52 results about "Automated algorithm" patented technology

Disclosed in the present invention are an edge-deployed semi-supervised anomaly detection method and system for a railway track foreign object. The method comprises the following steps: an edge device encoding and decoding a video stream captured by a camera to obtain an image frame sequence, and performing frame extraction; and using a semantic segmentation model to perform image segmentation on a certain image frame obtained by means of frame extraction, to obtain a railway track region segmentation image. The use of a single image as input may generate an expert model result having a high weight value; however, the determination based on a single image is not stable, multiple consecutive images of the task scene need to be inputted, the frequency of each expert model obtaining the highest weight is computed, and the expert model corresponding to the highest frequency is the final solution. The present invention supports scene-adaptive foreign object detection algorithm automatic selection, and a user can perform selection on the basis of prior knowledge, or selection may be performed by a scene-adaptive automatic algorithm selection method; the user only needs to provide a batch of image data of the current scene, and the optimal algorithm selection can be evaluated.

The invention relates to the field of automatic algorithm design, in particular to an automatic algorithm design method for solving a vehicle path planning problem by utilizing a large language model, which comprises the following steps of: S1, constructing an AutoDH framework, and defining a decision form of an intelligent agent; s2, defining heuristics in an LLM pool, an improvement pool and a disturbance pool; s3, acquiring state information of the CVRPs, and intelligently selecting a heuristic mode through two-stage MDP; s4, designing a reward mechanism fusing solution quality improvement, heuristic time cost and LLMs API calling cost; and S5, training the intelligent agent to optimize the solution of the CVRPs. The method can intelligently select the most cost-effective heuristic algorithm according to the state information of the current CVRPs, and improves the flexibility and adaptability of algorithm design.

The invention provides a numerical simulation and stability analysis method and system based on monitoring data, and belongs to the technical field of monitoring and early warning. The problems that in the slope monitoring technology, data collecting and processing efficiency is low, calculation model precision and adaptability are poor, dynamic monitoring of slope stability lacks pertinence, the system integration degree is low, cost is high, and application is complex are solved. The method comprises the steps that slope data are collected in real time through a multifunctional sensor; transmitting the acquired data to a data acquisition system for real-time data monitoring and recording; a two-dimensional numerical model of the side slope is constructed through finite element analysis software, and the side slope is analyzed through the two-dimensional numerical model; and comparing and analyzing real-time monitoring data in the data acquisition system and a numerical simulation calculation result, analyzing the change trend of the side slope, identifying a potential instability area, updating the two-dimensional numerical model in real time by using an automatic algorithm, completing real-time dynamic evaluation of the side slope, and early warning potential side slope instability risks in advance.

The invention discloses a method and system for measuring and identifying a high crack of a house wall, and belongs to the technical field of buildings, and the method comprises the steps: collecting complete wall crack point cloud data, carrying out the geometric feature extraction of the wall crack point cloud data, and dynamically adjusting the deployment density of an environment and a strain sensor. And completing the prediction of the future width of the crack through the prediction model, completing the classification of the crack risk level based on the predicted value and the structure standard, and generating a final maintenance strategy according to the crack dynamic expansion condition, the risk level and the construction resource configuration rule. According to the scheme, by introducing dynamic data fusion and prediction analysis, through continuous data collection and environment change analysis, pre-judgment and intelligent decision making of the future development trend of the crack are achieved, the damage level of the crack is evaluated through an automatic algorithm, the system can generate personalized repair plans under different crack conditions, and the repair efficiency of the crack is improved. The repairing efficiency is improved.

The invention provides a warehouse receipt supervision method and platform based on the Internet of Things, and relates to the technical field of warehouse management, and the method comprises the following steps: deploying an Internet of Things device in a warehouse, collecting the state information of stored goods in real time, preprocessing the collected data, and generating a warehouse receipt; generating a unique initial hash value for each warehouse receipt, and combining the warehouse receipt with the current hash value when a new warehouse receipt is generated each time; the Internet of Things equipment is deployed in the warehouse, the state information of the stored goods is collected in real time, the warehouse receipt is generated, the warehouse receipt information is processed and verified in real time through cooperation of the dynamic hash chain algorithm and the block chain, and then the warehouse receipt information is stored in the block chain, so that the authenticity and integrity of data are ensured, and the storage efficiency is improved. Verification rules and business logic of warehouse receipt information can be defined through an automatic algorithm, data verification and updating operation can be automatically executed, follow-up query, monitoring and auditing are facilitated, the whole supervision process is matched with automatic operation of the algorithm, and a large amount of manpower, material resources and financial resources are saved.

The invention relates to the technical field of computer science and industrial engineering, in particular to a tool workshop scheduling method, which comprises the following steps of: acquiring production characteristics of a current tool workshop, generating an optimal batch scheduling strategy according to the production characteristics of the current tool workshop based on a preset evolutionary strategy, and automatically designing an algorithm by utilizing a preset large model, and calculating the minimum completion time and the maximum completion time of the optimal batch scheduling strategy, and completing tool workshop scheduling according to the minimum completion time and the maximum completion time based on the optimal batch scheduling strategy. Therefore, the problems of high workshop scheduling complexity and the like caused by strong coupling of batch division and working procedure sorting, strict constraint between working procedures and high resource networking degree in workshop scheduling are solved, an automatic algorithm design framework based on a large model is introduced, the expert algorithm design time is shortened, and the workshop scheduling efficiency is improved. Therefore, the efficiency and feasibility of aircraft tooling production scheduling are improved.

Methods for non-invasive or minimally invasive assessment of epithelial tissue structure are disclosed. Digital imaging and processing are used to identify cell locations. More specifically, an automated algorithm that may be used to identify epithelial tissue structure, and / or to specify the coordinates / locations of cells in the epithelial tissue structure, through non-invasive or minimally invasive imaging, and use of this information to extract values of epithelial structure related parameters are disclosed.

The invention relates to the technical field of tire design, in particular to an automatic conversion method and system for a tire type rubber outer contour based on coordinate conversion and a program product. According to the method, outer contour feature point coordinates of a conversion object and an overturning object are automatically read and processed through a program, the distance from feature points to an upper outer contour is calculated, and new conversion coordinates are generated. Thirdly, automatically drawing the upper part and the lower part of the new contour by a program, and connecting the upper part and the lower part through an automatic algorithm to form a complete conversion contour; according to the method, the design precision and the working efficiency are remarkably improved, manual operation and personal errors are reduced, and the problems of missing points and repeated points caused by manual measurement are avoided. The method has high adaptability, can process automatic conversion of irregular contours, and is suitable for different design software platforms. By reducing the workload of designers and improving the intelligent level, an efficient and accurate automatic solution is provided for tire rubber design, and the method has wide application prospects.

The invention discloses an automatic algorithm configuration method based on deep reinforcement learning in evolutionary computation. The method comprises the following steps: determining a problem set; obtaining initial data for constructing a Markov decision process; defining a state vector of each individual in a time step in an evolutionary computation algorithm by using an individual state vector according to the initial data; constructing feature vectors of population features, individual features and development features according to the initial data; defining a continuous action space according to the initial data, and jointly controlling the selection of all individuals on hyper-parameters; based on whether a better solution is found or not, a reward mechanism is set, and maximization of expected benefits is achieved; and the deep reinforcement learning agent uses an attention mechanism to control dynamic hyper-parameters of the algorithm, so that a Markov decision process is realized, and automatic algorithm configuration is completed. According to the method, direct mapping from high-dimensional sensing information to continuous action space output is realized, configuration of individual exploration and development tradeoff is dynamically adjusted, and a balance mechanism of exploration and development is optimized.

The application discloses an ancient landslide dam extraction method based on optical images, and specifically comprises the following steps: step one, image acquisition; step two, basic element indication; step three, ancient landslide interpretation; step four, river blocking analysis; step five, dam interpretation; step six, parameter calculation and verification; and step seven, database construction. The application relates to the technical fields of remote sensing interpretation and ancient disaster identification. The ancient landslide dam extraction method based on optical images combines the ancient landslide range outlined on the images with direct interpretation signs of ancient dams and indirect inference methods, can preferably extract the ancient dams, and is supplemented by field sampling verification. Compared with an automatic algorithm identification mode, the ancient landslide dam extraction correctness is greatly improved, and the time and labor of pure field investigation are effectively avoided, and the efficiency is greatly improved.

This invention belongs to the field of dental medical technology, specifically relating to an automatic positioning method, electronic device, and storage medium for dental template teeth. The method includes: acquiring a patient's dental arch model, the cervical margin of a restoration, and a template tooth model; determining the dimensional characteristics of the template tooth; performing preliminary overall matching of the template tooth; extracting the tooth portions from the dental arch and fitting a dental arch curve; calibrating the position and size of the template tooth based on the dental arch curve, including determining the target position and orientation of the restoration based on the cervical margin and the dental arch curve for restored teeth, and determining and locating the missing tooth based on the projection of adjacent restorations, and scaling it according to the morphology of the cervical margin or the arc length of the missing tooth interval. This invention achieves precise initial positioning and adaptation of the template tooth through an automated algorithm, solving the problems of cumbersome and inefficient manual positioning in existing technologies.

According to the system for generating the intraoperative neural state monitoring report, data are processed and analyzed in real time through an automatic algorithm, and the neural state monitoring report is generated based on electromyographic signals in an operation and before the operation and neural confirmation and contrastive analysis. The generation of the monitoring report does not need manual writing, and the generation efficiency of the monitoring report is greatly improved. Through the system provided by the invention, professional technicians do not need to manually monitor and intercept signals in real time and carry out subsequent mass data analysis, and the processing center automatically carries out real-time signal interpretation, so that the manual data processing pressure is relieved.

A computer-aided method for determining permissible topological states (Z) of a power grid (10) is given, wherein the power grid (10) comprises a plurality of network nodes (N1-N5) and a plurality of switches (S1-S5) arranged between the network nodes (N1-N5), wherein the method comprises the following steps: a) Providing a first subset (M1) of the switches (S1,S2) which must be closed during the regular operation of the power grid (10), b) Providing a second subset (M2) of switches (S3-S5) which may be open during the regular operation of the power grid (10), c) Generating a multitude of incidence matrices (B), each representing a topological state (Z) of the power grid (10), - where for the respective topological state (Z) the closed switches (S1-S4) of the first (M1) and second (M2) subset are represented by entries in the incidence matrix (B), - where the individual incidence matrices (B) differ with respect to the entries corresponding to the closed switches (S3,S4) from the second subset (M2), - wherein the generation of the incidence matrices (B) is computer-aided using an automated algorithm, d) Checking the multitude of generated incidence matrices (B) with regard to at least one predefined boundary condition (RB1) as well as e) Forming a first set (Z1) of admissible topological states (Z) from the tested incidence matrices (B) that satisfy the predefined boundary condition (RB1). Furthermore, a corresponding operating procedure for a power grid (1) and a corresponding computer program product are specified.

The invention relates to the technical field of intelligent maintenance and artificial intelligence crossing of civil aviation, and provides a civil aircraft health management method and device based on a large model agent. According to the method, through combination of natural language interaction and an automatic algorithm, the workload of glue code writing and parameter debugging is reduced, the PHM system development threshold is lowered, and non-algorithm professional engineers can quickly construct an analysis process; task planning and algorithm calling are autonomously completed, the process execution efficiency is optimized, and the task processing period is shortened.

The invention belongs to the technical field of underwater acoustic signal processing, and particularly relates to an active sonar man-machine cooperation multi-target tracking method. Comprising the following steps: carrying out balanced denoising processing on background noise, and automatically detecting a threshold value to obtain azimuth and distance information of a target; meanwhile, the detection threshold value is manually adjusted through manual intervention; establishing a track for the detection point through an automatic algorithm and distributing a track ID; meanwhile, through manual intervention, suspected target points are manually recorded; calculating the proximity of each detection point and each track through an automatic algorithm, and carrying out track association according to a proximity matrix; meanwhile, through manual intervention, track ID affiliation is manually corrected; according to the invention, a hybrid tracking mode of'automatic-based and manual-assisted 'is formed, a full-automatic tracking algorithm is used for processing a conventional process, an abnormal decision is manually intervened, a high-reliability man-machine cooperative active sonar target tracking system is established, and the tracking robustness in a complex environment is improved by dynamically switching a manual / automatic mode.

The application relates to the technical field of robots and discloses a method for remotely manually repositioning coordinates of an indoor robot, which comprises the following steps: step 1, identifying a positioning loss state, and the platform sends an action: when the robot cannot maintain reliable positioning, the platform timely sends an alarm, and after an operator receives and confirms the alarm, the repositioning process is started; step 2, constructing a new map: a new map construction mechanism is set; step 3, calculating a target pose: the operator manually aligns the new map and the old map to obtain the target pose; and step 4, repositioning: the robot realizes internal updating according to the target pose. The application combines intelligent guidance of an automatic algorithm and flexible decision-making of manual operation, effectively overcomes the defects that pure automatic repositioning is prone to failure in a large range loss or a complex environment and falls into a local optimum, and simultaneously avoids the problems that traditional manual intervention needs on-site operation and is low in efficiency.

This invention provides a device and method for acquiring a three-dimensional model of abdominal organs, comprising: a coil support assembly, an MRI scanner, and an image processing device; the coil support assembly is clamped and fixed to the side of the examination bed to support the abdominal coil and reduce pressure on the abdomen of the user; the MRI scanner is positioned above the abdomen of the user, with the axis of the examination positioning light aligned with the midpoint of the axis of the coil support assembly, and a magnetic resonance imaging (MRI) scan sequence is acquired by scanning with the MRI scanner; the image processing device is communicatively connected to the MRI scanner, acquires the MRI scan sequence, and obtains a three-dimensional model of the abdominal organs based on the MRI scan sequence. This invention achieves precise extraction of abdominal organs at all levels through multi-region seed point strategies and adaptive threshold segmentation, overcoming the inefficiency and subjective limitations of traditional manual segmentation, and ensuring the reliability of the results through automated algorithms, thus providing support for the image analysis of abdominal organs.

The invention is suitable for the technical field of forensic dentition science, and provides an automatic forensic dentition science individual identification method based on a 3D computer vision algorithm, and the method comprises the steps: carrying out the filtering, denoising and down-sampling processing of an initial 3D dentition model, so as to remove the noise and reduce the data size; segmenting a dental crown region as a region of interest through an automatic algorithm, and removing gingiva and palatal plica soft tissues; and performing rigid registration on the segmented ROI model by adopting an iterative nearest point algorithm, and calculating a root-mean-square error and a registration fitting degree after registration. According to the method, the steps of filtering, denoising and down-sampling preprocessing are added before model segmentation, the 3D dental model segmentation, registration and comparison method is designed based on an open source 3D vision algorithm, full-process automatic processing from model preprocessing, model segmentation, model registration to model comparison is achieved, the registration and comparison precision can be guaranteed, and meanwhile the registration and comparison efficiency is improved. Registration efficiency is greatly improved, computing resources are saved, and subjective errors caused by manual segmentation of a model are avoided.

The invention discloses an intelligent interactive remote sensing information extraction method combined with a visual language large model, which comprises the following steps of: 1, extracting a to-be-interpreted remote sensing scene slice which only supports an RGB (Red, Green, Blue) mode at present; step 2, extracting image visual features by using a visual coding module; thirdly, feature alignment from the visual features to the natural language text features is achieved through a mapping module; and 4, generating a description natural language text of the image scene by using a language reasoning module. According to the method, a multi-modal method fusing visual and text features is adopted, the character description capability of the remote sensing scene is improved, the scene content can be visually displayed, the remote sensing visual scene is visually described through an automatic algorithm, and manual understanding and use are facilitated. Meanwhile, visual features of the remote sensing image scene can be effectively extracted, accurate natural language description is generated, and therefore the automatic generation function of the remote sensing image scene description text is achieved.

The invention relates to the field of safety production, in particular to an intelligent safety production dynamic supervision and early warning system, and aims to solve the problems that traditional safety production supervision depends on a large number of human resources, and potential safety hazards and omission risks exist. The system comprises a behavior analysis and identification module which is configured to detect whether production behavior data in current production environment image data is safe or not, and if not, alarm information is issued to an automatic alarm module; the scene model construction module is configured to generate safety production supervision processes for different production places; the summarizing statistical analysis module is used for configuring and summarizing dynamic supervision data and sending the summarized dynamic supervision data to the safety production hidden danger assessment module; and the safety production hidden danger assessment module is configured to assess the safety production supervision technology and capability of the enterprise and the existing safety production hidden danger based on the safety production supervision data. According to the invention, human resources are saved through an automatic algorithm, the early warning capability is enhanced, and the risk of potential safety hazard omission is reduced.

This invention discloses a method for rapid and automatic identification of structural surfaces based on high-precision point cloud data. To address the problem of slow regional growth and significant influence of the threshold of the growth criterion when segmenting structural surfaces, the concept of fuzzy boundaries is introduced to reduce the impact of the threshold on regional growth. Simultaneously, a pre-allocated search space accelerates the segmentation of regional growth, resulting in a rapid and accurate method for segmenting structural surfaces. While ensuring accurate identification of structural surfaces, this method rapidly extracts geological information about structural surfaces and improves the efficiency of obtaining this information through an automated algorithm.

This invention proposes a system for generating intraoperative neurological status monitoring reports. It utilizes automated algorithms to process and analyze data in real time, generating reports based on electromyographic signals during and before surgery, as well as nerve confirmation and comparative analysis. The system eliminates the need for manual report writing, significantly improving report generation efficiency. Furthermore, the system eliminates the need for specialized technicians to manually monitor and capture signals and perform extensive subsequent data analysis; the processing center automatically interprets the signals in real time, alleviating the burden of manual data processing.

Mechanisms are provided for automatically determining a speaker or writer's viewpoint, intensity, veracity and intended audience of natural language text. Language understanding functions including question answering, information search and retrieval, text generation, machine translation, transcription, summarization and decision-support are more accurate when these indirectly articulated elements of intent are understood. The functions use explicitly coded and curated commonsense knowledge propositions that form a knowledge graph data structure to interpret these elements of intent using contextual, subtextual and extratextual cues. These mechanisms operate on input text that has already undergone lexical disambiguation and provide a framework for identifying bias, deception and literary devices including simile, metaphor, analogy, allegory, personification, hyperbole, metonymy, sarcasm, diplomacy, intimidation and allusion.

A thermodynamic process and system for material modification including software procedures and methodologies comprising specialized thermodynamics processes. The system applies thermal bonding and division through thermal contrasting in nuclear technology, medicine, computing, and military operations. The process uses heat to bond fibers while employing contrasting temperatures to maintain different thermal zones in confined areas for specific results. The system emphasizes absolute containment to prevent contamination and organizes disordered cells, atoms, and mechanisms. Both automated algorithms and operator technicians provide oversight to verify data within the system database using current and real-time information.

The present invention discloses a glue coating detection method and device based on laser projection video recognition. The detection method includes the following steps: S1, laser projection line acquisition and recognition; S2, determination of the recognition area; S3, binarization of the recognition area; S4, detection of the glue state. Through video image processing and automatic algorithms, the present invention realizes automatic monitoring and analysis of the glue coating process, breaking away from the high cost and low efficiency of traditional manual methods and significantly improving work efficiency. Utilizing the principle of glue refraction, the continuity and shape of the glue can be accurately detected, avoiding errors that may arise from human judgment, achieving a high degree of automation and increasing reliability. Through real-time video monitoring and image processing, it can promptly detect whether the glue line is abnormally interrupted, and automatically identify the interruption location and distance, so that glue can be quickly replenished to ensure product quality.

This invention discloses a remote, non-contact, intelligent identification method for dangerous rock masses on steep slopes and rockfall risk assessment. This method uses an automated algorithm to intelligently identify dangerous rock masses by analyzing the kinematics of structural surfaces and combining the spatial position information of each structural surface in a point cloud. The goal is to identify potential dangerous rock masses in slopes and assess their potential for instability through non-contact measurement technology, reducing the risk of manual surveys. This intelligent method improves the efficiency, objectivity, accuracy, and comprehensiveness of dangerous rock mass identification, while also providing guidance for implementing appropriate preventive measures based on the level and scale of rockfall risk.

The invention discloses an automatic algorithm design method for solving a vehicle path planning problem by using a partial reconstruction heuristic operator, which integrates a partial reconstruction heuristic method, is used for solving a complex combinatorial optimization problem, constructs an AutoSAF framework and defines a heuristic pool, and the heuristic pool comprises an improvement pool, a reconstruction pool and a disturbance pool. In order to realize strong learning and reasoning capabilities, a selected sub-path is subjected to solution reconstruction by using an adversarial converter model and the solution is updated according to an acceptance standard of a mountain climbing strategy, the adversarial converter model is a heavy encoder-decoder architecture, a bidirectional GAN training strategy is adopted in the optimization process, the quality and the cost effectiveness of a solution generated by a generator are improved, and the method is suitable for large-scale popularization and application. When complex optimization problems such as vehicle path planning and the like are solved, the performance is remarkably improved, and the efficiency problem of a manual design algorithm is solved.

The invention provides a camera alignment method and a camera alignment and monitoring method. The method comprises the following steps: for a single camera, firstly, correcting a course angle and a pitch angle of the camera by acquiring a connecting line of an optical center and a vanishing point of the camera and combining the direction of a map lane line, and projecting a lane boundary into a three-dimensional space through lane boundary extraction and known height information in an image, so as to align the lane boundary with a map lane boundary; therefore, the horizontal position of the camera is corrected. For a multi-camera system, the poses of multiple cameras are optimized by calculating relative transformation information of a camera view overlapping area and combining lane line and stop line matching information of a single camera. Through an automatic algorithm, under the condition that the position error is 5m and the angle error is 10 degrees, automatic alignment of the camera pose can be rapidly completed, and the problem that a large amount of manpower is consumed in a scheme of carrying out camera pose alignment based on a position initial value and an angle initial value provided by a construction party at present is solved.

The application provides a warehouse receipt supervision method and platform based on Internet of Things, and relates to the technical field of warehouse management, which comprises the following steps: deploying Internet of Things equipment in the warehouse, collecting the state information of warehouse goods in real time, preprocessing the collected data, and generating a warehouse receipt; a unique initial hash value is generated for each warehouse receipt, and each time a new warehouse receipt is generated, the warehouse receipt is combined with the current hash value; the application deploys Internet of Things equipment in the warehouse, collects the state information of warehouse goods in real time, generates a warehouse receipt, and realizes real-time processing and verification of the warehouse receipt information through a dynamic hash chain algorithm combined with a block chain, and then stores the warehouse receipt information in the block chain, thereby ensuring the authenticity and integrity of the data; the verification rules and business logic of the warehouse receipt information can be defined through an automatic algorithm, and the data verification and updating operations are automatically executed, which facilitates subsequent query, monitoring and auditing; the overall supervision process is automatically operated in cooperation with the algorithm, thereby saving a large amount of manpower, material resources and financial resources.

The invention relates to the technical field of robots, and discloses an indoor robot remote manual coordinate repositioning method, which comprises the following steps that: step 1, a positioning loss state is identified, and a platform makes an action: when a robot cannot maintain reliable positioning, an alarm is sent to the platform in time, and an operator starts a repositioning process after receiving and confirming the alarm; 2, constructing a new map: setting a new map construction mechanism; step 3, calculating a target pose: manually aligning the new map and the old map by an operator to obtain the target pose; and 4, repositioning is carried out, wherein internal updating is achieved through the robot according to the target pose. According to the method, intelligent guidance of an automatic algorithm and flexible decision of manual operation are fused, the defects that pure automatic repositioning is prone to failure and falling into local optimum in a large-range loss or complex environment are effectively overcome, and meanwhile the problems that traditional manual intervention needs field operation and is low in efficiency are avoided.