848 results about "Analytical algorithm" patented technology

This invention relates to intelligent video surveillance fuzzy logic neural networks, camera systems with local and network-shared communications for facial, physical condition and intoxication recognition. The device we reveal helps reduce underage drinking by detecting and refusing entrance or service to subjects under legal drinking age. The device we reveal can estimate attention of viewers of advertising, entertainment, displays and the like. The invention also relates to method, and Vision, Image and related-data, database-systems to reduce the volume of surveillance data through automatically recognizing and recording only occurrences of exceptions and elimination of non-events thereby achieving a reduction factor of up to 60,000. This invention permits members of the LastCall™ Network to share their databases of the facial recognition and identification of subjects recorded in the exception occurrences with participating members' databases: locally, citywide, nationally and internationally, depending upon level of sharing permission.

A system and methodology is provided for filtering temporal streams of information such as news stories by statistical measures of information novelty. Various techniques can be applied to custom tailor news feeds or other types of information based on information that a user has already reviewed. Methods for analyzing information novelty are provided along with a system that personalizes and filters information for users by identifying the novelty of stories in the context of stories they have already reviewed. The system employs novelty-analysis algorithms that represent articles as a bag of words and named entities. The algorithms analyze inter- and intra-document dynamics by considering how information evolves over time from article to article, as well as within individual articles.

Methods and apparatus for non-invasively assessing physiological hard of soft tissue of human and other species are described. In a preferred embodiment, tissue is vibrationally stimulated in vivo through a frequency spectrum. The tissue reacts against the stimulus and the reaction is preferably measured and recorded. Based on analytical algorithms or comparisons with previously taken measurements, changes within the tissue can be detected and used for diagnostic purposes. Further embodiments describe the usage of the device and methods for in vivo intra-operative and post-operative implant evaluations and as a therapeutic tool.

Disclosed herein are demographic based content delivery systems and associated methods that automatically and non-invasively detect information regarding one or more persons within a designated viewing area. In embodiments of the systems and methods, two different automatic viewer identification and demographic classification technologies may be utilized, either alone or on combination. The technologies include image recognition technologies that utilize computer driven image and pattern recognition analysis algorithms and radio frequency identification tagging and reading technologies. Data collected from either one of or both of the two technologies is used to feed a demographic classification algorithm which then delivers different multimedia content based upon the determination of the algorithm. The embodiments of the invention may therefore be advantageously utilized to market in real time to consumers at the point of sale.

A video coding method for surveillance videos allowing some regions of the scene to be encoded in an almost lossless manner. Such Regions of Interest (RoI) can be determined a priori or they can be automatically determined in real-time by an intelligent system. The user can set high priority in such regions a priori or the intelligent video analysis algorithm can automatically assign some windows a higher priority compared to the rest of the video. In a preferred embodiment, this can be achieved by canceling the motion estimation and compensation operations, and then decreasing the size of the quantization levels during the encoding process in the RoI. The present inventions can produce MPEG compatible bit-streams without sending any side information specifying the RoI.

An integration platform for managing long term processes. The platform includes a runtime engine that manages the process instances and communication with the processes to provide Web services. The runtime engine includes an execution engine module and a prediction module. The prediction module generates predictions for executing processes for inbound and outbound messages and eventual process outcomes. The predictions may be utilized when requisite data is unavailable or under defined circumstances. The prediction module utilizes process history, process description and process state data to generate relevant prediction data. The prediction module may include a learning module that applies analytical algorithms to the process history and descriptions to create models for use by the prediction module.

The invention provides an intelligent video analysis system and method based on PTZ video camera cruising. The system comprises a front-end PTZ video camera and a rear-end server. The rear-end server comprises a cruising configuration module, a PTZ video camera control module, a video analysis configuration module, a system control module, an intelligent video analysis module and an alarm management module. The cruising configuration module is used for setting cruising groups and cruising points of the system to generate a cruising list. The PTZ video camera control module is used for analyzing the cruising list and generating a cruising execution list automatically. The video analysis configuration module is used for configuring related intelligent video analytical algorithms for all the cruising points and configuring the cruising points into the cruising list. The system control module is used for calibrating video camera parameters of each cruising point, calling a video stitching module and automatically generating a panorama splicing map in the whole cruising period according to the execution sequence of the cruising points. The intelligent video analysis module is used for conducting target detection and event analysis according to related setting and giving a real-time alarm to a detected event. The alarm management module is used for conducting corresponding local management on the alarm.

The invention belongs to the field of communication technology and relates to a smart old-people caring system. The main structure of the system comprises an information collection subsystem, an information processing subsystem, a cloud platform and a command center, wherein the cloud platform is a data storage system and is used for storing information collected by the information collection subsystem and transmitting the information to the information processing subsystem for analytical processing; the information processing subsystem serves as an information processing pivot, is provided with an intelligent AI analytical algorithm module, performs intelligent analytical sort-out on collected big data and sends the processed information to the command center; and the command center can store all the information and provide information query and transmission services. Through the smart old-people caring system, 24-hour old people behavior analysis in videos is provided, a reasonable and effective judgment is made on suspicious abnormal behaviors of old people through bidirectional linkage in combination with sign data of old people, and abnormal behavior categories, confidence ofabnormal behaviors, risk levels, abnormal staff identities and first-aid projects are provided.

The invention provides a test paper detection card intelligent detection system and a test paper detection card intelligent analysis method. The test paper detection card intelligent detection system comprises a test card detection card and an intelligent terminal; the test paper detection card includes a test paper detection area; four corner positions of the test paper detection area are each provided with a positioning mark; a standard color area formed by a plurality of standard color blocks with different colors is arranged at one side of the test paper detection area; a reaction area formed by arraying a plurality of reaction blocks is arranged at the other side of the test paper detection area; the test paper detection area also is distributed with a validation area formed by a plurality of white shadow validation color blocks; a bar code area is arranged in the test paper detection area. The test paper detection card intelligent detection system and the test paper detection card intelligent analysis method have the advantages that environmental interference factors during image shooting are effectively eliminated, and the sensitivity and accuracy of detection results are improved; and in addition, a detection and analysis algorithm is simple, so the detection time can be shortened, and the system and the method can play an important role in urine routine, blood routine and other medical detection.

Information of failure events from various network devices and service objects is written into original event list. Correlation analysis is carried out for events read by analyzing and controlling engine from the original event list according to levels of original event and selection of classes. The dynamic analytical algorithm synthetically handles information in each area such as scene of historical fault analysis, dynamic performance parameters of network, dynamic topology information and time character of event etc. comparing with prior art, the invention overcomes disadvantages of overlooking information of dynamic network state, dependent on preset rule and lack of auto study ability etc. The invented method provides effective correlation analysis, fault reason analysis and location in real time when storm of network fault happens.

An air heavy pollution case judging method based on weather classification and meteorological element clustering comprises the following steps: classifying main weather systems and meteorological elements affecting PM2.5 concentration distribution from a meteorological point of view of factors influencing air pollution diffusion based on historical data of meteorological elements and monitored PM2.5 hourly data, and innovatively establishing a weather classification and pollutant source analysis algorithm based on a backward track pattern; and on the basis of the numerical WRF simulation result, looking for cases similar to typical weather situations of pollution weather in the same season in the history to serve for air quality forecast. A weather classification method and a numerical calculation simulation method are combined, which improves the accuracy of forecast results and reasonably avoids abnormality in numerical calculation results; an automatic classification similarity forecast method is built, which makes air quality forecast more business-oriented; and the cause and source of heavy pollution can be analyzed, and the applicability of the algorithm is improved greatly.

The invention discloses a big data collaborative analysis tool platform. The big data collaborative analysis tool platform comprises a multi-data source configuration module, a data retrieval module, a data processing module, a data analyzing module and a data visualizing module, wherein the multi-data source configuration module is used for importing and crawling configurations of any data; the data retrieving module is used for establishing a full text segmenting index for imported data; the data processing module is used for realizing a collaborated data indexing and collaborated analyzing functions of team collaboration; the data analyzing module is used for customizing an algorithm template, analyzing algorithm freedom selection and customizing data modeling and algorithm realization; and the data visualizing module is used for realizing visualized display and automatic form reporting of an analyzing result. The platform disclosed by the invention has the beneficial effects that the defects of specialty, difficult technology, high cost and low efficiency in industry information researching work are overcome; a function integrated information research big data operating platform is provided, and multi-dimensional big data services such as multi-data source configurable importing and crawling of big data, big data storage, a big data search engine, big data online collaborative analysis, big data online real-time statistics, analysis and excavation and visualization are provided.

The invention discloses a statistics static state sequential analysis method applied to a near/sub threshold digital circuit; the method comprises the following steps: reducing the standard cell library work voltage to near the threshold-voltage, and carrying out function simulation and characterization modeling for a near/sub threshold standard cell library; using a probability time-delay analysis algorithm to fast parse and rank path time-delays; using a Monte Carlo parse strategy and 3 [sigma] decision standard to concisely parse a suspected path, thus further improving sequential reliability. The provided accurate, reliable and fast statistics static state sequential analysis method can solve near/sub threshold digital circuit sequential analysis reliability problems, fully considers technology deviation influences on the path sequence, thus solving the near/sub threshold digital circuit sequential analysis reliability problems; compared with a conventional static state sequential analysis method and a Hspice-based sequential simulation method, the statistics static state sequential analysis method has obvious advantages on sequential analysis accuracy and efficiency.

The invention discloses a self-adaptive hyperspectral image unmixing method based on region segmentation. In consideration of coexistence of linear mixing and bilinear mixing, the method is implemented by adopting the following steps: inputting a hyperspectral image; estimating the number of end elements with a minimum error based hyperspectral signal recognition method; extracting end element matrixes with a vertex component analysis algorithm; clustering hyperspectral data with a K-means clustering method, and segmenting the image into a homogeneous region and a detail region; adopting a linear model for the homogeneous region and performing unmixing with a sparse-constrained non-negative matrix factorization method, and adopting a generalized bilinear model for the detail region and performing unmixing with a sparse-constrained semi-non-negative matrix factorization method. According to the method, characteristics of the hyperspectral data and abundance are combined, the hyperspectral image is represented more accurately, and the unmixing accuracy rate is increased. The sparse constraint condition is added to the abundance, the defect of high probability of local minimum limitation of the semi-non-negative matrix factorization method is overcome, more accurate abundance is obtained, and the method is applied to ground-object recognition for the hyperspectral image.

The invention discloses a user characteristic analysis-based multi-model load prediction method. The method comprises the steps of building a load prediction model based on different user characteristics by using a linear regression algorithm and a time sequence algorithm, building a data model through a specific data analysis algorithm to obtain multi-factor load prediction based on user characteristic analysis, and predicting line load values in the same period in the future through line load history data, micrometeorologic history data and regional GDP history data; and performing classification on line load data by utilizing a K-Means clustering algorithm, and classifying the line load data into a residential electricity consumption line, a commercial electricity consumption line and an industrial electricity consumption line according to electricity consumption types. According to the method, the difference among the lines different in electricity consumption type is fully considered, so that the model is more accurate; and the influence of multiple influence factors on loads is comprehensively considered, primary factors influencing the loads are found out by extracting primary components of the multiple influence factors, secondary factors are abandoned, and a primary influence factor-based prediction model is built by utilizing a data analysis algorithm.

The invention discloses an analysis method of remote sensing inversion of water color parameters of inland class II water. According to the method, inherent optical quantity of the class II water is inverted with an improved QAA (quasi-analytical algorithm), and concentration inversion of the water color parameters such as chlorophyll a and suspended matters is realized based on the inverted inherent optical quantity of the class II water. The method comprises specific steps as follows: (1), data of remote sensing reflectance above the surface of the water is input, parameter inversion of the inherent optical quantity of the water is realized based on the improved QAA, and absorption coefficients and scattering coefficients of the water and absorption coefficients of phytoplankton are acquired; (2), concentration data of the chlorophyll a and concentration data of the suspended matters of the water at a sampling point are input, a chlorophyll a concentration quantitative inversion model is established according to the concentration data of chlorophyll a and the absorption coefficients of the phytoplankton, and a suspended matter concentration quantitative inversion model is established according to the concentration data of the suspended matters and the absorption coefficients of the water after removal of pure water; (3), hyperspectral data finishing atmospheric correction are input, and concentration inversion of the water color parameters of the inland class II water in a monitoring area is realized according to the step (1) and the step (2).

A system for organizing data in multiple text data files and for automatically parsing data directly into named variables within a single, user-configurable graphical user interface (GUI) or workspace. Results organization, analysis algorithms, and file viewing all take place from within the workspace. No apriori knowledge of the dimensionality or length of data is required. Results may readily be placed into reports. Multiple data files may be processed using batch commands. Metafile capability is provided wherein data within a first level file may be the name of another data file allowing cascaded processing of data files. This unification of a user-configurable workspace, an integrated file viewing method and a parsing method minimizes distraction to a researcher allowing him or her to focus on analysis of experimental data rather than on the mechanics of data processing. The system of methods interacts with third party data analysis application software packages.

The invention relates to a beam bridge state evaluation method, which comprises the following steps of: 1) installing an FBG (fiber bragg grating) sensor on a beam bridge to be detected; 2) carrying out a quasi static load experiment on the beam bridge to be detected to obtain a control point deflection value; 3) comparing the control point deflection value obtained in the step 2 with a control point deflection value obtained by a finite element calculation model to obtain a beam bridge correction coefficient h; and 4) judging the working situation of the bridge structure according to the beam bridge correction coefficient h calculated in the step 3. The elevation method disclosed by the invention is convenient and practical; light load is used for replacing heavy load to lower the experiment expenditure; traffic interruption time is reduced; quasi static load or dynamic load is used for replacing static load; the analytical algorithm of the load experiment result can be directly obtained from a quasi static load experiment result so as to greatly lower the later-stage analysis difficulty of the quasi static load experiment data; and a quick load experiment method is used for replacing the traditional load experiment method.

The invention designs a novel parallelization scheme, particularly relates to a parallelizing method of association analytical algorithm in order to overcome the defect that a conventional association rule analysis algorithm Apriori cannot well adapt to parallelization. The parallelizing method includes blocking computation tasks via a master control node, allocating and distributing to various subsidiary computation nodes; parallelly computing via the various subsidiary computation nodes to screen frequent item sets, finally combining the nodes and returning results for statistics, and generating the frequent item sets; distributing the frequent item sets again and generating rules via various nodes. Since each computation node only processes a part of computation tasks, the problem that massive data cannot be processed by being read into an internal storage by one machine and processing speed is too slow is solved; the various nodes can be parallelly involved in processing, and processing efficiency is effectively improved; synchronous dependence, network communication overload, high frequency in I/O (input/output) operation among the nodes during computation are correspondingly improved, and scanning and computing speed of a database are improved.