322 results about "Identification rate" patented technology

The invention relates to a ground penetrating radar parameter optimization and general survey method for highway subgrade condition detection. The method comprises four core modules: one is ground penetrating radar parameter optimization configuration, and automatic parameter correction during medium change is realized through static basic configuration and dynamic adaptive adjustment in combination with real-time sensing and PID (Proportion Integration Differentiation) control; the second method is rapid general survey execution, traffic flow, pavement vision and historical disease data are fused, a route is optimized through a Dijkstra algorithm, and abnormity is judged through multi-feature fusion; thirdly, imaging processing is optimized, multi-layer medium correction time delay imaging and complex disease classification imaging are provided, and the deep disease recognition rate is increased; and fourthly, maintenance decision linkage is realized, the disease level and priority are quantified, the maintenance scheme is automatically output, and the decision period is shortened. The method solves the problems of poor dynamic adaptation, difficulty in complex disease identification and disjunction in decision making in the prior art, has been verified in multiple sections, and is suitable for highway subgrade disease full-chain monitoring and maintenance.

The invention discloses a ground wire full life cycle management method, relates to the technical field of power equipment asset management, and aims to solve the problems that manual recording is easy to tamper, cloud diagnosis delay is high and an alarm mode is single and easy to neglect in traditional ground wire management. According to the method, operation fingerprints and spatio-temporal information during grounding wire hooking are recorded through the block chain technology, and it is ensured that data cannot be tampered; a built-in lightweight AI model of the edge computing unit is used for analyzing sensor data in real time to carry out localized fault diagnosis; and triggering a multi-mode sound-light alarm controlled by the PWM signal according to the diagnosis result grade. The system realizes full-life-cycle credible traceability, sub-second fault response and high-recognition-rate alarm of the state of the grounding wire, and is suitable for intelligent operation and maintenance of the grounding wire in the field of a transformer substation and the like.

The invention discloses an end-to-end radar working mode identification method based on hierarchical density clustering and an encoder, and the method comprises the steps: firstly obtaining the pulse repetition interval PRI, PW pulse width and carrier frequency RF data of a radar pulse, and constructing a radar pulse sequence; and meanwhile, recording an initial working mode label corresponding to each pulse. Secondly, inputting a radar pulse sequence for preprocessing to obtain a preprocessed radar pulse sequence; and finally, performing working mode identification on the preprocessed radar pulse sequence through an encoder structure neural network model, and outputting a working mode corresponding to the radar. According to the method, the input sequence quality is remarkably improved, data interference is reduced, and the recognition rate of a complex radar working mode is improved.

The embodiment of the invention provides an association identification method and device, equipment, a storage medium and a program product, and relates to the field of big data and financial science and technology. The method comprises the steps of obtaining business data of a target object; based on the business data, a semantic baseline of a preset element corresponding to the target object is constructed, and the semantic baseline comprises a relation type set, a role type set and a constraint rule set; generating a path template based on the semantic baseline; based on the path template, an initial path meeting the semantic baseline is obtained through mining in a preset knowledge graph, a target path of the target object is obtained through closed-loop consistency verification, and the target path is used for representing the incidence relation corresponding to the target object. According to the method, path search is prevented from being separated from compliance semantics, and the error recognition rate is reduced. A logic closed loop of path search and compliance verification is ensured, limitations on semantic expression, path search and evidence chain recheck are solved, and the accuracy and traceability of associated transaction identification are remarkably improved.

The invention relates to a paper archive digital processing method and device, equipment and a medium. The method comprises the following steps: extracting image quality characteristics of a paper archive image, identifying a flaw region based on the image quality characteristics, and classifying flaw types through a pre-trained neural network; according to a preset defect type-image restoration algorithm mapping rule, restoring the defect area by adopting a corresponding algorithm to obtain a finished image; extracting a character sequence from the trimmed image by using an optical character recognition algorithm, encoding the character sequence into a semantic vector, and determining a document theme label in a pre-established archive theme library through vector similarity matching; and a multi-dimensional index structure is generated by combining the original image, the trimmed image and the character sequence, so that efficient retrieval and management of archive contents are realized. According to the method, the accuracy of defect repair and the recognition rate of optical character recognition are improved, and the automation level of digital processing is enhanced.

The invention discloses a fishery management system and method based on lightweight anomaly detection and intelligent agent collaborative decision, and belongs to the technical field of fishery management, and the system comprises a visual information collection and processing module which is used for obtaining and processing the visual data of a breeding area; the lightweight anomaly detection module is used for periodically scanning the breeding area and identifying and classifying potential anomaly conditions in real time; the intelligent decision agent module is used for starting an inference engine when an abnormal condition is detected, and generating and executing a corresponding decision instruction; and the message notification Internet of Things communication module is used for realizing real-time transmission of information and equipment control. According to the invention, the problems of high energy consumption, low efficiency and difficulty in realizing intelligent monitoring in a resource-limited scene of existing fishery management are solved. According to the invention, by constructing a cascaded technical architecture of low-consumption inspection, accurate triggering and dynamic decision, efficient abnormal response can be realized in a resource sensitive scene, the computing power consumption is reduced, and the high-precision abnormal recognition rate is ensured.

The invention discloses an ore sorting system and method for real-time positioning and classification recognition based on deep learning, and relates to the technical field of AI vision, and the system comprises the following modules: an ore image multi-source fusion module which is used for collecting and preprocessing multi-modal ore images and fusing the multi-modal ore images to form a multi-modal image set; and the end-to-end deep learning reasoning module is used for combining an improved pyramid network structure and a local attention mechanism, extracting and fusing multi-scale feature information, performing reasoning calculation to output a recognition result, and obtaining a position coordinate, a category result and a judgment confidence coefficient of the ore. According to the method, through multi-modal image fusion and end-to-end deep learning reasoning, the position, category and contour information of the ore can be accurately recognized, the problem that the recognition rate of complex ore features, especially hard-to-process ores such as pock ores and linear ores, of a traditional method is low is effectively solved, the sorting precision is remarkably improved, and the sorting efficiency is improved. And mistaken discarding of concentrates and mistaken selection of tailings are reduced.

The invention belongs to the technical field of image processing and privacy protection, and particularly relates to a deep face anonymization standardization processing and evaluation system. The system comprises a data preprocessing module, a multi-technology integration anonymization module, a cross-scene test set module, a multi-dimensional evaluation module and a visual interaction module. The data preprocessing module realizes face data standardization processing by unifying image size, resolution and color depth standards and combining Dlib face detection and feature point positioning; the multi-technology integration anonymization module packages mainstream anonymization technologies such as FIT, G2Face, DeepPrivacy and the like, and provides an extensible interface; the cross-scene test set module covers face samples of different illumination, postures and ages; the multi-dimensional evaluation module constructs an evaluation system from privacy protection intensity (re-recognition rate, cosine similarity and the like), image quality (FID, SSIM and the like), attribute maintenance (difference measurement of expressions, hairstyles and the like) and processing efficiency (time / memory consumption); and the visual interaction module supports data uploading, technology selection, result viewing and automatic report generation. According to the method, the problems of no unified architecture and evaluation fragmentation in the prior art are solved, standardized processing and fair evaluation of a face anonymization technology are realized, and a privacy protection technology is promoted to be applied to the ground.

The invention discloses a building construction drawing key information intelligent extraction method and system. The system comprises a text recognition and error correction module, a CAD structure analysis module, a multi-modal information fusion and understanding module and a conflict resolution and structured output module. According to the method, text recognition robustness is improved by fusing an OCR technology and a large language model error correction mechanism, a spatial relation graph is constructed in combination with CAD vector analysis, building field knowledge is generated and injected through retrieval enhancement, and finally structured information is output through multi-modal collaboration and conflict arbitration. According to the method, the problems of high text misrecognition rate, incomplete component information extraction and weak semantic relationship reasoning in the prior art are solved, the accuracy and integrity of key information extraction are remarkably improved, and digital and efficient processing of architectural drawings is supported.

The application belongs to the field of power system fault diagnosis, and discloses a single-phase grounding fault determination method for a small current grounding system based on Bi-LSTM, which comprises the following steps: analyzing a model of a power system and actual dispatching data, building a computer numerical simulation model through the analyzed data, labeling fault data and non-fault data in the simulation data and the actual dispatching data, normalizing simulation and real-time dispatching characteristic data, making a training set and a test set of a Bi-LSTM model, and designing a fault line selection model based on Bi-LSTM and training and testing of the model. In the case of single-phase grounding, the application can quickly determine a fault line according to discrete dispatching data; compared with existing zero sequence current identification methods, the application does not require zero sequence current data, and has a higher identification rate in the case of neutral point grounding through an arc suppression coil or a fault point with high transition resistance, thereby realizing simple and efficient small current grounding fault line selection.

A multi-level privacy protection method based on bimodal PSNR quality control. A bimodal PSNR evaluation system is constructed to separate the calculation of Face-PSNR and Background-PSNR. An adaptive blur kernel is used for dynamic desensitization, and the kernel is calculated as max(w, h) / 6 based on the ROI size. A PSNR-driven iterative blur mechanism is introduced, and each round is checked until it meets the standard and is backed up with a mosaic. Differential privacy protection is applied to the log, and a heterogeneous detector is used to verify the desensitization effect independently. An accuracy attenuation check and compliance audit mechanism is established. This method achieves real-time processing at 17.8 FPS on Raspberry Pi 4B, with a desensitization incremental delay of 3.6 ms, ROIPSNR<22 dB and global PSNR>35 dB, accuracy attenuation≤3.41%, and privacy unidentifiability rate of 100%, meeting the GDPR / PIPL compliance requirements. It is suitable for privacy protection target detection in community security edge computing scenarios.

The present application belongs to the technical field of medical data recognition, and particularly relates to a depression recognition method and system based on differential evolution feature selection. Original electroencephalogram data is acquired, and the acquired original electroencephalogram data is preprocessed; based on the preprocessed electroencephalogram data, a differential evolution-based feature selection method is used to select an optimal feature subset; a self-adaptive hierarchical fusion network model based on Transformer is constructed, and training is performed based on the optimal feature subset to obtain a depression recognition reference model and a reference recognition result. The purpose of feature selection on electroencephalogram data is to remove redundant features and irrelevant features, thereby obtaining an optimal feature subset. The screened feature subset is put into training, which not only saves training time but also obtains a higher recognition rate than original data training.

An objective of the present disclosure is to obtain an object recognition rate via an operator of a monitoring center.　Therefore, the present disclosure relates to an object recognition rate calculation device that calculates a final object recognition rate on the basis of a first parameter relating to the video quality of a video acquired from an imaging device provided on a moving body and a second parameter relating to the transmission status of the video to an operator terminal, wherein the final object recognition rate indicates the ratio of people who can recognize an object within a time period in which it is possible to prevent the moving body from colliding with the object when the object appears at a target distance from the moving body when a plurality of people watch the video.

The application discloses an intelligent image recognition equipment testing tool and testing method, which comprises a message queue middleware and a testing result confirmation device. The message queue middleware comprises a module for simultaneously acquiring identification results of contraband identification of multiple intelligent image recognition equipment for the same X-ray machine. The identification results from different intelligent image recognition equipment are stored in different topics. The testing result confirmation device comprises a module for subscribing to the identification results in different topics, labeling the identification results with manufacturer labels of the intelligent image recognition equipment based on different topics, selecting corresponding audit labels from optional audit labels and configuring the audit labels to the identification results with the manufacturer labels, counting the number of the audit labels, and obtaining an identification rate according to the number. With the aid of storage technology and computer human-computer interaction technology or automatic comparison technology, the testing is realized on a complete online processing, so that the loss, discontinuity and inaccuracy of archives are avoided, the testing result is reliable, and the testing labor cost is reduced.

The application provides a calibration method and device of a vehicle-mounted surround view system, and relates to the technical field of intelligent vehicles. The method comprises the following steps: acquiring an image of a calibration object collected by a vehicle-mounted camera, the calibration object comprising multiple regions, at least one region comprising a pattern containing identification information, identifying the identification information of the pattern and the two-dimensional coordinates of the corner points of the pattern, and determining the calibration external parameters of the camera according to the identification information of the pattern and the two-dimensional coordinates of the corner points. Thus, the external parameter calibration of the camera is performed by identifying the corner points of the pattern of the calibration object, the distortion generated when the edge region of the image is identified is reduced, the requirement for ambient light and the arrangement difficulty of the calibration object are reduced, the identification rate of the corner points is improved, and the efficiency and accuracy of the calibration are improved.

The present disclosure provides a bearing fault online diagnosis method, device, storage medium and program product, the method comprising: based on the comprehensive importance weight corresponding to each sample data, sample management operation is carried out to obtain the latest sample data;In all the latest sample data, based on the preset sampling rule and the comprehensive importance weight corresponding to each latest sample data, sampling operation is carried out to obtain target sample data;Based on the target sample data, the first fault diagnosis model trained in advance is updated, so as to predict the current first level fault category diagnosis result corresponding to the bearing to be diagnosed. The present disclosure realizes intelligent screening and playback of sample data, provides high value training data support for online updating of the first fault diagnosis model, the design of comprehensive importance weight retains key samples, effectively solves the problem that online learning is prone to catastrophic forgetting and the recognition rate of old fault categories drops sharply, and at the same time ensures the diagnosis accuracy of the first fault diagnosis model.

This invention relates to the field of casting technology, and more particularly to an efficient casting process based on a high-strength cast pouring cup outer frame. The casting process includes acquiring process state parameters at detection points, obtaining outer frame thermal field parameters and interface pressure field parameters, fusing them to generate a comprehensive pouring feature vector, inputting it into a process state recognition model, determining the current process state category and outputting the category confidence score, thereby obtaining a process category selection set and an accurate recognition rate to determine whether the recognition result meets the standard; selecting the corresponding process state category from the selection set and determining the corresponding compensation degree, thereby performing compensation control operations; finally, determining the control efficiency index based on the re-acquired process state parameters after the compensation control operation, and determining whether to adjust the preset compensation degree until the control effect meets the standard. This invention solves the problem of isolated process parameters and inability to adaptively adjust in existing technologies, effectively improving the demolding success rate and reducing the casting defect rate.

The application discloses a kind of municipal drainage pipeline image recognition and analysis system, it is related to image recognition technical field, the system includes: data acquisition module gathers pipeline inner wall image, water quality, flow and equipment speed data;Scene variable coupling module is constructed model by dynamic weight calculation and multi-peak membership analysis, and outputs comprehensive scene identification;Image enhancement collaborative module is sequentially self-adapting enhanced image according to filtering, standardization, correction, sharpening;Defect identification analysis module extracts feature, classifies defect and generates structured report;Adaptive iteration module compares identification rate threshold, and optimizes and updates parameter library;The application constructs multivariate coupling model, combines sewage noise adaptive filter kernel formula and dynamic distortion correction strategy, solves the problem that traditional system scene is not adapted, image processing is not good, and identification is not accurate, adapts diversified pipeline detection demand.

This invention provides a real-time identity recognition method, device, and electronic device based on electromyography (EMG) signals. The real-time identity recognition method includes: acquiring a target EMG signal characterizing limb movement from collected raw EMG signals; determining the target time-frequency features of the target EMG signal; and performing recognition processing on the target time-frequency features based on a preset convolutional neural network to obtain target identity information. Using this method, a terminal device can receive raw EMG signals collected by an EMG wristband in real time and perform time-frequency feature extraction and identity recognition processing on the received raw EMG signals, thereby achieving the goal of real-time identity recognition based on EMG signals and improving the recognition rate of identity information while meeting real-time requirements.

The application discloses a kind of vehicle identification methods based on multi-sensor data fusion, comprising the following steps: S1. synchronously collecting the video information of vehicle and radar point cloud information;S2. extract vehicle image feature information in video information, and 3D detection is carried out to radar point cloud information, and vehicle prediction frame information is obtained;S3. vehicle image feature information and vehicle prediction frame information are fused, and vehicle fusion result is obtained;S4. whether the maximum outer rectangle overlap of vehicle image contour projection and vehicle prediction frame contour in vehicle fusion result is greater than set value, if yes, vehicle fusion result is effective, and the outline of vehicle is determined;If no, vehicle fusion result is invalid.The application solves the problem that the recognition rate of traditional video monitoring scheme is not high and the monitoring range is small in low visibility environment, and improves the robustness of vehicle identification.

The invention provides a contract term intelligent comparison and performance risk monitoring method and device and electronic equipment, and belongs to the technical field of software engineering.In the method, contract version comparison and difference analysis can be automatically completed, and the examination efficiency is improved; risk points in the contract terms can be identified through preset compliance rules, and the risk identification rate is high; the system can be connected with an external business system to realize performance risk monitoring and effectively prevent passive default; and data analysis can be carried out on the historical contract document, so that the contract template and / or the contract compliance rule are / is optimized, and effective utilization of data is realized.

The application discloses a kind of train vehicle fault image intelligent analysis methods based on lightweight deep learning technology, comprising: obtaining train image data, then the train fault component to be detected in image is labeled, then the overall data is divided into training set, test set two parts;Data enhancement operation is carried out, the hyperparameter required by algorithm is configured, and the training sample is input into the MobileDetectNet neural network model formed after improvement to carry out feature learning;The learned feature model is input to the new vehicle image to identify faults, and finally the area of the fault occurs is framed and the corresponding alarm information is output.The MobileDetectNet network intelligent identification model of the application has few parameters, occupies less memory and has high recognition rate, can improve the recognition rate of EMU fault detection under the original equipment CPU environment, reduce the work intensity of artificial, shorten the maintenance operation time, reduce the missed detection probability, so as to ensure the safe operation of EMU.

Embodiments of the present application relate to a fingerprint matching method and device, computer equipment and a storage medium, comprising: obtaining a first fingerprint image to be matched; expanding the first fingerprint image to generate an expanded second fingerprint image; performing fingerprint matching based on the first fingerprint image and the second fingerprint image to obtain a target matching fingerprint image corresponding to the first fingerprint image. Thus, by using the fingerprint latent distribution rule, the small-area fingerprint can be expanded, and the recognition rate of the small-area fingerprint can be improved.

The application discloses a multi-modal image data automatic recognition system and method based on machine learning, relates to the technical field of image data processing, and comprises the following steps: image data is called, pixel point feature information is extracted, and a pixel point feature sequence is constructed; feature matching degree analysis is performed on the pixel points, and hierarchical division is performed; spatial feature joint index analysis is performed on the pixel points in the same layer, and hierarchical pixel correlation index is obtained by integrating the analysis data; interlayer pixel classification analysis is performed on the pixel points between adjacent layers, pixel classification correlation analysis is performed on the adjacent layers, and pixel classification correlation index of the adjacent layers is determined; correlation evaluation of pixel layer correlation logic is performed on non-repeated image data, and the best image matching combination is output; the application realizes accurate recognition of multi-modal image data, reduces the misrecognition rate and interference rate caused by coarse recognition and coarse matching, and improves the accuracy of image recognition.

The invention belongs to the technical field of industrial automation control, and discloses a fluidized bed fan circuit fault diagnosis method and system based on IDBO-RBF. Comprising a sensor group, a signal conditioning circuit, a data acquisition card, a processor, a memory and a human-computer interaction interface, and the method comprises the following steps: S1, data acquisition and preprocessing; s2, constructing an IDBO-RBF fault diagnosis model; s3, model training; and S4, fault diagnosis. Through the improved IDBO-RBF neural network, accurate analysis of the fault signal of the fluidized bed fan circuit is realized, and the recognition rate of the circuit fault signal is greatly improved.

The invention provides a buried pipe network optical fiber sensing signal classification method and system by using an optimized and improved extreme learning machine. The method comprises the following steps: firstly, converting a vibration signal sample into a normalized one-dimensional matrix and generating a corresponding one-hot coding label; then network parameters are initialized, an optimization model with classification accuracy and recall ratio as objective functions is constructed, optimal parameters are searched through collaborative iteration of an exploration group, a mining group and an alternative group, the exploration group is responsible for expanding a search range around the whole space, the mining group is responsible for searching for a better solution between member ranges, and the alternative group is responsible for providing an optional solution; and finally, training a classifier by using the optimal parameter, and verifying the model performance on a test set. According to the method, the network weight is iteratively updated according to the rule, the purpose of quickly searching the optimal weight is achieved, and the recognition rate of the algorithm can be remarkably improved.

The invention provides a pulsed eddy current detection device and detection method based on roller pair adjustment, and relates to the technical field of nondestructive detection. The device comprises a supporting frame, a roller pair, a telescopic driving mechanism, a detection probe, a sliding adjusting mechanism, a distance measuring unit and a control unit. The control unit maintains the constant distance between the detection probe and the surface of the detected rod body based on a feedback signal of the distance measuring unit; based on the distance and the roller state signal, a telescopic driving mechanism is controlled to adjust the posture of the roller pair, so that the axis of the roller pair is aligned with the axis of the measured rod body; and when a preset spacing stability condition and an alignment condition are met, controlling the detection probe to transmit a pulsed eddy current excitation signal to the detected rod body, and receiving and outputting an eddy current response signal for defect identification. The double-shaft lift-off effect is actively inhibited through dynamic closed-loop control, the stability and accuracy of a detection signal are guaranteed, and then the recognition rate and detection reliability of the rod body surface and near-surface defects are remarkably improved.

The invention discloses a method and device for recognizing and reconstructing vector characters of a CAD drawing, and relates to the technical field of data processing. The method comprises the steps that a PDF file is analyzed, and vector elements and attribute information of the vector elements are extracted; character stroke candidate lines are screened out according to attribute grouping; carrying out connectivity analysis and density-based spatial clustering on the lines, and identifying a text region; rendering the vector lines of the text region into an image with high resolution, carrying out single character segmentation, and recognizing characters by combining drawing-level font library template matching with a single character recognition model; combining the single characters into word groups according to the positions and the context information, and performing word order correction by adopting a language model; the recognition result is written back into an editable CAD text object; according to the method, vector geometric information is directly utilized, distortion of bitmap sampling is avoided, the small character recognition rate and the anti-jamming capability are effectively improved, and end-to-end semantic recovery from vector lines to searchable and editable texts is achieved.

The application belongs to the field of electroencephalogram signal processing, and is a fatigue driving classification method and system based on electroencephalogram multi-scale fuzzy entropy features of CEEMDAN. The method comprises the following steps: collecting electroencephalogram data, and labeling part of the data; pre-processing the electroencephalogram signal to remove artifacts in the electroencephalogram signal; extracting features from the pre-processed electroencephalogram signal, constructing and training an SVM classifier, classifying unmarked data using the trained SVM classifier to obtain pseudo-labeled data; performing CEEMDAN processing and scale transformation processing on the pseudo-labeled data and the previously obtained labeled data to obtain electroencephalogram multi-scale fuzzy entropy features based on CEEMDAN, and establishing a fatigue state classification model to classify the electroencephalogram signal. The collection method is simple, is more suitable for application in an intelligent driving system, and has a high fatigue state recognition rate.