155results about How to "Realize intelligent analysis" patented technology

The invention discloses a non-scene law enforcement high-speed dynamic weighing detection system and method, and belongs to the field of traffic control. The system comprises a dynamic weighing system, a vehicle license plate recognition system, a variable message sign display system, an opposite-direction driving snapshot system and an upper computer software system. The dynamic weighing system, the vehicle license plate recognition system and the opposite-direction driving snapshot system are in signal connection with the upper computer software system; and the output end of the upper computer software system is connected with the input end of the variable message sign display system. Each vehicle is allowed to form a piece of complete vehicle pass information, comprising vehicle speed, axle load, number of axles, axle base, complete vehicle weight, license plate, speed, pass time and information about whether a vehicle overloads; if the vehicle overloads, a reminding message is sent to the variable message sign display system according to user setting to guide the vehicle to pull in to recheck; and meanwhile, the vehicle license plate information and weight information are displayed on a variable message sign, and are updated to a branch center database, thereby facilitating searching, sharing and analysis of vehicle pass records.

The invention discloses an IP camera in a video monitor system, a video monitor system, and a signal processing method of the IP camera. The IP camera can be used for detecting and tracking a moving object in a monitored scene represented by a digital video signal at the same time when acquiring the digital video signal, so as to determine whether to send an alarm or not according to the tracking data analysis, thereby achieving intelligent analysis while video monitoring. For the video monitor system and the IP camera, a moving object detection and tracking module and an intelligent video analysis module can be embedded in the hardware of the IP camera in a software manner to facilitate the function expansion and updating of the video monitor system.

The invention discloses a video monitoring intelligent early-warning system and a video monitoring intelligent early-warning method on the basis of target space relation constraint. The system comprises a video acquisition module, a target component intelligent analysis functional module, a video content visual analysis module, a general behavior rule sample database and an alarm module. On the basis of behavior comprehension on target components, aiming at spatial correlation between targets, the system carries out behavior characteristic analysis, evolution and classification, establishes a spatial correlation mode sample library, implements detection and identification on abnormal behaviors, can more effectively and accurately carry out identification and judgment on emergencies and can lock an abnormity trigger point to provide evidence for post-mordem forensics. The system introduces target space relation constraint, divides a video scene region to complete defining related concepts of the target components of a video, eliminates a certain fuzziness and faults in space semantic information, solves the problems of incompletion and inaccuracy in video content analysis and is more accurate and more efficient to detect abnormal events in a video monitoring range.

The invention discloses an industrial data intelligent analysis system, which belongs to the technical field of intelligent analysis and comprises an industrial data access DA, an interactive visual IV and an intelligent analysis AI. The method comprises the following steps: step 1, adopting a data access + big data underlying analysis platform + application layer Web visualization architecture mode by the whole industrial data intelligent analysis system; 2, adding a data source; 3, collecting data, and storing and integrating the data; 4, preprocessing the data; 5, realizing algorithm analysis and task development: modeling the data after data preprocessing according to a business scene, adopting intelligent analysis AI, providing analysis algorithm configuration of a bottom layer, customizing a task flow in a draggable manner, and carrying out periodic or immediate processing configuration on the task; and 6, carrying out operation management and display. The method has the beneficial effects that intelligent analysis of a large amount of industrial data is realized, and the potential value of the industrial data is mined; and abundant graphs explore a visual interface, and one-stop service is provided.

The invention discloses an anti-theft monitoring system for electric power facilities. The anti-theft monitoring system for the electric power facilities mainly comprises a outdoor monitoring host, a communication server, a data bank server, a WEB server, a multistage alarm receiving host, a user host and a cell phone terminal. A outdoor terminal is installed on a transformer pole and powered by a three-phase alternating current, and is connected with a various function detection module. The communication server, the data bank server and the WEB server are located at a monitoring center; the outdoor terminal monitors the environmental conditions and various operating parameters to guarantee the safety of the transformer, and transmits data with a communication server of the monitoring center through a GPRS (General Packet Radio Service) wireless network access way and keeps in touch with the communication server; the alarm receiving and management contact with a control center server through an internet access. The anti-theft monitoring system for electric power facilities can accurately analysis and judge the conditions of people approaching, facility vibration, facility slant and the current and voltage of the transformer, and is suitable for popularization and application.

The invention discloses a laser-electric arc composite welding real-time regulation and control system and a regulation and control method thereof. The regulation and control system comprises a laser-arc composite welding heat source system, a sensing system, a welding data collection system, a control unit, a control signal output system and a welding mechanical-movement control system. According to the laser-electric arc composite welding real-time regulation and control system and the regulation and control method thereof, information is collected through the sensing system, is input to the welding data collection system and is converted into respective digital signals, the digital signals are input into a data collection module of a control unit and are input into a human-computer interaction module through classification, the simple digital signals are converted into referable relative variable signals, then the relative variable signals are input into an intelligent module to be calculated, when the calculated result data deviates from a set range of the intelligent module, the electric arc current, the laser energy, the movement speed and the like are regulated and controlled in real time so that the welding comprehensive evaluation value is returned to the set range again, and therefore the purposes of stabilizing the welding process and improving the welding quality can be achieved.

The invention discloses a criminal case aid decision making method based on a fuzzy deep belief network. The method comprises steps of: establishing a vector model describing a case according to a legal document and a measurement of penalty method; establishing an appropriate deep neural network structure according to the complexity of a judged problem; optimizing a model parameter by using supervised and unsupervised learning methods in virtue of a known plus or minus judgment relation; and performing modeling by using the neural network and predicting the judgment result of a judge. The method can guarantee the objectivity and the fairness of case trial and improves trial efficiency.

The invention provides a log analysis method based on machine learning, and the method comprises the steps: obtaining a source log of a first time period from a target system according to a log analysis request when the log analysis request is received; Analyzing the source log through a preset random forest analysis model so as to judge whether the target system is abnormal or not according to ananalysis result; And if the target system is abnormal, determining an abnormal type according to the source log, and giving an alarm according to the abnormal type. The invention further provides a log analysis device and equipment and a computer readable storage medium. An analysis model is constructed in a machine learning mode, intelligent analysis of system logs is achieved, the workload andthe labor cost of people are effectively reduced, and the analysis efficiency and the operation and maintenance efficiency are improved; And the analysis can be automatically carried out at any time,so that the abnormal condition of the target system can be timely discovered to give an alarm, and the system abnormity can be timely discovered.

The invention discloses a large-volume concrete multi-field real-time online collaborative intelligent simulation method and a large-volume concrete multi-field real-time online collaborative intelligent simulation system. The method comprises the following steps: step S1, a sensing step of model establishment and real sensing input; S2, an analysis step of initializing parameters to be determinedand performing simulation calculation analysis; S3, an evaluation step of evaluating the truth degree of the simulation result; S4, a feedback step of judging simulation results and making process optimization adjustmentst; S5, a control step of performing closed-loop intelligent control; and S6, a learning step of learning the above steps, and starting a new loop. Through dynamic circulation ofthe steps, the frequency and precision of traditional simulation analysis and the application efficiency of simulation results are improved. Real perception, intelligent analysis and dynamic control of simulation objects are achieved, and the method can be applied to intelligent construction, full life cycle simulation analysis and the like of large-scale civil engineering and water conservancy structure engineering.

The invention discloses an intelligent power distribution system based on a cloud-management-edge-end architecture. The intelligent power distribution system comprises a cloud layer Internet of Thingsplatform, a management layer remote communication network, an edge layer edge computer terminal, and end layer data acquisition equipment. The intelligent power distribution system depends on an edgecomputing terminal which is edge layer equipment, realizes unified access of end layer data based on a local communication network, accesses the data to an electric power internet-of-things platformin a unified manner based on a remote communication network, and realizes unified acquisition access and intelligent analysis of data such as electrical quantity, environment monitoring, equipment state and video monitoring in a power distribution room station in combination with business sections such as an intelligent monitoring system in the power distribution room station and transformer arealow-voltage intelligent equipment monitoring.

The invention discloses a method and a system for monitoring underground engineering portrait information and monitoring information safely. The system comprises a monitoring measurement information processing system, a portrait information processing system, a comprehensive safety monitoring analyzing module, an early-warning and warning output module and a next-stage geologic structure reporting module. The monitoring measurement information processing system and the portrait information processing system are respectively connected with the input end of the comprehensive safety monitoring analyzing module, and the output end of the comprehensive safety monitoring analyzing module is respectively connected with the next-stage geologic structure reporting module and the early-warning and the warning output module. The method and the system for monitoring the underground engineering portrait information and the monitoring information safely acquire monitoring measurement information and portrait information in real time, analyze safety level of acquired data, perform early warning and warning on dangerous situations and forecast geological situations at the next stage, and have the advantages of performing remote control, timely finding and recording abnormal data information and warning data information, being convenient for technical personnel to analyze, and the like.

The invention provides a test case generation method based on the data code separation technology. The test case generation method comprises the following steps: S1, pre-processing program source codes and generating a program control flow graph; S2, generating a path matrix A according to the program control flow graph and setting corresponding priority factors Beta to all predicates in all paths P; S3, reversely ordering the paths according to the priority factors Beta to obtain a matrix A'; S4, as for the matrix A', selecting a path Pi to calculate test data, and finishing path selection when the order of a matrix A'' constituted by the selected path Pi equals to the order of the path matrix A; S5, generating test data according to the types of predicates in the path Pi; and S6, generating a test case according to the test data and the path Pi. The test case generation method realizes intelligent code analysis, reduces working hours for manual design of both test cases and test data, shortens the test script writing time and improves the test efficiency.

The invention discloses a method for intelligently analyzing a fault of a traction converter. The method includes: 1) detecting corresponding various different fault information in advance when faults of different fault sources in the traction converter occur, forming a set of fault fusion representation information with various fault information detected corresponding to each fault source, and establishing a model of corresponding relations between the fault sources and the fault fusion representation information; and 2) monitoring a running state of the traction converter in real time, obtaining various real-time fault information to form fault fusion real-time representation information, and analyzing a fault source matched with the fault fusion real-time representation information according to the corresponding relation model. The method for intelligently analyzing the fault of the traction converter has the advantages of simple implementation method, capability of realizing intelligent analysis of the fault of the traction converter, and high fault analysis efficiency and precision.

An embodiment of the invention discloses an intelligent interrogation system based on XGBoost disease prediction and a method. The system comprises a client and a service end which are connected through a network. The client comprises an interrogation module and a display module. The service end comprises a data processing module, a data storage module and an XGBoost disease predicting module. Through inputting patient symptom information which is acquired by a client into a trained XGBoost multi-class model, a disease probability predicted value and predicted disease information are output. Furthermore the patient can directly check the related information on the client. Based on machine learning and big data technology, a self-service diagnosis service of the patient can be realized. Thesystem and the method further have advantages of supporting intelligent analysis and prediction on mass patient data, realizing high accuracy and high timeliness in prediction, and realizing simple operation and low cost.

The invention discloses a network event processing method and device. The method comprises the steps that log information of network equipment is collected; events in the log information are analyzed, the events are stored in an event data base; correlating analysis is carried out on the events in the event data base through a correlating analysis algorithm, connection between the events is found; classifying is carried out on the connected events through a classifying algorithm, categories of the events are obtained; processing modes corresponding to the events after classifying are determined. The problem caused by the fact that network behavior is analyzed through a rule model aiming at correlated technologies is solved, and intelligent analysis on the network events is achieved.

The present invention relates to a mobile phone communication monitoring system device of a controllable cradle head. The device comprises a communication monitoring mobile phone device, a cloud server, an intelligent control mobile phone terminal and the controllable cradle head. The communication monitoring mobile phone device is installed on the controllable cradle head and is connected to the cloud server to carry out data transmission and data processing. The cloud server is connected to the intelligent control mobile phone terminal which is used for remotely controlling the controllable cradle head to realize cradle head rotation to capture a face image and searching best video communication and monitoring angles. A camera of the communication monitoring mobile phone device carries out automatic focusing, and the clarity of the face image is improved. The mobile phone communication monitoring system device is combined with the cloud platform, the storage capacity and the safety factor of the information are improved, and the information fusion and intelligent analysis of a communication video and a monitoring video image are realized.

The invention provides an intelligent medical detection method and an intelligent medical detection system based on an Internet cloud service technology. The method comprises the steps that 1, patient information is stored in a cloud server, and a detection device gives a detection result corresponding to the patient information; 2, a ZIGBEE transmission module transmits a ZIGBEE signal including the detection result; 3, a ZIGBEE receiving module monitors a received wireless signal for judging that whether the ZIGBEE signal having the detection result is included in the wireless signal; 4, if the ZIGBEE signal having the detection result is not included in the wireless signal, monitoring is continued, and if the ZIGBEE signal having the detection result is included in the wireless signal, the detection result of the patient is extracted after analyzing the ZIGBEE signal; 5, the detection result information is packaged into an Internet signal, then the Internet signal is encrypted and sent to the cloud server; and 6, the cloud server sends the detection result to a mobile terminal through the Internet. According to the method, the medical detection report is paperless, privacy of the personal information of the user is protected, the detection data can be monitored in real time and intelligently analyzed, remote medical guidance can be implemented, and statistics and analysis of big clinical medical data can be achieved.

The invention discloses a cluster scene intelligent monitoring method and system based on cluster trajectory classification. The method comprises the steps of first, adopting a key point tracking algorithm for obtaining a motion trajectory of a moving object in a monitoring video; then, adopting a spanning tree algorithm of a global uniform Markov random field model for connecting fragment trajectories with the time-space coherence to obtain a complete trajectory tree; then, adopting a confidence-based correlated topic model for learning middle layer features in the trajectory tree and redescribing the middle layer features; then, adopting a multi-class support vector machine classifier for classifying the trajectory tree on the basis of the trajectory tree described by the middle layer features; finally utilizing the confidence-based correlated topic model and a model of the multi-class support vector machine classifier and adopting a minimization information entropy algorithm for classifying the fragment trajectories. The complete cluster scene intelligent monitoring system is designed to replace manual analysis on cluster scene monitoring videos, manpower resources are saved, and the system robustness is good.

The invention discloses a smart grid big data-based electricity price execution checking method. The method comprises the following steps: 1) processing the power consumer power consumption data by utilizing a data pre-processing module; 2) constructing a power consumer typical power consumption track expert library by utilizing a clustering algorithm; and 3) realizing the electricity process execution checking of the power consumers by utilizing a distance discriminant analysis algorithm. According to the clustering algorithm, an initial clustering center can be determined through the density parameters of sample points; by utilizing a method of combining the cluster evaluation index and the score evaluation index, the optimum number of clusters can be determined, so that a power consumer typical power consumption track curve is formed; according to the distance discriminant analysis algorithm, electricity price execution checking discrimination is carried out on new power consumers, and through calculating the electricity price abnormal suspected coefficient, a list of the final the electricity price execution abnormal consumers is determined. According to the method, the intelligent analysis and identification of the power consumption behavior tracks can be realized, the remote online diagnosis of the consumer electricity price execution can be realized, and the pertinence, correctness and timeliness of the marketing checking can be improved.

The invention discloses a low-background spectrum unscrambling method based on a sodium iodide detector. The method comprises the following steps: (1) a gamma detector after energy calibration actually measures a radionuclide spectrum and a natural background spectrum, and noise reduction is carried out; (2) a deconvolution algorithm is adopted to process a gamma energy spectrum after noise reduction; (3) a peak searching algorithm carries out searching on all potential peak positions of the gamma energy spectrum after noise reduction, and a peak position list is acquired; and (4) the peak position list information and a nuclide library are combined to be together inputted to a fuzzy logic system, the system gives the peak position recognition confidence for any nuclide characteristic peakin the nuclide library, constraints are set to further adjust the peak position recognition confidence, and thus, a nuclide list and the nuclide recognition confidence are acquired. The low-background spectrum unscrambling method based on the sodium iodide detector has the advantages of accurate peak position reduction, remarkable improvement of energy resolution, decomposition of an overlapped peak, low error recognition rate, accurate background nuclide analysis and accurate low activity nuclide analysis, and can be used for nuclide recognition and activity calculation for radiation monitoring equipment such as a gated radiation monitor and a nuclide recognition instrument.

The invention discloses a road detection vehicle and a road detection method and device, and the method comprises the steps: obtaining a road image, performing disease target identification and extraction on the road image by adopting a deep learning model, and carrying out classification and parameter statistics on the extracted disease targets. According to the road detection method provided bythe embodiment of the invention, the damage information of the road surface can be automatically identified, classified and recorded through multi-lane parallel detection, so that automatic and intelligent analysis of road surface videos and image data acquired by a high-speed running road surface disease detection vehicle is realized.

The invention relates to an expressway traffic information display method, device and system based on an intelligent lamp post. The method comprises the steps: collecting vehicle information, determining a road section for displaying expressway traffic information, interactively transmitting expressway traffic information, and tracking images of a vehicle and an adjacent vehicle; analyzing the visual field of a driver of the vehicle according to the tracked images of the vehicle and the adjacent vehicle, and displaying highway traffic information according to the visual field of the driver; evaluating the road condition in front of the vehicle; and displaying the highway traffic information according to whether the front road is abnormal or not. According to the invention, real-time traffic information guidance is provided for drivers for different vehicles through the intelligent lamp post; the intelligent lamp post cooperates with the guardrail screen and the ground sign display screen in the expressway section covered by the intelligent lamp post, so that traffic monitoring blind spots are perfected, drivers can obtain traffic information conveniently, the traffic accident rateis effectively reduced, and the traffic efficiency of the expressway is improved.

The embodiment of the invention discloses a driver-vehicle association processing method and device. The method comprises the following steps that: the first positioning information of each mobile terminal on a road surface and the second positioning information of each intersection are obtained, inter-object distances between the mobile terminals are calculated; clustering analysis is carried outon the inter-object distances, so that a plurality of mobile terminal classes can be obtained, and driver terminals in the mobile terminal classes are determined; the driving tracks of vehicles are generated according to the second positioning information and the first positioning information of each driver terminal; and the license plate information of each vehicle shot at each intersection is acquired, and the vehicles with the license plate information are associated on the basis of the driving tracks, so that driver-vehicle association is completed. According to the method and device, evidence acquisition is facilitated through the first positioning information; the driver terminals are recognized through clustering analysis; the intelligent analysis of the behaviors of the vehicles is achieved; judgment on suspicious vehicles, black operation vehicles and the like is completed; and matching is performed on the license plate information of each vehicle shot at each intersection, the driving tracks of the vehicles are applied, and therefore, accurate matching can be achieved, and accurate driver-vehicle association is realized.

The invention relates to a metering device fault diagnosis method based on RF-GBDT. The specific method comprises the following steps: based on related data such as metering device attribute data, historical electric quantity data, meteorological data, metering abnormality record data and the like, extracting related indexes such as electric quantity characteristics, weather characteristics and the like from data 15 days before abnormality discovery time, constructing an RF-GBDT model, and identifying and diagnosing whether the metering device has a fault or not. The RF-GBDT model can be applied to fault diagnosis of the metering device. According to the method, the suspected fault list of the metering device is output, fault feature analysis of the metering device is carried out accordingto the suspected list, the feature rule of the suspected fault metering device is comprehensively understood, accurate first-aid repair service of the metering device is implemented in a targeted mode, the fault management level of the metering device is improved, and the user service satisfaction is improved.

The invention discloses front-end equipment in a video monitor system, a video monitor system, and a signal processing method in the front-end equipment. Moving objects in a monitored scene represented by digital video signals are detected and tracked at the same time of acquiring the digital video signals, and the tracking data is analyzed to obtain a result of whether or not to send an alarm, thereby achieving intelligent analysis while achieving video monitor. For the video monitor system and the front-end equipment, a moving object detection and tracking module and an intelligent video analysis module can be embedded in the hardware of the front-end equipment in the software manner so as to facilitate the function expansion and updating of the video monitor system..

The invention discloses a clinical image recording method. The method comprises the steps that when a photographing instruction is received, a disease template and an image template are obtained; image photographing is conducted on a current target based on the disease template and the image template to obtain disease feature images; labeling remark and saving are conducted on the disease featureimages. The invention further discloses a clinical image recording device and a computer readable storage medium. By means of the method and the device, accurate recording of clinical image information is achieved, and through labeling classification and related saving of other information, an existing data system of a hospital is combined to further achieve automated data management of clinical image information and intelligent analysis.

The invention relates to an intelligent navigation device based on cloud computation and used for a scenic area. The intelligent navigation device comprises an intelligent entrance ticket bracelet and is characterized in that the navigation device further comprises a scenic spot access control system which is connected to a cloud data center, the cloud data center is connected with a scenic spot program list timetable module, a scenic area route planning module, a tourist car positioning and scheduling module, a scenic spot carrying capacity module and an intelligent terminal module, and the intelligent terminal module is connected with an electronic display terminal and an intelligent client terminal. By means of the intelligent navigation device, scenic spots and routes in the scenic area can be intelligently analyzed, tourists can play according to the routes analyzed by the cloud data center, time is saved, and jamming is avoided.

A system used or realizing intelligent analysis of computer hang-up is prepared as utilizing monitor module to store hardware and driving program as well as application software information of computer into databank of computer, utilizing search module to search out corresponding information in databank of computer according to information key word obtained from journal file generated at time of computer hang-up then to provide relevant settlement plan.

The invention relates to a semi-supervised neighbor propagation learning and multi-visual dictionary model-based intelligent video analysis method. The method includes the following steps that: key video frames are extracted based on a sample-and-hold strategy; the key frames are converted into a grayscale image, and the grayscale image is averagely segmented into a plurality of image blocks, and OM feature vectors are generated based on the sequence of the average brightness of the image blocks; a video sample is divided based on AP clustering, so that mutually exclusive video sub clusters can be formed; the category labels corresponding to each sub cluster are determined according to the video sub clusters obtained through the division of the AP clustering, and a multi-visual dictionary is constructed; the category label of a video to be detected is judged online according to the minimum distance judgment criterion; and closed-loop feedback adaptive reconfiguration learning is adopted to reconstruct a multi-visual dictionary adaptive to a new environment, so that the category label of the video to be detected can be re-judged. With the method of the invention adopted, the problem that efficiency and precision in key frame extraction cannot be taken into account simultaneously can be solved, and enhanced video features have very high robustness for approximate global linear change; and accurate and intelligent detection adaptive to video mode change can be realized.