39 results about "Acquisition technique" patented technology

The invention relates to the technical field of industrial visual inspection, and particularly discloses a waste metal classification and recognition method and system based on image recognition, and the method comprises the steps: obtaining metal surface visual information through an image collection system, extracting multi-level depth features after preprocessing, and generating a preliminary classification result and confidence evaluation; when the confidence coefficient is insufficient, starting a multi-mode verification mechanism, acquiring element composition data by adopting a laser-induced breakdown spectroscopy technology, and acquiring surface topological characteristics by adopting a structured light three-dimensional scanning technology; matching the element data with a component database to generate a component verification result, and comparing the morphology features with a morphology database to generate a morphology verification result; and finally, three types of results are integrated based on a weighted fusion algorithm to generate a final classification decision, and a sorting mechanism is controlled to complete accurate sorting.

A power transmission tower grounding resistance intelligent measurement method based on ground potential field distributed sensing comprises the steps that a ground potential sensing network is arranged around a power transmission tower grounding device, and a ground potential field distributed measurement system with an iron tower as the center is established; a broadband characteristic coding current signal is injected into the grounding system through a weak current injection module; acquiring potential response data of each node of the ground potential sensing network excited by the feature coding current signal by adopting a distributed synchronous acquisition technology, and reconstructing complete ground potential field spatial distribution from sparse node observation data based on a compressed sensing theory through a signal reconstruction algorithm; and inputting the reconstructed complete ground potential field spatial distribution map into a pre-trained map neural network model, and outputting an intelligent identification result and an anomaly diagnosis conclusion of the ground resistance. According to the invention, non-contact, distributed and high-precision grounding resistance measurement under the condition that a grounding lead does not need to be disconnected is realized, and the measurement precision and reliability under a complex working condition are remarkably improved.

The invention discloses a psychological health assessment system and method based on big data, and relates to the technical field of physical and psychological health assessment, the system comprises a data acquisition module, a data processing module, a psychological mechanism analysis module, an ability quality modeling module, a multi-modal detection module and an assessment result generation module; the technical key points are as follows: in the implementation process of the technical scheme, a data acquisition module synchronously acquires static, dynamic, self-assessment and real-time emotion data of an assessed person in multiple scenes of campus, security and protection and the like; with a campus scene as an example, static information such as homework documents and social platform records of students is collected, dynamic behavior data such as classroom interaction and extracurricular activities are collected through an intelligent camera and voice equipment, and meanwhile, real-time emotion data are obtained by means of a face video stream collection technology to form a multi-dimensional data source; the heterogeneous data are processed into standardized coded data in a unified mode, and the limitation that traditional evaluation depends on a single questionnaire or simple observation is broken through.

The invention relates to a centrifugal compressor PIV speed measurement and analysis method, device and equipment and a medium. The method comprises the following steps: acquiring flow field optical measurement data of the centrifugal compressor through a particle image velocity measurement technology; inputting optical measurement data of the flow field into a deep learning model, and extracting dynamic characteristics of the flow field through a characteristic extraction layer of the deep learning model; reconstructing the dynamic characteristics of the flow field into a three-dimensional flow field velocity vector field by using a velocity vector reconstruction layer in the deep learning model; and generating a velocity measurement analysis result based on the three-dimensional flow field velocity vector field. By adopting the method, the multi-scale dynamic characteristics in the complex three-dimensional rotating flow field of the centrifugal compressor can be effectively captured, high-precision three-dimensional velocity vector reconstruction is realized, and the spatial resolution and robustness of flow field measurement are improved.

The invention relates to a high-precision historic building digital building block system and a modeling method thereof, and relates to the technical field of digital cultural heritage protection and education, and the system comprises a three-dimensional modeling module which is used for obtaining millimeter-level precision geometric data of historic building components through employing a laser point cloud and multi-view image fusion technology, the module acquires data through a laser scanner and a digital single lens reflex and realizes spatial registration by using a feature point matching algorithm. Through the laser point cloud and multi-view image fusion technology, millimeter-level precision geometric data acquisition of the ancient building component is realized, the accuracy of a digital model and the application of a parametric modeling algorithm are ensured, the building block units can be generated and scaled in proportion to adapt to different application scenes, and meanwhile, the consistency of geometric characteristics is kept; the design of a standardized plug interface module is based on topological optimization, connection requirements of different ancient building elements are automatically adapted, and the characteristics of a tenon-and-mortise structure are reserved.

The invention discloses a rock particle interface deformation identification method based on an image identification algorithm, and belongs to the field of rock particle interface deformation identification, and the method comprises the steps: obtaining a particle distribution range through a high-resolution image collection and multi-scale scanning technology, and when the scale span is detected to exceed a preset threshold value, determining the particle distribution range; and separating particle regions with different sizes by adopting a layered segmentation method. The texture mode of each layer is analyzed by using a convolutional neural network, a deformation feature vector is extracted, and algorithm granularity parameters are adjusted through iterative optimization so as to match the actual particle size. And when the matching degree is insufficient, a feedback loop mechanism is introduced to reprocess the image data and a noise filtering link is fused, so that the deformation identification precision is remarkably improved. Finally, an integrated calculation method is adopted to fuse multi-scale analysis results, accurate judgment of the deformation state of the whole particle interface is achieved, reliable comprehensive indexes are provided for rock stability evaluation, and the technical problem that a traditional method is insufficient in precision in cross-scale particle deformation recognition is solved.

Embodiments of the present application disclose a vector elbow centerline acquisition method and device, computer equipment and a storage medium. The method comprises: importing a three-dimensional model; using OpenCascade technology to acquire the outer surface of the pipe material of the three-dimensional model; using OpenCascade technology to make an equal parameter curve of the outer surface along the generatrix direction; using OpenCascade technology to create an intersection surface with the parameter curve inside the equal parameter curve as a boundary; and acquiring the intersection line of the intersection surface to obtain the elbow centerline. The method of the embodiments of the present application can improve the accuracy of the elbow centerline acquisition, and is suitable for various complex bending component models and has high efficiency.

The present invention discloses a method for constructing a large language model for predicting molecular properties based on a knowledge graph. The method includes multiple steps: first, high-throughput data acquisition technology is used to obtain pre-training corpus, including relevant data such as molecular properties and metabolic pathways. Then, based on these data, a knowledge graph about fourteen endpoints is constructed, and the knowledge graph represents the relationship between molecules, their properties and endpoints in the form of triples. Next, the large language model is pre-trained using LoRA technology, and the model is optimized using an incremental learning strategy to form an expert large model. Subsequently, RAG technology is used to retrieve triplets related to the target endpoint from the knowledge graph, and the retrieval results are sorted. Through prompt word design, the expert large model filters the retrieved triplets and extracts key information. Finally, the filtered triple list is converted into natural language input to the expert large model to generate prediction results of molecular properties and provide a basis for reasoning. This method significantly improves the accuracy and efficiency of predicting chemical reactions between molecules and endpoints.

The application discloses a kind of semi-supervised iterative training methods for complex scene remote sensing image interpretation, including the following steps: preliminary annotation is accumulated by semi-automatic acquisition or artificial labeling mode in early model training, training sample set is constructed to obtain annotation model by training, and interpretation model library is expanded;Cross pseudo-label constraint self-training is used to construct annotation model, and unlabeled data is gradually introduced to obtain more representative scene sample annotation, realize sample expansion and refinement.The application uses multi-modal network, based on the iterative training technology of semi-supervised learning such as self-training and cross pseudo-label, obtains the model set under large-area complex scene, based on scene rule set and model search, constructs model matching and integration technology, realizes large-area remote sensing image classification interpretation under complex scene;Through the connection of multi-modal data, sample library and model library, the continuous iteration of model and sample optimization is realized.

We provide the system. [Solution] A means of receiving inquiries from users in natural language, A means for analyzing the aforementioned query and extracting keywords, A means of obtaining related information using vector search technology with extracted keywords, A means of automatically generating a response in natural language based on the acquired information, A means of providing the generated answer to the user, A means for monitoring the operation of the aforementioned means and accumulating feedback for use in learning, An information processing system that includes this.

The application provides a seismic data acquisition and transmission method and device, computer equipment and a storage medium, the method comprising setting a time period for daily seismic data acquisition; acquiring seismic data within the time period; obtaining a broadcast message of a preset protocol based on narrowband Internet of Things technology; uploading state information of a seismic data acquisition system according to the broadcast message of the preset protocol; detecting whether the seismic data acquisition system is working normally by judging the uploaded state information based on an automatic return mechanism; when it is detected that the seismic data acquisition system is working normally, transmitting the acquired seismic data according to a preset mode and storing the seismic data according to a preset format based on wideband Internet of Things technology, which can reduce the acquisition of invalid data, improve data effectiveness, improve data transmission efficiency and reduce battery consumption of the seismic data acquisition system during the seismic data acquisition and transmission process.

The invention discloses a data processing method for automatic deployment of an enterprise-level PaaS platform. The method comprises four core steps of basic environment data acquisition and preprocessing, intelligent task allocation and resource scheduling, deployment process data interaction and collaborative management, and resource and data monitoring optimization after deployment. In the data acquisition stage, multi-dimensional basic environment data of an enterprise network, a server, storage and the like are acquired by using a multi-source data acquisition technology, and are subjected to cleaning, standardization and verification processing; during intelligent task allocation, precise resource allocation and dynamic capacity expansion and shrinkage are realized; in the deployment process, component collaboration and exception handling are guaranteed by means of a unique data interaction identifier, a message queue and a monitoring platform; and a big data analysis and evaluation model is applied to monitor resource use and data quality in real time and dynamically optimize platform configuration. According to the method, the automatic deployment efficiency and stability of the PaaS platform are remarkably improved, resource utilization is optimized, and the problems of data processing rigidness, unreasonable resource scheduling, insufficient management and the like in traditional deployment are solved.

This invention relates to the field of pose estimation technology, specifically to a multi-view, multi-person 3D pose estimation method, apparatus, and device. The method includes: acquiring image information and depth information of the target using multi-view acquisition technology; segmenting the person in a single view based on the image information and extracting 2D semantic features; reconstructing the point cloud based on the depth information to obtain a 3D point cloud in a preset world coordinate system; segmenting the point cloud for different human figures based on the 3D point cloud and calculating the point cloud boundaries as human body boundaries; mapping the 2D semantic features to the 3D point cloud to obtain 3D features; and classifying different human figures and constructing 3D poses based on the 3D features and human body boundaries. This technical solution can segment the point cloud for different human figures based on the 3D point cloud, accurately define the joint positions of different figures, and achieve differentiation and matching of different figures. Furthermore, based on discrete point cloud data, it has better flexibility and fault tolerance.

This invention relates to a computer-implemented method comprising: - acquiring (406) first training images (108) using a first image acquisition technique (104), each first training image depicting a plant-related motivation; - acquiring (402) second training images (106) using a second image acquisition technique (102), each second training image depicting a motivation depicted in a corresponding first training image; - automatically assigning (404) at least one label (150, 152, 154) to each acquired second training image; - describing the motivation... The first and second training images with the same motivation are spatially aligned (408) into aligned training image pairs; - a machine learning model (132) is trained (410) based on the aligned training image pairs and the labels, wherein, during training, the machine learning model (132) learns to automatically assign one or more labels (250, 252, 254) to any test image (205) acquired using the first image acquisition technique that depicts plant-related motivations; and - the trained machine learning model (132) is provided (412).

This invention relates to a method and system for intelligent sensing of drilling tool information based on RFID and visual recognition collaboration. It includes: a data acquisition terminal, used to acquire identification information and coded graphic information on drilling tool tags based on RFID and visual recognition collaboration technology during the drilling tool tripping process, and transmit this information to a data processing terminal; and a data processing terminal, used to perform image processing and damage assessment inversion on the acquired identification information and coded graphic information using image recognition and coded graphic damage inversion algorithms, obtaining quantitative damage indicators for the drilling tool tags, and recording them to form a dataset of operating conditions and tag status throughout the entire lifecycle of the drilling tool. This invention can be widely applied in the field of intelligent monitoring and digital management technology for downhole drilling tools in oil and gas exploration and development.

The present application relates to power distribution diagnosis technical field, specifically, it is a kind of intelligent electric appliance box power distribution fault self-diagnosis method, it includes the following steps: using sensor is deployed at the node and environmental point of intelligent electric appliance box, the multi-source data of voltage current, temperature vibration is obtained by synchronous acquisition technology, the voltage sag, harmonic distortion characteristics are extracted by power quality analysis, combined with temperature vibration and image information constructs multidimensional feature vector, the fault of power distribution module is decomposed into bottom event, constructs fault tree, determines top event by bottom event state evaluation and logic reasoning, uploads fault diagnosis result and processing suggestion to cloud platform, obtains the diagnosis module of continuous optimization by dynamic adjustment model parameter and feature extraction strategy of reinforcement learning algorithm.The present application solves the problems that traditional intelligent electric appliance box power distribution fault diagnosis relies on artificial inspection, diagnosis result reliability is poor and diagnosis module lacks continuous optimization ability.

The application relates to the technical field of industrial visual detection, and particularly discloses a waste metal classification and recognition method and system based on image recognition, which obtains metal surface visual information through an image acquisition system, extracts multi-level depth features after preprocessing, and generates a preliminary classification result and a confidence evaluation; when the confidence is insufficient, a multi-modal verification mechanism is started, laser-induced breakdown spectroscopy technology is used to obtain element composition data, and a structured light three-dimensional scanning technology is used to obtain surface topological features; the element data is matched with a composition database to generate a composition verification result, and the topological features are compared with a topological feature library to generate a topological feature verification result; finally, a weighted fusion algorithm is used to integrate the three types of results to generate a final classification decision, and a sorting mechanism is controlled to complete accurate sorting.

The application discloses a kind of based on multi-axis state signal's numerical control processing technology identification method and system, belong to machining process type identification technical field.Method includes to workpiece is standardized clamping positioning;Make numerical control machine tool execute standardization preoperation program;Adopt non-synchronous acquisition mode based on working condition trigger, obtain multi-axis current data;Introduce timestamp alignment algorithm to realize the consistency of multi-axis current data on time sequence;For the multi-axis current data of timestamp alignment, combined with empirical mode decomposition and wavelet threshold denoising algorithm, construct adaptive selective denoising preprocessing framework, determine the current data of sample segmentation, and utilize multi-axis synchronous sliding window technology, obtain three-dimensional data set;Process feature set is constructed;Adopt principal component analysis to process feature set is nonlinear feature dimension reduction, and combined with grid optimization's radial basis function kernel support vector mechanism constructs multi-class process identification model.The application effectively realizes the automatic identification of multiple processing technology.

Techniques for correcting an overlay misalignment between an external camera image and a system camera image are disclosed. A first system camera image and a first external camera image are acquired. A first visual alignment is performed between those two images to produce an overlaid image. Some of the content in the overlaid image is surrounded by a bounding element. A position of the bounding element is modified based on movements of the system camera and / or the external camera. In response to performing a second visual alignment using new images, an update vector is computed. Relative movement between the two cameras is determined. Based on the movement and based on the update vector, the bounding element is progressively transitioned to a corrected position in the overlaid image. A speed by which the bounding element is progressively transitioned is proportional to the amount of movement.

Improved magnetic resonance imaging systems, methods and software are described including a low field strength main magnet, a gradient coil assembly, an RF coil system, and a control system configured for the acquisition and processing of magnetic resonance imaging data from a patient while utilizing a sparse sampling imaging technique.

The invention discloses a comprehensive product type selection evaluation method based on autonomy of an information system, and relates to the technical field of autonomy of the information system, and the method comprises the steps: carrying out the all-directional and multi-angle evaluation of candidate products through setting a plurality of dimensions, including performance, safety and technical suitability; specific characteristic parameters are subdivided under each dimension, in the performance aspect, such as processing speed, response time and resource occupancy rate, in the security aspect, such as encryption, authentication, authorization, auditing and vulnerability management, in the technical suitability aspect, such as compatibility with the system and integration difficulty. In addition, big data and Internet acquisition technologies are adopted to obtain feedback records of the user on the information system product, and bid evaluation of the characteristic parameters is determined according to the feedback records; meanwhile, by calculating the distance from the characteristic parameter corresponding to each quality attribute of each candidate product to bid evaluation and dividing a characteristic parameter evaluation interval, powerful support is provided for product model selection.

This technology enhances user satisfaction with the results of assessments regarding the possibility of acquiring intellectual property rights. [Solution] The information processing device includes: a related technology acquisition unit that acquires related technical information related to the intellectual property to be analyzed based on information about the intellectual property to be analyzed described in natural language; a positive opinion generation unit that generates a positive opinion regarding the possibility of obtaining rights to the intellectual property to be analyzed using a first large-scale language model; a negative opinion generation unit that generates a negative opinion regarding the possibility of obtaining rights to the intellectual property to be analyzed using a second large-scale language model; and a conclusion generation unit that generates a conclusion regarding the possibility of obtaining rights to the intellectual property to be analyzed using a third large-scale language model.

The application discloses a self-learning and detection method, in particular to a self-learning and detection method of product manual assembly process on a production line. Firstly, the working area is divided by the edge detection and rectangle detection technology of OpenCV, the spatial information of the working area is saved, and the learning and debugging stage is started; in this stage, multiple skilled workers simulate the assembly action on the work station, the hand positioning frame with direction is acquired by the palm recognition technology in MediaPipe to position the palm position, the region and trajectory information of the palm are determined by comparing the region information, and the trajectory data is aligned to ensure the consistency in time; next, the genetic algorithm is used to find the optimal trajectory; when the workers actually assemble, MediaPipe tracks the palm position and judges the region information, and then compares it with the standard process. The application solves the technical problem of providing the self-learning and detection method of product manual assembly process on a production line, and has the advantages of faster speed, lower cost, easy popularization and the like.

The present disclosure relates to deriving one or more of a precision, a recall, or an F-score metric. In accordance with the disclosed techniques a set of data is acquired or accessed. The data is continuously distributed. Based on an expected signal and an observed signal, one or more of a precision, a recall, or an F-score are determined for one or more automated routines configured to operate on the set of data. One or more of the precision, the recall, or the F-score provide a comparative measure of the accuracy of the one or more automated routines.

The invention relates to the technical field of logistics, and discloses a logistics abnormal cost node detection method, device and equipment and a storage medium, and the method comprises the steps: obtaining logistics node data based on a multi-channel asynchronous collection technology, fusing the logistics node data based on a graph neural network feature fusion method, so as to obtain a comprehensive cost feature vector, training a conditional quantile regression model based on historical events, inputting the comprehensive cost feature vector data into the conditional quantile regression model, screening out suspected abnormal nodes, obtaining a preset threshold value of node abnormal scores, calculating the node abnormal scores of the suspected abnormal nodes based on an improved isolation forest algorithm, and when the node abnormal scores exceed the preset threshold value of the node abnormal scores, determining that the abnormal nodes are abnormal. According to the scheme, the method and the device have the advantages that the defects of false report and missing report of the traditional scheme are overcome, the scene adaptability is high, and the accuracy of abnormal cost node detection is ensured.

Techniques are described for complex preprocessing steps to ensure consistency of sorted sets of reference measurement data, which have conventionally been required when sorting reference measurement data sets for DPG algorithms captured using an EPI technique, to be omitted because items of reference measurement data captured via the described techniques are already consistent in themselves. Therefore, for each polarity of the read-out gradients, a set of fully sampled reference measurement data is available, which is already suitable for carrying out a dual-polarity (DP) algorithm without any further measures.

The invention discloses a personalized credit product intelligent recommendation system for small and micro enterprises. The system comprises a data acquisition and preprocessing module, a risk assessment and demand analysis module, a credit product knowledge base module, a personalized recommendation algorithm module, a user interaction and feedback module and a system management and monitoring module. The data acquisition module acquires enterprise full-dimension information through multiple interface technologies such as RESTFulAPI, JDBC / ODBC and HTTPS protocols, abnormal value detection is carried out by adopting an isolated forest algorithm, and data preprocessing is carried out by using methods such as a K-nearest neighbor interpolation method and one-hot coding. The risk assessment module adopts a feature screening method combining recursive feature elimination and random forest feature importance to construct a random forest and support vector machine fusion model, and realizes credit risk quantitative scoring through logarithmic function mapping. According to the invention, credit demands of small and micro enterprises can be accurately matched, recommendation accuracy and credit approval efficiency are significantly improved, and credit risks of financial institutions are effectively reduced.

The invention relates to a computer-implemented method comprising: - acquiring (406) first training images (108) using a first image acquisition technique (104), each first training image representing a plant-related motif; - acquiring (402) second training images (106) using a second image acquisition technique (102), each second training image representing the motif depicted in a respective first training image; - automatically assigning (404) at least one label (150, 152, 154) to each of the acquired second training images; - spatially aligning (408) the first and second training images representing the same motif into a pair of aligned training images;- training (410) a machine learning model (132) based on the aligned training image pairs and labels, wherein during training the machine learning model (132) learns to automatically assign one or more labels (250, 252, 254) to any test image (205) acquired with the first image acquisition technique depicting a plant-related motif; and - providing (412) the trained machine learning model (132).

An information processing apparatus includes a related art acquisition unit that acquires related art information indicating related art related to an analysis target intellectual property based on analysis target information, a positive opinion generation unit that generates a positive opinion regarding a possibility of acquiring a right to the analysis target intellectual property based on the analysis target information and the related art information by using a first model, a negative opinion generation unit that generates a negative opinion regarding the possibility based on the analysis target information and the related art information by using a second model, and a conclusion generation unit that generates a conclusion regarding the possibility based on the positive opinion and the negative opinion by using a third model. This information processing apparatus enhances patentability evaluation through AI-driven decision making support.