28results about How to "Improve build quality" patented technology

This invention relates to a method and system for automated generation of TSN network configurations based on a reflective agent. The system comprises a network state perception and intent reflection module, a YANG model configuration parameter and structure decision module, and a YANG model intelligent generation module. The network state perception and intent reflection module is responsible for structurally perceiving network state and resource information. The YANG model configuration parameter and structure decision module integrates multi-source knowledge and contextual information, intelligently generating deployable TSN scheduling parameters and YANG configuration structures based on business needs. The YANG model intelligent generation module automatically generates the corresponding YANG configuration model based on the optimized results, performs syntax and semantic verification, and then distributes it to the target device, achieving automated deployment of network configurations. This invention significantly improves the intelligence level and deployment automation capability of TSN network configuration by introducing reflective reasoning and feedback optimization techniques.

This invention relates to the field of decorative engineering technology, and discloses a method and system for generating virtual renderings of decorative engineering projects. The method includes: S1, parsing CAD drawings; S2, identifying professional intent; S3, defining and configuring key aesthetic elements; S4, attributing and correcting model parameters; S5, conflict detection; S6, adaptive fusion and correction; S7, material mapping and lighting calculation; and S8, outputting the virtual rendering. This invention solves the problem that the complexity of CAD drawings leads to incomplete or inaccurate geometric structures in 3D models, which in turn affects material mapping and lighting calculation, thus impacting the visual realism of virtual renderings.

The application discloses a radar target recognition method and device with azimuth enhancement under physical scattering constraints. The core is that a feature level azimuth expansion module (FAEM) is constructed, sparse view angle features are augmented in a feature space, and effective expansion of azimuth information is realized; meanwhile, a scattering center reconstruction constraint module (SCRCM) is innovatively introduced, physical authenticity and semantic fidelity of generated features are improved; in addition, multi-view perception enhancement modules (MVFM) are used to fuse generated multi-azimuth features, and intra-class compactness constraints are applied to optimize feature space structures, and the recognition ability of the model is further improved. The method realizes optimization from feature generation, physical constraints to semantic enhancement, and effectively improves the target recognition performance under sparse view angles.

This invention relates to the technical field of programming task code generation methods based on multi-agent collaboration and iterative reasoning. Specifically, it discloses a method for generating programming task code based on multi-agent collaboration and iterative reasoning, comprising: obtaining a task description with a primary-secondary functional dependency structure; generating initial code by a generating agent based on a dependency-guided thought chain and performing static structure verification; collecting runtime execution trajectories in an isolated sandbox through dynamic agents; having a diagnostic agent attribute errors to the trajectories, distinguishing between inter-function collaboration errors and intra-function logic errors, and generating structured repair guidance; and combining historical perception feedback and a controlled reset mechanism to guide subsequent iterative repairs and avoid local optima. This invention solves the problems of ambiguous error location, easy code degradation, and low repair efficiency in existing code generation methods for multi-functional dependency tasks, achieving accurate error attribution, preventing code rollback, and improving generation success rate and repair convergence efficiency.

The application discloses a ship body segmentation splicing method based on a laser point cloud device, and comprises the following steps: acquiring X partial point cloud data of a ship body segmentation; splicing the X partial point cloud data by using the spatial overlapping part and the same features of the point cloud data; deleting the point cloud data of a part which is not needed to be compared with a ship body segmentation design model, and reserving the ship body segmentation point cloud; generating a target point by the key position of the ship body segmentation point cloud data, performing coordinate conversion on the coordinates of the target point and the corresponding position of the ship body segmentation design model, converting the coordinate system of the ship body segmentation point cloud data into the same as the theoretical coordinate system of the ship body segmentation design model; outputting the required point cloud format, saving the complete ship body segmentation point cloud data, and realizing the ship body segmentation splicing. The precision and efficiency of the segmentation splicing directly determine the linear smoothness, structural strength, assembly stress, construction period and cost of the final ship body, and is one of the core key processes of ship construction.

The application discloses a kind of T profile intersection panel double-angle transition operation methods, and the method specifically includes the following steps: S1, web in T profile is respectively welded and fixed with bulkhead and outer plate, and the circular radius R of arc angle transition is measured;S2, the panel of T profile on bulkhead is installed on the web on bulkhead, and the panel width B is measured;S3, the panel of T profile on outer plate is installed on the web on outer plate, and the angle A of panel and web on outer plate is measured;S4, make dome long circle bottom platform device, according to the circular radius R cutting dome long circle bottom platform, obtain the transition panel of arc angle transition;S5, the transition panel is installed on the web;S6, the transition panel is cut to panel excess after installation is completed, and the installation of bulkhead and outer plate transition panel is completed.The application is cut by making dome long circle bottom platform to realize the manufacture and installation of transition panel, improve the installation efficiency of transition panel and construction quality.

This invention discloses a backdoor attack method, system, and medium based on multi-object semantic coexistence in text-to-image diffusion models. This method utilizes the perspective of multi-object semantic coexistence in text-to-image diffusion models to develop MOBA backdoor attack schemes. First, by constructing a trigger alignment dataset and optimizing backdoor implantation and semantic preservation in parallel, the attack success rate and visual concealment are effectively improved, reducing the impact of semantic corruption on attack effectiveness. Second, during model training, an attention-based decoupling backdoor enhancement mechanism is adopted. By decoupling the attention regions of different objects, the semantic integrity of the input prompt is maintained while the backdoor is activated and the attention distribution is reasonably adjusted to reduce the impact of generation bias on attack concealment. The method of this invention ensures both visual concealment and model performance under benign input while achieving efficient backdoor attacks.

The application discloses a kind of machine learning training topology generator and the method for generating topological discriminator model, comprising the following steps: with matrix pair form characterizing real topological data, and pre-processing;Topology generator is constructed, and topological pseudo-matrix pair data is generated;Topological discriminator is constructed to identify the source of topological data;The topological generator and the topological discriminator are based on the adversarial training of generative adversarial network;Low-voltage area topological matrix pair data is generated, and is converted into visual topological graph.The application realizes the coevolution of topological generator and topological discriminator by constructing topological generator and topological discriminator and through the adversarial training of both, and then enables topological generator to automatically generate a large number of topological data in accordance with real station area statistical distribution and engineering rationality from simple input, with the technical effects of high generation quality, high automation efficiency, guaranteeing physical rationality, solving data scarcity problem and strong flexibility and adaptability.

This invention relates to a dynamic retrieval enhancement generation method based on critical path entropy increase trends and document selection. Addressing the issues of retrieval response lag and false triggering caused by single-point threshold triggering in existing technologies, as well as noise interference and resource waste caused by direct splicing of Top-K retrieval documents, this invention creatively proposes a critical path entropy increase trend triggering mechanism. It optimizes single-point entropy monitoring of all tokens to critical path token entropy increase trend monitoring, achieving on-demand retrieval triggering by dynamically determining uncertainty increments, thereby suppressing false triggering caused by noise fluctuations in non-critical locations and alleviating retrieval response lag. Furthermore, this invention proposes a document selection model based on direct preference optimization of DPO, optimizing the coarse-grained injection of Top-K documents into a fine-grained filtering and injection mechanism. By filtering out irrelevant and redundant documents, it reduces the invalid occupation of the context window, thereby solving the problems of reduced generation quality and wasted computational resources caused by noise interference.

Embodiments of the present disclosure provide a text information generation method and device, electronic equipment and storage medium. The method comprises: obtaining a target video, and extracting a video feature of the target video, the video feature being used to represent content in at least one content dimension of the target video; generating a target prompt according to the video feature, the target prompt being used to represent a generation rule of text information for the target video; and generating the text information for the target video based on the target prompt. By extracting the video feature of the target video, generating the target prompt, and using the target prompt in combination with the capability of a language model to generate the text information for the target video, the method realizes fast and efficient generation of the text information, while ensuring that the content of the generated text information matches the video content of the target video, and improves the generation efficiency and content quality of the text information corresponding to the target video.

This application discloses a video generation method, apparatus, storage medium, electronic device, and product. The main implementation scheme is as follows: A video creation page is displayed, on which a first input control is shown; in response to an interaction event triggered by the first input control, video creation requirements are obtained; in response to receiving a video generation instruction, based on the video creation requirements, feature analysis is performed on the determined basic video elements to generate video generation elements matching the video creation requirements; the video generation elements are then merged to obtain at least one video content, which is then displayed on the video creation page. Based on this scheme, video generation efficiency and quality can be improved, thereby optimizing the user experience.

The application discloses a kind of based on machine learning's tunnel unmanned plane autonomous flight system, comprising: multi-sensor data acquisition and pre-processing module, for collecting multi-source data and generating pre-processing data set;Improved Hector SLAM positioning module, for executing the introduction information entropy weight and the scanning matching of multi-source constraint, output two-dimensional map and horizontal pose estimation;Weak light vision incremental estimation module, for output light flow estimation visual incremental displacement;Unscented Kalman filter fusion module, for fusing multi-source information, output three-dimensional pose fusion result;Risk tensor construction module, for updating map and generating three-channel risk tensor;Improved A3C strategy control module, for fusing strategy output flight control instruction;Flight control and data acquisition module, for driving flight and generating inspection report.The application realizes the data fusion control effect of unmanned plane autonomous flight accurate navigation and efficient inspection in tunnel scene.

This application discloses a method and apparatus for searching and comparing technical solutions. The method includes: acquiring a technical solution input by a user and extracting a set of keywords corresponding to the technical solution; constructing a search strategy based on the keyword set, selecting multiple relevant documents from a preset literature database according to the search strategy, calculating a quality index corresponding to the search strategy based on the selected multiple relevant documents, updating the search strategy based on the quality index, until the quality index meets preset conditions; determining multiple target documents from multiple relevant documents based on the technical elements corresponding to the technical solution; generating search comparison information based on the multiple target documents, and displaying the search comparison information. Thus, this application can effectively solve the problems of low efficiency and high false negative rate of existing search solutions, realize the automation of the search comparison process, and improve the generation quality of search comparison information of technical solutions.

This invention relates to the field of video editing technology, and provides a method, apparatus, device, and medium for optimizing categorized electronic photo albums of family images. It solves the problem that currently automatically generated electronic photo albums cannot meet the diverse needs of users, leading to low user engagement. The method includes: acquiring a target video and user operation information for each electronic photo album corresponding to various automatically generated image types; optimizing a preset image classification rule based on the operation information to obtain a target classification rule; classifying each frame of the target video according to the target classification rule to obtain target image groups corresponding to each image category; and generating target electronic photo albums corresponding to each target image group based on each target image group, each electronic photo album, and the images already existing in each electronic photo album. The electronic photo albums obtained through this method are more in line with user preferences and usage habits.

Embodiments of the present disclosure disclose a large language model-based electronic contract generation and sending method, device and equipment. A specific implementation of the method includes: performing semantic analysis on an electronic contract demand information set to obtain a contract demand semantic information set; performing business data mapping processing on the contract demand semantic information set to obtain a business database information set; performing format conversion on the contract demand semantic information set to obtain a contract generation execution instruction set; generating a demand semantic dependency graph set; generating a database call statement set; generating a target electronic contract; performing abnormality detection on each contract segment information included in the target electronic contract to obtain a segment abnormality identification information set; and after adaptive display of the target electronic contract, performing sensitive encryption sending to a corresponding terminal for electronic signature processing. The implementation can improve the efficiency and quality of electronic contract generation, improve the security and display quality of the electronic contract, and reduce the waste of video memory and transmission resources.

The application provides a large language model-based question and answer method, device, equipment and medium. The application introduces Chosen reinforcement constraints and Rejected weakening constraints to optimize the large language model, uses Chosen reinforcement constraint parameters to maintain the stable generation ability of the large language model for Chosen answers, effectively maintains or even improves the generation quality of the large language model. The use of Rejected weakening constraint parameters effectively suppresses the generation of incorrect or unreliable low-quality answers, avoiding the reduction of the generation quality of the large language model. The two work together to effectively improve the tendency of the large language model to generate high-quality answers and maintain the stability of the output answer quality, so that the large language model can stably output accurate and reliable answers in mathematical reasoning, code generation and other question and answer scenarios, effectively solving the problem of output quality fluctuation and insufficient accuracy in existing question and answer scenarios.

The utility model belongs to engineering construction equipment technical field especially relates to a kind of for anti-settling well lid reinforcing plate foundation formwork device, it includes outside formwork, inside formwork and horizontal inclined brace, outside formwork outer facade is provided with several vertical arrangement horizontal stiffener, outside formwork outer facade is provided with several transverse arrangement vertical stiffener, horizontal stiffener and vertical stiffener intersect, inside formwork is located inside outside formwork, inside formwork inner wall is provided with several vertical arrangement stiffener ring, horizontal inclined brace quantity is at least two, horizontal inclined brace is crossed and is arranged in outside formwork and inside formwork top end, outside formwork and inside formwork are all with horizontal inclined brace fixed connection, the utility model can be made in construction site at any time, the material used is conventional, and production mode is fast and convenient, and applicability is extensive, can improve engineering construction efficiency and quality, increase construction safety, shorten construction period, reduce investment cost, reduce the influence to city traffic and environment.

The application discloses a point supervision semantic segmentation model construction and segmentation method based on SAM and CNN feature fusion, comprising the following steps: step one, acquiring remote sensing image data as a training set, and performing point labeling on target ground objects in the remote sensing image data to obtain a point label set, wherein the point label set is the category and position of the target ground objects; a point supervision semantic segmentation model is constructed; object structure information generated by using a convolutional neural network branch and a SAM driven fusion branch is used to realize collaborative modeling of image semantic information and structure information in combination with a cross-level fusion module CFM; meanwhile, under the condition of point supervision, the model can enhance the integrity control of the target region, improve the boundary positioning accuracy, and reduce the category confusion phenomenon, thereby significantly improving the effect of remote sensing image semantic segmentation. The technical problem that the existing point supervision semantic segmentation method is difficult to fully utilize image structure information and semantic information under the condition of sparse supervision information is solved.

This invention discloses an intelligent content generation and precise distribution method based on AIGC, comprising the following steps: constructing user individual feature vectors; constructing user association graphs; performing community detection processing to obtain user groups and calculate group distribution parameters, and constructing corresponding group semantic resonance fields; calculating group condition vectors, obtaining generation prompt information, and constructing conditional input sequences; using the AIGC model to generate target content consistent with the semantic features of the corresponding user groups; performing semantic projection calculation to obtain content semantic vectors, and calculating group fit and individual matching degree; generating distribution results for different user individuals; collecting user feedback data on the distribution results, and updating user individual feature vectors, user association graphs, and group semantic resonance fields, thereby realizing integrated collaborative processing of content generation and content distribution.

This invention relates to a large language model (RAG) method and system based on fine-grained retrieval text optimization. The method includes obtaining original question-answer pairs, retrieving several relevant texts using a semantic vector-based retrieval system, constructing a fine-grained retrieval text optimization model, which includes a text quality assessment module, a text optimization module, and a reinforcement learning agent. The text quality assessment module models the information content and distribution structure of the retrieved text, while the text optimization module selects corresponding optimization strategy templates from four types of strategy action signals: competition, backoff, cooperation, and complementarity. The reinforcement learning agent streamlines the optimization process, dynamically selecting and executing the optimal optimization action from the text optimization module to train the fine-grained retrieval text optimization model. The trained fine-grained retrieval text optimization model is then used for retrieval enhancement generation tasks. This invention, through intelligent decision-making text optimization, facilitates the achievement of higher quality and more interpretable and robust retrieval enhancement generation.

This invention relates to a method for determining the design life of asphalt pavement for highways. Currently, there are only general principles for determining the design life of highways, lacking specific methods. This method initially estimates the design life, calculates the pavement structure thickness; calculates the social benefit evaluation value and environmental carrying capacity evaluation value; estimates the life-cycle cost; determines the technical benefits; establishes a standard decision matrix for the social benefit evaluation value, environmental carrying capacity evaluation value, life-cycle cost, and technical benefits for each design life, and ranks them to determine the optimal design life. This method is the first to propose using social benefits, environmental carrying capacity, life-cycle cost, and technical benefits as analytical indicators for the design life of asphalt pavement for highways. By establishing an indicator judgment matrix, a decision matrix, and a proximity ranking, the design life of asphalt pavement for highways is determined, filling a methodological gap in current standards and providing a methodological and technical approach for the design, construction, maintenance, and operation management of highways.

A garment color generation system and method for human body images belong to the field of garment auxiliary design. In view of the problem of garment automatic coloring, a human body image dataset is first constructed, including original human body images, local gray human body images and corresponding human body analysis images; then a locally controllable garment color generation system is designed, including a color feature extraction module and a color generation adversarial network. The color feature extraction module extracts the color matching scheme from the reference material and generates the color feature map, realizing the local controllability of color rendering. The color generation adversarial network includes a generator with a "double encoder-single decoder" structure and a discriminator, and the generator fuses human body features and color features to automatically color the garment area in the human body image. The color generation system adopts the generator and discriminator in an adversarial manner for training, and is matched with various auxiliary loss functions, realizing high-quality human body image generation.

This specification discloses a model training method and apparatus, and a data processing method and apparatus. The training method includes: training a configuration prediction model based on model feature vectors and device feature vectors, capable of outputting configuration numbers for each layer of the model. The processing method includes: using the configuration prediction model to predict the configuration numbers of the model to be loaded and run on the device, thereby predicting the configuration numbers for each layer of the complete model. The device performs loading inference based on the predicted configuration numbers for each layer of the model, and can generate target content based on user input information.

The application discloses a kind of experience intention anchor point guide multi-modal trajectory planning method, constructs online map topology generation module based on multi-modal fusion perception, realize high-precision topology reconstruction under no map mode, through multi-modal fusion transformer real-time extraction environment characteristic and generate dynamic lane anchor point, improve the generalization ability and environmental adaptability of automatic driving system in unknown environment and unstructured road;The application introduces risk gradient guide mechanism in diffusion generation process, realizes the explicit safety correction of inference stage;Provide strong zero-sample safety guarantee at inference end, ensure that the generated trajectory can avoid in real time when facing sudden obstacles, improve the safety of systemThe application can effectively fuse perception topology and driving intention, in no high-precision map, dynamic game and long tail complex scene, significantly improve the geometric rationality of planning trajectory, intention consistency and safety, ensure the robust operation of automatic driving vehicle in variable environment.

This application provides a method, collaborative processing method, and device for allocating AI-generated content tasks in an industrial computing network, relating to the field of digital data processing. The allocation method includes: edge servers performing multiple rounds of federated learning; constructing a training consortium based on the comprehensive evaluation scores of terminal devices; collaboratively training and aggregating a hybrid expert model equipped with a resource-aware gating network; and simultaneously, cyclically executing resource competition. Based on terminal device requests, available resource capacity, and their resource contribution status, a competition mechanism is used to determine the target device and resource allocation scheme, and update the resource contribution status. The updated status serves as a feedback signal, dynamically influencing the comprehensive evaluation of subsequent federated learning, forming a closed-loop collaborative optimization of model training and resource allocation. This application can solve the problems of low performance of AI-generated content service models, poor resource allocation efficiency, and weak system sustainability under conditions of heterogeneous data, privacy sensitivity, and dynamically changing resources in the Industrial Internet of Things (IIoT).

The application provides a kind of image rain removal method and device, belong to image processing field, method includes: S1: each original rain image and each clear rain image is carried out data enhancement processing respectively, obtains rain day training image and rain day clear training image;S2: initial rain removal model is constructed;Wherein, initial rain removal model includes image processing module and image discrimination module;S3: based on image processing module, each rain day training image and rain day clear training image are carried out image processing respectively to obtain reconstructed rain day image, rain stripe removal image and reconstructed clear image.The application solves the problem that fog-related feature extraction is not accurate and image generation quality is low, obtains clear image with richer image content and higher generation quality, and improves the accuracy of clear image.

The application relates to the technical field of intelligent space layout generation, and discloses an indoor layout generation method and system based on double-track constraints, wherein the method comprises the following steps: acquiring indoor space structure data; constructing a current layout state and a candidate action set; scoring the candidate action based on a conditional scoring model; generating a score sequence after hard constraint modulation; calculating soft constraint energy; generating a candidate action probability distribution; and performing a step-by-step indoor layout generation task. Compared with the prior art which mainly relies on a single scoring model for layout generation, especially in a complex indoor scene with clear space non-placing areas and traffic constraints, it is difficult to guarantee layout legality and generation diversity at the same time. Due to the double-track constraint mechanism of hard constraint Boolean gate modulation and soft constraint energy hierarchical accumulation, the application realizes the probabilistic modeling of the prior legality screening and subsequent rationality differentiation of the candidate layout action, and improves the overall indoor layout generation quality.

The application relates to a door type scaffold with a vertical detection function and a vertical detection method, and belongs to the technical field of building construction. In order to overcome the fact that the verticality of the installation of steel pipes in a scaffold is difficult to be detected in real time in the prior art, and the construction efficiency and construction safety are affected, the door type scaffold with the vertical detection function comprises a base and a plurality of support pipes, the base and the support pipes both comprise a socket assembly, the socket assembly comprises a socket pipe, a gravity mechanism and a plurality of detection mechanisms, the gravity mechanism comprises a plumb bob, a connecting rod and a leveling disc, one end of the connecting rod is connected with the plumb bob, the other end of the connecting rod penetrates through the leveling disc and is connected with a fixed disc in a universal mode, and the detection mechanism comprises a detection rod and a marker rod, the lower end of the detection rod is abutted against the leveling disc, the upper end of the detection rod is connected with the marker rod, and the other end of the marker rod extends to an observation window. The application aims to realize real-time detection of the verticality of the connection between the steel pipe and the base, the steel pipe and the steel pipe, improve the construction convenience and the construction efficiency, and guarantee the construction safety and quality.