96results about How to "The statistical results are accurate" patented technology

The invention discloses a crowd density estimation method based on video analysis and a pedestrian volume statistical method based on the video analysis. The crowd density estimation method includes the flowing steps of (1) off-line training: manually counting crowd density data, extracting characteristics and conducting training; and (2) on-line estimating: extracting the characteristics and conducting regression prediction by utilizing trained model parameters. The pedestrian volume statistical method includes the step of setting up a robust relationship between a scene and a line-passing number of people by combing the crowd density and a micro-region pedestrian flow speed before a line is passed. Characteristics such as foregrounds, edges and gray scale co-occurrence matrixes are extracted based on a whole area to conduct crowd density estimation, problems of dense crowds, sheltering and the like can be well solved through mixing of the characteristics, and real-time crowd density estimation is achieved. In addition, on the basis of area crowd density estimation, pedestrian volume estimation is conducted through combination of the pedestrian flow speed based on an optical flow, detection and tracking of a large number of individuals under a complex environment are avoided, and two-way pedestrian volume counting of accurate robust under dense crowds is achieved.

The invention provides an Internet information classification method and system, and belongs to the technical field of computers. The system provided by the embodiment of the invention comprises a word segmentation module, a preset keyword library module and a matching module. The method comprises the following steps of: obtaining comments input by users, and segmenting the comments to obtain keywords; matching the keywords with a preset keyword library to obtain an emotion value corresponding to each keyword; storing at least two classes of keywords in the preset keyword library, pre-storing at least one keyword and the emotion value corresponding to the keyword in each class respectively; and obtaining the emotion value of the comment according to the emotion value corresponding to each keyword. In the embodiment of the invention, by the preset keyword library and the emotion values corresponding to the keywords, the emotion values corresponding to the comments of the users obtained by a word segmentation and matching mode are obtained. Accordingly, compared with an artificial statistic method in the prior art, the statistical results are more accurate and efficient.

The method includes steps: obtaining talk tone in user communication; using speech recognition to convert content of talk tone to text information, and analyzing and obtaining user id corresponding to character of voice print of talk tone, saving user's behavioral characteristic information and related user id. The invention also discloses equipment for collecting user's behavioral characteristics. The said equipment includes unit for acquiring voice, speech recognition unit, semantic analysis unit, user in recognition unit, and unit for storing behavioral characteristics. Through talk tone, the invention can acquire user's behavioral characteristics.

The invention discloses a medical standard term ontology management system and method, a device and a storage medium, belonging to the technical field of life and health. The system comprises a disease terminology library, a terminology inquiry module, a terminology customization module and a foundation platform construction module, wherein the disease terminology library displays the collected data in a tree structure, and simultaneously displays the details of the terminology, synonyms and parent-child node information; the term query module is used for creating term index, constructing termquery system, and realizing term fast positioning. The terminology customization module is used for establishing the terminology system belonging to different departments by relying on the disease terminology database and combining the practical application of different departments' business systems. The basic platform construction module is used for managing different organizations in the medical union to manage users of different organizations. The system has a uniform classification definition, can meet the needs of the development of medical and health field and industry in our country, and can be maintained and updated in a timely manner.

The invention discloses a people counting method based on video analysis. The method comprises the following steps of inputting a video image, acquiring a foreground picture by using a background subtraction method, and clustering the foreground picture into a plurality of blocks; extracting features, and performing perspective correction on the features; estimating the number of people by using a two-layer regression model, wherein the first-layer regression divides the different blocks in each frame into different disperse density layers, and the second regression combines virtual standardization and regression counting as a joint learning process; training a counting model synchronously covering virtual standardization and regression problems for the different disperse density layers; and at last counting each block by using the different regression models according to the different crowd density layers, and accumulating the numbers of all blocks to obtain the number of people. According to the people counting method, based on the two-layer regression model, and through combining the vision normalization with the number regression, the defect of a single regression model is overcome, and better robustness and adaptability are provided for shielding multi-density crowd scenes and crowds and segmenting incomplete images.

The invention relates to the technical field of image processing, and aims to provide a pedestrian flow statistics method based on deep learning and multi-target tracking. The method mainly comprisesthe following steps: S1, shooting a pedestrian monitoring video and reading an image in the video; S2, setting an effective area and a flow count of the image; S3, constructing a pedestrian detectionmodel based on deep learning and training the pedestrian detection model; S4, performing current pedestrian detection to obtain coordinates and image blocks of a current pedestrian frame; S5, trackingthe current pedestrian by using a multi-target tracking algorithm based on deep learning, and generating coordinates of the current pedestrian; S6, generating a moving track of the current pedestrian; S7, judging whether the current pedestrian leaves the effective area or not; If yes, entering the step S8, and if not, entering the step S4; S8, a noise threshold value is selected, and noise judgment is carried out; And S9, deleting the coordinates of the current pedestrian in the continuous video frames. According to the invention, an accurate flow statistics result can be provided in an actual use scene.

The invention relates to a people counting system. The people counting system is characterized by comprising a human body target detection module and a human body target tracking counting module. The human body target detection module comprises a Kinect depth transducer, an interesting target detection module and a human face target detection module. The human body target tracking counting module comprises a target model set, a first matching module, a second matching module and a counting module and is used for counting human body target contours in the target model set. Compared with the prior art, the people counting system has the advantages that the system conducting people counting based on the Kinect depth sensor is adopted, counting results are accurate, and the system can adapt to any scene of a controlled environment and can conduct people counting in a detection space in real time.

The invention relates to a traffic flow statistics method based on the image visual processing technology. Compared with the prior art, thedefect that comprehensive statistics of the traffic flow can not be achieved according to the video image technology is overcome. The traffic flow statistics method based on the image visual processing technology comprises the following steps of conducting pre-processing, wherein a virtual circle and a detection line are drawn for a detected lane, and manual statistics is conducted under the vehicle saturation condition; establishing an initial background image, wherein the background image BG is constructed according to the frame difference method; conducting vehicle monitoring and forming a vehicle binary image Object through a difference image DI; obtaining the image information of vehicles in the virtual circle; judging the state of the vehicle in the virtual circle, wherein whether vehicles exist in the virtual circle and the virtual circle is in a vehicle saturated state or not is judged by calculating the ratio of overlapped images to the detection line; conducting traffic flow statistics, wherein the traffic flow in the unsaturated state and the traffic flow in the saturated state are calculated, and thus a traffic flow statistics result is obtained. By the adoption of the traffic flow statistics method based on the image visual processing technology, the traffic flow in the unsaturated state can be obtained, the traffic flow in the saturated state can also be obtained, and the statistics result is more accurate.

The invention relates to a method and device for intelligently switching to an HDR mode of a camera. Computer program instructions are used for controlling a computer to complete the camera method for intelligently switching the HDR mode through establishing of a functional module framework, and the computer program instructions are stored in computer readable storage media. The method includes the steps that the duty ratio of pixels of which the brightness value is smaller than a preset low brightness value and the duty ratio of pixels of which the brightness value is larger than a preset high brightness value in an image of a current scene of a camera are estimated, and switching to the HDR mode can be achieved only when the duty ratio of the pixels of which the brightness value is smaller than the preset low brightness value is larger than a preset low brightness ratio threshold value and the duty ratio of the pixels of which the brightness value is larger than the preset high brightness value is larger than a preset high brightness ratio threshold value. In combination with the characteristics of the HDR mode, the estimation of the duty ratio of the pixels of which the brightness value is smaller than the preset low brightness value and the duty ratio of the pixels of which the brightness value is larger than the preset high brightness value in the image of the current scene can really reflect brightness distribution conditions of the actual scene, and accurately judge whether or not switching to the HDR mode needs to be performed to take photos.

The invention relates to a wind power climbing event probability scene prediction method. According to the method, distribution characteristic consistency and time sequence characteristic consistency are achieved through building of a multi-objective fitness function of a cumulative density function and a high-order moment autocorrelation function. On the basis of a genetic algorithm of the multi-objective fitness function, iterative optimization is conducted on probability generation model parameters to obtain a large number of prediction scenes, and climbing event probability characteristics excavated from scene capture zones are used for evaluating the prediction method. Actual data of some aboard wind field are selected for conducting example calculation and statistical analysis, the result shows that the statistical result of the multi-objective function is more accurate than that of a single-objective function, the probability scene prediction method can accurately estimate the characteristic quantity of a climbing event, the correctness of the method is proved, and the method can provide guidance for a probability scene generation method and a climbing event prediction model.

The invention relates to a floating car counting method based on the video monitoring processing technology. Compared with the prior art, the defect that floating cars can not be comprehensively counted through the video image technology is solved. The floating car counting method includes the following steps: carrying out preprocessing, wherein virtual line circles and detection lines are drawn for measured lanes, and manual counting in the car saturation state is carried out; building an initial background image, and constructing a background image BG with a frame difference method; monitoring cars, and forming binary images O<bject> of the cars through difference images DI; obtaining the information of images of the cars in the virtual line circles; judging the state of the cars in the virtual line circles, wherein whether the cars exist in the virtual line circles or not and whether the virtual line circles are in the saturation state or not are judged by calculating the ratio of intersection images to the detection lines; counting the floating cars, wherein the floating cars in the non-saturation state and the floating cars in the saturation state are respectively calculated, and the floating car counting result is obtained. By means of the floating car counting method, the flow number of the cars in the non-saturation state can be obtained, the flow number in the saturation state can further be obtained, and the counting result is more accurate.

The invention relates to the technical field of education informatization, and discloses a classroom efficiency detection method and system. The method comprises the following steps of: shooting an image comprising all the students in a classroom; obtaining a sub-image corresponding to each human face area in the shot image; and recognizing the sub-images and judging whether the corresponding human face areas comprise front faces or not, and if the judging result is positive, carrying out classroom efficiency statistics according to front face information. According to the method, the front faces of all the students during the teaching of a teacher are judged by utilizing human face recognition, so that the classroom efficiencies of the students and the teacher can be detected, the classroom efficiencies can be assessed and the assessment is more perspective.

The invention provides an automatic layout method of a building pipeline system. The layout method comprises following steps: S1. unfolding graphing paper to manipulate layers and blocks in the graphing paper for easy identification and retrieval of pipelines and annotations for pipelines required layout operation; S2. obtaining pipeline segments and annotations from the operated graphing paper; S3. performing pipeline resolving according to a set algorithm according to the obtained pipeline segments and annotations; S4. automatically generating multiple components having attributes, methods and being responding to specific events according to the resolved data; a building information model is formed by combining the components, that is, the automatic layout of pipeline system is completed. By means of the method of the invention, the layout is accelerated; the number of personnel for layout is reduced; the accuracy of layout is increased; the capacity requirement for the layout workers is reduced; the layout is more reasonable and raw material is saved; there is less error in the layout graphs so that the method is easy to be accepted by all parties with less divergence and the work progress is accelerated.

The embodiment of the invention discloses a user status accounting method, a device and a system. The method comprises steps: after a center server detects that an accounting server working abnormally exists, a target accounting server is selected from idle accounting servers; the network identifier of the accounting server working abnormally in a preset online server identifier collection is deleted, the network identifier of the acquired target accounting server is recorded into the online server identifier collection, and an online server identifier collection after upgrading is obtained; and the online server identifier collection after upgrading is sent to a service server so as to enable the service server to send service data to the accounting server indicated by the network identifier in the online server identifier collection after upgrading. Thus, an abnormal accounting server can be quickly replaced, and data loss can be effectively reduced so as to obtain a finally-accurate accounting result of users online at the same time.

The invention provides a plugin processing method, a client side and a cloud server. The method comprises the steps that log information of multiple plugins of the client side is acquired and uploaded to the cloud sever, the cloud sever carries out statistics on the downloading quantity of each plugin within a time threshold value and the use state of each downloaded plugin according to the log information, and statistical results are obtained; a feedback message issued by the cloud server is received, wherein the feedback message carries the statistical results corresponding to the multiple plugins; the multiple local plugins are processed according to the statistical results. According to the plugin processing method, the statistical results which are fed back by the cloud server and correspond to the plugins in the current client side are received, and therefore the plugins can be processed more reasonably according to the statistical results, and user experience can be improved.

The invention belongs to the technical field of intelligent traffic, and discloses an intelligent flow identification and statistics method for a complex traffic intersection, and the method comprisesthe steps: constructing a complex traffic intersection video data set, and training a depth detection model; automatically dividing the intersection into detection areas through mouse point selection, and numbering the detection areas; detecting a vehicle in the video picture by using the trained deep learning model; predicting the position of the next frame of the vehicle by using a target tracking algorithm according to the detected vehicle and the tracking vehicle of the previous frame; counting the traffic flow according to the change condition of the number of the intersection where thevehicle is located; and repeating the step 3 to the step 5 until the video is ended or manually stopped. According to the method, the vehicle types can be automatically recognized in various complex traffic intersection environments, the number is counted according to the vehicle types, the traffic flow between different intersections is counted, and the method has the advantages of being accuratein statistical result and good in real-time performance and has high engineering practical value.

The invention discloses a method for carrying out decision-making road section division on a basic road section based on road condition indexes. The system comprises a central processing module, an information transmission module, a road surface type collection module, a traffic information collection module, a road surface structure and a road surface detection method. The central processing module comprises a wireless monitoring module, a data storage module, a data analysis module and a road section maintenance division module, and the road surface type collection module comprises a road surface structure collection module, a road surface disease collection module and a road surface structure collection module. The method has the advantages that the method for carrying out decision-making road section division on a basic road section based on road condition indexes can comprehensively judge the damage degree of the vehicle to the road surface, the statistical result of the traffic volume can be more accurate, road surface diseases can be maintained in a targeted mode, the blindness of maintenance work is reduced, inaccurate basic road section division caused by a single basis isavoided, and road section maintenance is more facilitated.

The invention discloses a method and device for recognizing user access trace. The method includes: acquiring an equipment identity and an IP address correlated to the equipment identity in a websitelog file, and performing md5 verifying to acquire a user ID value; screening log files according to the user ID value to acquire log files with the same user ID value; inquiring whether each log fileacquired by screening contains a user access identity, excluding the log files not containing the user access identity to determine access log of a same user, and recognizing access trace of the useraccording to the determined log file, wherein the user access identity is automatically generated and stored by a web server when the same user accesses to a network address not for the first time andstored in the log file of the user by the web server when the user accesses to the network address again. By utilizing the method and the device, the access trace of the user can be recognized through the network access log.

The invention discloses an automatic test system and method for switching equipment in a rail transit system. The system comprises a test control module used for performing test control, an equipmentmanagement module used for managing accessed network device and more than one piece of test equipment used for executing tests. The test control module is connected with the test ends of the test equipment and the tested equipment through the equipment management module. The debugging end of the tested equipment is connected with the test control module, and the test control module configures thetest equipment and the connection between the tested equipment through the equipment management module so as to carry out required network topology configuration and controls the test equipment to test the tested equipment according to a required test task. The method can achieve the automatic testing of the switching equipment in the rail transit system, and has advantages of wide testing range,high testing efficiency and testing precision, and facilitating the implementation of complex topology testing and case reproduction.

The present invention discloses a pedestrian recognition method, an apparatus and an electronic device. The method includes the following steps that: a plurality of frames of scanning data of a laser radar are obtained, wherein the scanning data are distance information of each scanning point which is obtained by the laser radar through obliquely scanning the passage surface of pedestrians at a preset angle; the gray-scale values of foreground points contained in the plurality of frames of scanning data packets are determined according to the distance information of the scanning points contained in the plurality of frames of scanning data packets, and the grey-scale map of the foreground points is obtained according to the determined gray-scale values; region growing is performed on the basis of the priori model of the gray-scale distribution of the foreground points obtained by scanning the pedestrians, and connected areas in the grey-scale map are determined; and the pedestrians on the passage surface of the pedestrians are recognized according to the determined connected areas. According to the method provided by the invention, the gray-scale values of the foreground points contained in the plurality of frames of scanning data packets are spliced into the gray-scale map, so that the pedestrian targets on the passage surface can be recognized, and therefore, the accuracy of the statistical result of passenger traffic volume can be improved.

The invention discloses a method for making a statistic of a hot degree of a subject, a push method and device of programs. The push method comprises the following steps of acquiring the EPG (Electronic Program Guide) data of each television channel, wherein the EPG data of one channel comprise names of programs broadcasted in the channel and the broadcast time; searching information relevant to each program according to the EPG data, wherein the relevant information comprises one or several of program types, program subjects, program main performers and program hosts; determining the programs which satisfy the preset subjects according to the searched information relevant to each program; acquiring the determined relevant data of programs through the EPG; and finally making a statistic of the hot degree of the subject according to the relevant data of the program. According to the method for making a statistic of the hot degree of the subject provided by the invention, the EPG data of the television programs are used for making a statistic of the hot degree of the preset subject, which not only saves a large amount of operation cost of data collection, but also makes the statistical results of the subject hot degree more convincing.

The invention relates to an optimization method of a PIK3CA gene H1047R mutation digital PCR detection system, wherein the detection system includes upstream primers, downstream primers, wild-type probes, mutant-type probes, a wild-type template and a mutant-type template; the wild-type template is normal human gDNA after enzyme digestion, and the mutant-type template is a mutant plasmid insertedwith a PIK3CA gene H1047R mutant fragment after enzyme digestion; a standard substance is prepared from the mutant-type template and the wild-type template according to a certain proportion of copy number; reaction is performed with a medium mutation frequency standard substance as a template by digital PCR, a data statistical graph is prepared according to the reaction data, and a wild-type fluorescent region and a mutant-type fluorescent region are selected. Reaction is performed with the wild-type template as the template by digital PCR, and the background threshold value is determined. A detection product optimized by the optimization method of the PIK3CA gene H1047R mutation digital PCR detection system has higher accuracy degree.

The invention provides a method for solving historical data change in a big data system, which comprises the following steps of: setting an update time field for a business table to mark the update time of the field; executing the operation type data storage ODS task once in the early morning every day, checking the updating time field of the service table , and generating a piece of operation type data storage ODS layer data; enabling the zipper table module to check the latest record of all the data of the operation type data storage ODS layer, comparing the latest record with the corresponding record of the last time, and recording the event that the table changes in the day into a change notification table if the latest record changes; and after the zipper table task is finished, executing a task for generating a service daily table, checking the change notification table by the task, executing a corresponding task unit according to the changed table and the changed date, and generating service daily table data. According to the method, the problems of when to recalculate and which part to recalculate after historical data changes can be solved, so that a large number of computing resources are saved.

The invention discloses an online foreign matter removing device for tobacco shred making. The foreign matter removing device comprises a supporting machanism, a detection mechanism and a removal mechanism; the detection mechanism comprises an X-ray machine and a detection module composed of a plurality of detectors, wherein the X-ray machine is located above a conveyor belt, and the detection module is located below the conveyor belt; and the removal mechanism is located below the discharge end of the conveyor belt, and comprises a foreign matter remover and a remover bracket, wherein the foreign matter remover comprises a plurality of high-pressure nozzles, and electromagnetic valves each are arranged at the air inlet end of each of the high-pressure nozzles. The online foreign matter removing device has the advantages that foreign matters in the tobacco shred making process can be imaged and displayed so that online impurity identification and accurate removal can be more accuratelyrealized, the online identification and removal of magnetic, low-magnetic and non-magnetic metal objects can be realized at the same time, removal effect is good, precision is high, practicability ishigh, and quality of tobacco leaves and working efficiency of tobacco shred making are indirectly improved.

The invention discloses a virtual queue time counting method and a virtual queue time counting system based on a Bluetooth beacon. The counting method comprises the following steps: a client uploads Beacon broadcast information, GPS information and hardware information, which are received in real time, to a server in real time; the server carries out Beacon broadcast information verification, hardware information verification, and geographic location comparison verification of the GPS information and the Beacon broadcast information in real time; and the server records the virtual queue time in real time when and only when the Beacon broadcast information verification, hardware information verification, and geographic location comparison verification of the GPS information and the Beacon broadcast information carried out in real time succeed. Compared with the prior art, the statistical result obtained by the virtual queue time counting method and the virtual queue time counting system based on a Bluetooth beacon of the invention is accurate and fair, and the statistical result can be used for reference for businesses to offer discounts and convert into virtual currency.

The invention relates to an optimization method of a BRAF gene V600E mutation digital PCR detection system, wherein the detection system includes upstream primers, downstream primers, wild-type probes, mutant-type probes, a wild-type template and a mutant-type template; the wild-type template is normal human gDNA after enzyme digestion, and the mutant-type template is a mutant plasmid inserted with a BRAF gene V600E mutant fragment after enzyme digestion; a standard substance is prepared from the mutant-type template and the wild-type template according to a certain proportion of copy number;reaction is performed with a medium mutation frequency standard substance as a template by digital PCR, a data statistical graph is prepared according to the reaction data, and a wild-type fluorescentregion and a mutant-type fluorescent region are selected. Reaction is performed with the wild-type template as the template by digital PCR, and the background threshold value is determined. A detection product optimized by the optimization method of the BRAF gene V600E mutation digital PCR detection system has higher accuracy degree.