Patents
Literature
Patsnap Copilot is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Patsnap Copilot

145 results about "Background analysis" patented technology

The background analysis is also referred to as a situation analysis, and is the nuts and bolts of the marketing plan.

Systems and methods for recognizing objects in radar imagery

ActiveUS20160019458A1Low in size and weight and power requirementImprove historical speed and accuracy performance limitationDigital computer detailsDigital dataPattern recognitionGraphics
The present invention is directed to systems and methods for detecting objects in a radar image stream. Embodiments of the invention can receive a data stream from radar sensors and use a deep neural network to convert the received data stream into a set of semantic labels, where each semantic label corresponds to an object in the radar data stream that the deep neural network has identified. Processing units running the deep neural network may be collocated onboard an airborne vehicle along with the radar sensor(s). The processing units can be configured with powerful, high-speed graphics processing units or field-programmable gate arrays that are low in size, weight, and power requirements. Embodiments of the invention are also directed to providing innovative advances to object recognition training systems that utilize a detector and an object recognition cascade to analyze radar image streams in real time. The object recognition cascade can comprise at least one recognizer that receives a non-background stream of image patches from a detector and automatically assigns one or more semantic labels to each non-background image patch. In some embodiments, a separate recognizer for the background analysis of patches may also be incorporated. There may be multiple detectors and multiple recognizers, depending on the design of the cascade. Embodiments of the invention also include novel methods to tailor deep neural network algorithms to successfully process radar imagery, utilizing techniques such as normalization, sampling, data augmentation, foveation, cascade architectures, and label harmonization.
Owner:GENERAL DYNAMICS MISSION SYST INC

Method and system for positioning WiFi (wireless fidelity) application focused areas

The invention relates to a method for positioning WiFi (wireless fidelity) application focused areas. The method comprises the following steps: a WiFi signal acquisition terminal arranged at a test site monitors and captures a part of or all WiFi wireless data messages from surrounding areas; the WiFi signal acquisition terminal analyzes the captured WiFi wireless data messages and screens out a detection request frame from the messages; the WiFi signal acquisition terminal acquires WiFi user information with the MAC(media access control) address and signal strength of a network card, and WiFi user data records are formed one by one in a form of user records and stored; the WiFi signal acquisition terminal uploads the stored WiFi user data records to a background analysis system; and the background analysis system summarizes and analyzes the received WiFi user data records so as to obtain the distribution situation of a part of or all WiFi application focused areas in a specific area. The invention relates to a WiFi signal acquisition terminal and a positioning system for WiFi application focused areas. By using the method and system disclosed by the invention, the WiFi user scale of each site can be analyzed, and WiFi application focused areas can be positioned precisely.
Owner:CHINA TELECOM CORP LTD

Configurable method for multi-vehicle model body information acquisition system

The invention relates to a configurable method for a multi-vehicle model body information acquisition system. The method includes the following steps that: acquisition configuration files and analysis configuration files which have corresponding relations are generated based on the different vehicle body buses of different vehicle models and the types of bus information and according to requirements of bus information acquisition; the acquisition configuration files and the analysis configuration files are introduced to a target vehicle-mounted bus information acquisition terminal and a server respectively; the vehicle-mounted bus information acquisition terminal acquires and generates standard encrypted compressed files according to the acquisition configuration files; the standard encrypted compressed files are uploaded to the server; and vehicle body bus information can be analyzed completely according to corresponding analysis configuration files. With the method adopted, acquisition and analysis of different vehicle body bus information can be realized fast through different configuration files for different vehicle models, and a situation in which one set of system is needed to be developed for each vehicle model can be avoided, and the standardization of the vehicle-mounted bus information acquisition terminal and a background analysis program can be realized, and the research and development efficiency of new products can be greatly improved.
Owner:WUHAN KOTEI TECH CORP

Intelligent detection method for electric vehicle charging control

ActiveCN104022552AAchieving Dynamic Modeling CapabilitiesBatteries circuit arrangementsElectrical testingData streamCharge control
The invention provides an intelligent detection method for electric vehicle charging control. The intelligent detection control method includes the following steps that A, an formation acquisition sub module acquires information from a front end charging terminal, then a data stream record is formed, and information is acquired every a period of time in the whole charging process; B, background modeling analysis is conducted, and then a background analysis model is formed; C, an access control sub module sends control instruction information to the front end charging terminal according to a preset charging mode value acquired by the background analysis model; D, an intelligent charging control detection module is based on the data stream record acquired in real time to construct a decision tree classification model through an increment classification decision tree method; E, a local analysis model and the background analysis model are compared; F, a comparison result acquired by the step E is transmitted to background analysis system software, a classification target value is prejudged with the reference to the local analysis model, and the background analysis model is optimized. According to the intelligent detection method, the local efficient detection function is achieved, and a background system can continuously adjust and optimize the background analysis model according to the received result.
Owner:ELECTRIC POWER RESEARCH INSTITUTE, CHINA SOUTHERN POWER GRID CO LTD +1

Grid equipment state inspection and evaluation training system

The invention discloses a grid equipment state inspection and evaluation training system. The grid equipment state inspection and evaluation training system comprises an equipment failure simulation module, a data information acquisition module, a state evaluation module and a risk evaluation module. According to the grid equipment state inspection and evaluation training system, grid data are generated through operation of the equipment failure simulation module and then transmitted to a background analysis module through a data acquisition system for failure diagnosis and risk elevation. According to the grid equipment state inspection and evaluation training system, the actual condition of a grid is simulated through equipment, that is to say, the big grid is moved into a laboratory; a novel and open advanced application comprehensive training system is established, an electric transmission and transformation equipment state information advanced application system and a dynamic simulation system are combined together organically, the correctness of an electric transmission and transformation equipment failure simulation system is verified through advanced application software, the correctness and reliability of the advanced application software also can be verified on the contrary, the combination realizes reuse of the grid equipment state inspection and evaluation training system, and therefore resources are saved greatly.
Owner:STATE GRID CORP OF CHINA +1

Systems and methods for recognizing objects in radar imagery

ActiveUS9978013B2Low in size and weight and power requirementImprove historical speed and accuracy performance limitationNeural architecturesRadio wave reradiation/reflectionGraphicsData stream
The present invention is directed to systems and methods for detecting objects in a radar image stream. Embodiments of the invention can receive a data stream from radar sensors and use a deep neural network to convert the received data stream into a set of semantic labels, where each semantic label corresponds to an object in the radar data stream that the deep neural network has identified. Processing units running the deep neural network may be collocated onboard an airborne vehicle along with the radar sensor(s). The processing units can be configured with powerful, high-speed graphics processing units or field-programmable gate arrays that are low in size, weight, and power requirements. Embodiments of the invention are also directed to providing innovative advances to object recognition training systems that utilize a detector and an object recognition cascade to analyze radar image streams in real time. The object recognition cascade can comprise at least one recognizer that receives a non-background stream of image patches from a detector and automatically assigns one or more semantic labels to each non-background image patch. In some embodiments, a separate recognizer for the background analysis of patches may also be incorporated. There may be multiple detectors and multiple recognizers, depending on the design of the cascade. Embodiments of the invention also include novel methods to tailor deep neural network algorithms to successfully process radar imagery, utilizing techniques such as normalization, sampling, data augmentation, foveation, cascade architectures, and label harmonization.
Owner:GENERAL DYNAMICS MISSION SYST INC

Abnormal power consumption monitoring and positioning system and method

InactiveCN106780115AQuick extractionSolve the difficult problem of finding abnormal power consumptionData processing applicationsCharacter and pattern recognitionFeature extractionReal time analysis
The invention relates to an abnormal power consumption monitoring and positioning system and method. Power consumption information of all users in a transformer area is input into an abnormal power consumption detection model, effective features are extracted and classified, and a suspected abnormal user list is output; in a range of the suspected abnormal user list, smart monitoring and data image collection devices are mounted for strongly suspected abnormal users of which abnormal power consumption evidences are judged insufficiently on site, and real-time monitoring and evidence obtaining are carried out on the power consumption condition of the strongly suspected abnormal users; and the power consumption data of the strongly suspected abnormal users collected by the smart monitoring and data image collection devices is received through a background analysis and push system for real-time analysis, the condition that the abnormal users are discovered is warned, and abnormal power consumption warning information is pushed to a plurality of power consumption real-time monitoring smart terminals contained in a power consumption real-time monitoring application platform. According to the system and the method, the problem that the abnormal power consumption is difficult to discover, obtain evidence and meter can be solved, and the abnormal power consumption monitoring is relatively smart and efficient.
Owner:SHANGHAI MUNICIPAL ELECTRIC POWER CO

Internet-of-things-based intelligent management system for kitchen waste vehicle

InactiveCN106892231AAvoid influenceAvoid impact, but also avoid clogging the sewerData processing applicationsRefuse receptaclesOperational costsData information
The invention relates to an intelligent management system for kitchen garbage trucks based on the Internet of Things, including garbage trucks for transporting garbage and an information control system for supervising the working conditions of the garbage trucks; the information control system includes a data acquisition device, a data acquisition Terminal, intelligent background analysis system and client equipment; the data collection device is used to collect information on the working conditions of garbage trucks and the garbage they carry; the data collection terminal is used to collect and summarize the data information obtained by the data collection device; the intelligent The background analysis system is used to analyze and process the data information sent by the data collection terminal. The intelligent management system of the kitchen garbage truck reasonably designs and plans the sanitation operation management mode, improves the sanitation operation quality, and reduces the sanitation operation cost. At the same time, the data collected by a large number of smart sensors is processed and analyzed, and then the potential commercial value of sanitation data is deeply excavated. The market prospect is promising.
Owner:CHONGQING SAGE RENEWABLE RESOURCES UTILIZATION CO LTD

Intelligent all-service fault diagnosing system

InactiveCN104320653ALow costBusiness application transparencyTelevision systemsData acquisitionCloud storage
The invention discloses an intelligent all-service fault diagnosing system. The intelligent all-service fault diagnosing system comprises a data gathering and testing terminal, a cloud storage platform and an analysis server, the data gathering and testing terminal receives data gathered by a plurality of test points, the data gathering and testing terminal is accessed into an IPTV optical network, and the data gathering and testing terminal, cloud storage platform and analysis server are in communication connection in sequence; the data gathering and testing terminal completely sends the gathered data to the cloud storage platform, and the analysis server compares the different data to confirm a problematic point location. A background analysis system of the intelligent all-service fault diagnosing system uses a mature free software to compile custom scripts to realize effective treatment, the system uses a lot of free open source platforms to perform secondary development, the comprehensive cost is greatly lowered, the field staff do not need to perform complex configuration and deployment and correct user terminals or network equipment, and the intelligent all-service fault diagnosing system is entirely transparent to the business application of the user and capable of diagnosing and analyzing faults under a good user sensing state.
Owner:中国电信集团工会上海市网络操作维护中心委员会 +1

System and method for farmland intelligent-control water distribution management

InactiveCN101836573AEfficient managementPromote voluntary water conservationWatering devicesCultivating equipmentsAridWater use
The invention discloses a system and a method for farmland intelligent-control water distribution management. The farmland intelligent-control water distribution management system comprises a terminal user card, an intelligent water supply controller and a water resource management system, wherein the terminal user card exchanges information with the intelligent water supply controller through a wireless radio frequency technique; the intelligent water supply controller is connected with the water resource management system; and background analysis software is loaded on a PC server of the water resource management system. The farmland intelligent-control water distribution management method comprises four steps, namely a step of analyzing data and formulating a scheme, a step of setting quota according to the scheme, a step of setting parameters and intelligently supplying water and a step of acquiring information, feeding back and adjusting. The invention can realize planned water use of agricultural irrigation, scientific water distribution, precise irrigation and intelligent control, and provides an effective water saving-distributing management scheme particularly for the problem of reasonably utilizing water resources in arid water-deficient areas.
Owner:青岛恒讯水利科技发展有限公司

Detection device and method for appearance and assembly of electric paper book based on machine vision

ActiveCN102353683AEnsure clarity and completenessGuaranteed infallibilityOptically investigating flaws/contaminationMachine visionEconomic benefits
The invention discloses a detection device for apparent defects and an assembly of an electric paper book. Three-section type online detection is adopted in the detection process of the device and back side detection of a product is carried out in a gap of the sections; and detection for the requirements on pollution, scratches, missing parts, assembly errors and the like is carried out by detection software based on a pattern recognition algorithm. According to the electric paper book detection device defined by the invention, the position of the electric paper book is transmitted to a camera through a sensor to trigger and collect images; and then background analysis is carried out according to the collected images. The design of the device guarantees an automatic flow of the detection process and the detection state can be monitored; and then products can be processed by a qualified/unqualified classification according to a detection result. By using the electric paper book detection device disclosed by the invention, objective and accurate detection can be obtained. In addition, the operation of the device is convenient and easy to understand and the defects of the electric paper book can be accurately detected without professional skills, so that careless mistakes caused by the judgment of artificial subjective factors can be effectively prevented and the detection deviceis good for improving the work efficiency of staff and the economic benefits of enterprises.
Owner:苏州优纳科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products