight domain knowledge base; generating an intelligent module-level code according to the multi-level thinking map; carrying out multi-dimensional code quality evaluation on the intelligent module-level code to generate a quality problem report, and establishing an intelligent feedback mechanism based on a graph structure according to the quality problem report and the multi-level thinking graph; and inputting an intelligent feedback mechanism based on a graph structure into the field specialized and enhanced large language model, and outputting a module-level software code. The problems that in the prior art, the complex requirement processing capacity is limited, the field specialization degree is insufficient, code quality lacks a dynamic optimization mechanism, and cooperative management between modules is difficult are solved, and efficient and reliable code generation is achieved.
Owner:BEIJING INST OF ASTRONAUTICAL SYST ENG

A method and apparatus for generating a denture insertion connection structure

ActiveCN116898606BAvoid self-intersection situationsPrecise offsetArtificial teethMechanical engineeringStructure generation
The application discloses a kind of denture insertion connection structure generation method and device, wherein the method includes: based on denture and denture plate model generation bottomless denture model grid, and the edge of this denture model grid is optimized, and optimized denture model grid is obtained;The two-dimensional discrete boundary curve generated by the model grid is inward equidistant, and the bottom boundary curve is obtained;Based on the curve, the first connection structure bottom surface grid generated by the model grid is cut, and the second connection structure bottom surface grid is obtained;The equidistance and edge deformation of the bottom surface grid are carried out, and the third connection structure bottom surface grid is obtained;The boundary vertex of the bottom surface grid and the boundary vertex of the model grid are subjected to CDT processing, and the connection structure side surface grid is obtained;The side surface grid and its junction are subjected to smoothing processing, and the denture insertion connection structure is obtained.The method and device of the application can solve the problem that the connection structure may exist in the insertion direction.
Owner:SHENZHEN UP3D TECH CO LTD

Retrieval-driven workflow generation method and system for domain knowledge enhancement

The invention discloses a domain knowledge enhanced retrieval-driven workflow generation method and system, and the method comprises the steps: receiving a natural language demand of a user, and retrieving a high-correlation candidate workflow template from a pre-constructed domain structured knowledge base through a mixed strategy of fusing sparse retrieval and dense retrieval; systematically extracting task step definitions, dependency graphs, API (Application Program Interface) calling specifications and domain constraint conditions in the candidate templates to form a structured knowledge element set; deeply fusing the knowledge set with user requirements, and constructing a structured cue word containing task description, knowledge reference and generation constraint; and generating a high-quality executable workflow code meeting the standard of the target platform by utilizing the large model. According to the method, the problems of poor code normalization caused by lack of domain knowledge, low retrieval correlation caused by a single retrieval strategy, insufficient flexibility caused by a hard coding template and the like in an existing method are effectively solved, and the accuracy, the normalization and the practicability of workflow code generation are remarkably improved.
Owner:SHANHU INTELLIGENT COMPUTING TECHNOLOGY (HUNAN) CO LTD

Internet of Things terminal equipment security identity authentication method

The invention belongs to the field of identity authentication, and particularly relates to an internet of things terminal equipment security identity authentication method, which comprises the steps of constructing an electronic authentication key module, generating an OTP (One Time Programmable) key, performing a first-stage authentication stage, performing a second-stage authentication stage and performing communication evaluation. According to the multi-party key fusion method based on the one-time password book protocol, it is ensured that even if a system is damaged, it is difficult for a third party to deduce confidential information, and the method has high practicability and confidentiality in various mixture networking scenes and Internet of Things terminal equipment with limited processing capacity; according to the double-layer identity authentication method based on the electronic key, the electronic key is automatically generated and managed, user intervention is not needed, double-factor authentication is achieved, the validity of the key is verified by introducing a hash function and random position selection, error correction and privacy protection of the key are achieved, and it is ensured that no error or malicious tampering exists in the calculation process.
Owner:JIANGSU ZHONGSHI ELECTRONIC TECH CO LTD

Cutter code generation system and method

The invention discloses a tool code generation system and method, relates to the technical field of tool coding, and aims to solve the problem that one object has multiple codes or multiple objects have one code in the existing tool code generation process. Inputting the initial coded data to generate primary input information; the auditing and judging module is used for auditing and judging the content and the coding digits of the first-level input information and generating second-level auditing information; the automatic verification module is used for automatically verifying the secondary auditing information, carrying out secondary matching on the cutter article according to the image identification information and an AI database, and determining a matching result; and the duplicate removal module is used for removing duplicate of the first-level input information and judging whether the first-level input information is repeated. The invention further comprises a corresponding method, the entered code information can be audited and automatically verified, the conditions that one object has multiple codes and multiple objects have one code are effectively prevented, and efficient and accurate management and control of the tool are achieved.
Owner:HANGZHOU STEAM TURBINE

A method for predicting human eye scanning trajectory for panoramic images

The present invention proposes a method for predicting human eye scanning trajectory for panoramic images, namely, a three-dimensional visual scanning trajectory Transformer (3D Visual Scanpath Transformer, Pathformer3D). First, a two-dimensional panoramic image is mapped to a three-dimensional spherical coordinate system, and a three-dimensional convolutional network is used to extract its features; then, a 3D Transformer encoder is used to convert the features of each image area in the spherical coordinate system into a context-aware representation; then, the context-aware representation of the learned feature information, all historical gaze point query information, and the query information of the current gaze point are input into the 3D Transformer decoder together to calculate the hidden information of the gaze point at the current moment; finally, the hidden information is input into a three-dimensional mixed density network to learn the joint three-dimensional Gaussian distribution of the position of the gaze point at the current moment in the three-dimensional spherical coordinate system, and the final focus point position is sampled therefrom. The Pathformer3D model proposed in the present invention can generate a more realistic and reliable panoramic image scanning trajectory for 360° panoramic images.
Owner:NANJING UNIV OF AERONAUTICS & ASTRONAUTICS