60results about How to "No need for human judgment" patented technology

The invention provides a microblog rumor recognizing method and system based on propagation structures. The microblog rumor recognizing method includes the steps that according to the machine learning method of a support vector machine, feature data of information features, user features and propagation features of microblogs serves as data, a graph kernel function is calculated with a random walking method, and a support vector machine model is accordingly built in cooperation with a classical radial basis kernel function and used for recognizing rumor; each microblog is represented with a propagation tree and a feature vector, wherein the propagation trees are used for representing the propagation structures in the propagation features, and the dimensions of the feature vectors are the information features, the user features and the propagation features of non-propagation structures; when the microblogs are recognized, the to-be-recognized microblogs are classified through the support vector model built with a model building module, and whether the to-be-recognized microblogs are rumor or not is judged according to the classification result. According to the microblog rumor recognizing method and system, due to comparison of the propagation structures of microblog information, the microblog rumor identifying accuracy and the microblog rumor recalling rate can be greatly increased.

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.

A method for quickly detecting bump at an end of a highway bridge comprises: receiving a startup signal to enter a pavement detection mode; transmitting driving parameters to a test vehicle so that the test vehicle is driven according to the driving parameters; acquiring vertical section elevation data that is a distance between a vehicle body and a pavement when the test vehicle is driven; judging whether bump occurs on the pavement according to the acquired vertical section elevation data; reporting pavement bump results.The method enables quick and accurate automated judgment for bump positions on the pavement; no manual judgment is required, and the whole detection process is safer and more accurate.

Embodiments of the present invention provide a neural network based pulmonary nodule image recognition method and device. The method includes: acquiring pulmonary nodule training sample data, whereinthe pulmonary nodule training sample data includes pulmonary nodule training sample data marked with benign attributes and pulmonary nodule training sample data marked with malignant attributes; training a three-dimensional convolutional neural network by using the pulmonary nodule training sample data as input data; acquiring a pulmonary nodule image to be tested, inputting the pulmonary nodule image to be tested into the three-dimensional convolutional neural network, extracting image features of different dimensions of the pulmonary nodule image to be tested by using the three-dimensional convolutional neural network, and obtaining the benign and malignant discrimination results of each pulmonary nodule in the pulmonary nodule image according to the image features of different dimensions. The three-dimensional convolutional neural network of the embodiment of the invention can accurately distinguish the benign and malignant pulmonary nodules, effectively improve the accuracy of thebenign and malignant distinguishing results of the pulmonary nodules, and improve the working efficiency without manual judgment by the doctor.

The invention discloses an Ethernet long network cable stream-hitting performance testing method and system. The method comprises the following steps: S1, connecting a device to be tested with a testing device and a control terminal; S2, through configuring different port parameters, testing stream parameters and testing stream transmission mode parameters, to obtain stream-hitting testing data ofthe device to be tested under the different parameters by the testing device; and S3, according to the obtained stream-hitting testing data, calculating the stream-hitting performance of the device to be tested by the control terminal. The provided Ethernet long network cable stream-hitting performance testing method is capable of realizing each duplex configuration of the Ethernet port, and theautomation test under various byte lengths and various speed configurations, greatly improving the testing efficiency, and checking whether the auto-negotiation of the Ethernet port is valid and the packet loss probability. The method uses an automated script based on a TCL script, codes are simple, the operation is convenient, a testing result can be automatically output without the artificial judgment, the testing time is greatly reduced, and the testing efficiency is improved.

The invention discloses a vehicle accident early warning method and system. The method comprises the following steps: acquiring current target feature information about at least one of a driver, a running vehicle and a driving environment within a preset time period; acquiring historical target feature information of historical accidents in a database within a preset time period before the accidents occur; calculating the similarities between the current target feature information and the historical target feature information of the historical accidents respectively, and determining the numberof similarity values greater than a preset similarity threshold to obtain the number of suspected danger judgments; calculating the suspected danger proportion of the number of suspected danger judgments to the total number of historical accidents; and when the suspected danger proportion is larger than a preset proportion threshold, determining that a current vehicle is a vehicle which needs attention. The safety accident probability of the current vehicle is obtained through comparison of the similarity between the current target feature information and the historical target feature information. Moreover, monitoring and early warning are performed on the driving of the vehicle automatically, so that the probability of accidents is lowered.

The invention provides an automatic winding unit for rubber parts. The automatic winding unit comprises an extruding machine device used for extruding rubber, a profiling device used for carrying outrolling forming on the extruded rubber, a winding main machine used for winding profiled rubber strips and a control system; the winding unit further comprises an automatic rubber strip treating device, wherein the treating device is located between the profiling device and the winding main machine, and a cutting device is arranged on the automatic rubber strip treating device; and the automatic rubber strip treating device is used for automatically cutting the profiled abnormal rubber strips through the cutting device, and the cut abnormal rubber strips are automatically removed. The windingunit has the advantages of being high in automation degree, manpower is saved, the winding efficiency is high, and the quality is improved.

The invention provides a structural damage recognition system and a recognition method based on time series symbolization, and relates to the technical field of large civil engineering structure health monitoring. The invention aims at the problems of big data volume, complexity in calculation, big possibility of failing to report, and incapability of recognizing the structural damage in time of the traditional structure health monitoring method. The method comprises the following steps: 1) collecting and storing a structural non-healthy state series histogram; 2) obtaining the real-time accelerated speed time interval data of a current structure to be monitored, and processing to generate a real-time state series histogram; 3) judging whether the real-time state series histogram is matched with the structural non-healthy state series histogram or not, if so, representing that the current structure to be monitored is damaged, and if not, representing that the current structure to be monitored is under the healthy state. According to the invention, the data transmission amount between system functional units is reduced, the data transmission speed is quickened, all abnormal states of the structure are guaranteed to be recognized, the report failure phenomenon is avoided, and the accuracy of a judgment result is guaranteed.

The invention relates to a method for determining a photoetching process window and relates to the semiconductor integrated circuit process. The method is characterized in that firstly, selecting a plurality of types of graphs as graphs for judging an optimal exposure process window, then setting an energy-focal length matrix during exposure of a photoetching machine, exposing wafers, measuring line widths of the graphs of different exposure units, collecting line width sizes of the selected various types of graphs, forming a Poisson curve according to the line width sizes, removing the energy-focal length matrix corresponding to unreasonable angle data, and thereby accurately determining the optimal process window of photoetching. The method is advantaged in that new products are importedquickly, the production capacity is improved, manual judgment is not needed, and errors caused by manual judgment are avoided.

The invention provides a grass height detection device and a mower. The grass height detection device is mounted on a moving part and includes a shift plate and a height measuring mechanism; the upperend of the shift plate is rotationally connected with the moving part through a rotary shaft; the lower end of the shift plate is above a ground; the axis of the rotary shaft is perpendicular to a moving direction of the moving part such that when the moving part is in motion, the shift plate is driven by grass to rotate; the height measuring mechanism helps detect real-time rotational angle of the shift plate; the greater the rotational angle of the shift plate, the taller the grass, and on the contrary, the smaller the rotational angle of the shift plate, the more dwarf the grass; grass height can be judged according to angle changes, the grass is quantified to great extent, fewer manual judgments are required, and obtained results are accurate; since manually judging grass height by eyes is not required, automation level is high, workload is reduced for workers, and working efficiency is improved; a work strategy can be adjusted according to grass height, the work strategy with lowlabor intensity can be achieved, and energy consumption is reduced.

The embodiment of the invention provides a claim settlement reference information generation method and device. The method is applied to a server, and comprises the steps of receiving medical record data transmitted by a user terminal; sending the medical record data to an input terminal, and receiving prescription data sent by the input terminal based on the medical record data, wherein the prescription data comprises disease information, targeted drug information, target information and a target detection result; judging whether a preset special drug insurance claim settlement condition is met or not according to the disease information, the targeted drug information, the target information and the target detection result; and generating claim settlement reference information according to a judgment result. According to the embodiment of the invention, the prescription data can be collected online, whether the special drug insurance claim settlement condition is met or not is automatically judged according to the prescription data, manual judgment is not needed, and the probability of auditing errors is reduced.

The invention is applicable to the technical field of electronics, and provides a vehicle program updating system and method, an updating tool and a storage medium. According to the method, related information of a plurality of to-be-updated vehicles is sent to the plurality of to-be-updated vehicles through a server. The updating tool obtains updating packages of the to-be-updated vehicles from the server. When the updating tools update the to-be-updated vehicles, vehicle types are determined according to intelligent key flag position of the to-be-updated vehicles. Then, equipment corresponding to the to-be-updated vehicles according to the vehicle types and other relevant information is updated to target versions. Corresponding terminals of the to-be-updated vehicles according to types of the to-be-updated vehicles are updated to appointed versions. Meanwhile, manual judgment is not needed. Accuracy is high, and the loss of an online car-hailing management company can be effectivelyreduced.

The invention discloses a ship plate groove diagram drawing method based on CAD. The method comprises the steps of starting a CAD program interface; displaying an input box for a user to input ship plate thickness parameters, six groove commands for the user to select, a profile code selection box for the user to select, an amplification factor selection box for the user to select an amplificationfactor, and a proportion selection box for the user to select a printing proportion on the program interface; obtaining a ship plate thickness parameter input by a user, obtaining a profile code selected by the user, obtaining a target groove command clicked by the user, obtaining an amplification factor selected by the user, obtaining a printing proportion selected by the user, drawing a grooveapplication graph corresponding to the target groove command, and generating a profile code identifier right above the groove application graph; and generating a drawing proportion identifier right below the profile code identifier, and generating an amplified graph amplified by the amplification factor. After the plate data is input, the groove application graph can be automatically drawn and identified.

The invention discloses an automatic temperature control system and method, and belongs to the field of temperature control. The system comprises an outdoor temperature sensor, a controller, a heatingventilation air conditioning system and an outdoor temperature sensor which is arranged outdoors; the controller is connected with the outdoor temperature sensor and the heating ventilation air conditioning system and used for judging that the current time is within the preset refrigeration time period or in the preset heating time period in real time; if it is judged that that the current time is neither within the preset refrigeration time period nor in the preset heating time, the controller is further used for stopping the process; if it is judged that that the current time is within thepreset refrigeration time period, the controller is further used for determining the working mode of the heating ventilation air conditioning system to be in the refrigeration mode; and if it is judged that that the current time is within the preset heating time period, the controller is further used for determining the working mode of the heating ventilation air conditioning system to be in the heating mode. Manual judgment is not needed, the controller can conduct automatic judgment, manpower resources are saved, and the efficiency of the method of the determining work is improved.

The invention discloses a method and a device for determining network events in a communication network. The invention adopts a main technical scheme that: at least one signaling corresponding to a set field is extracted from a selected test data source according to the field, wherein each signaling corresponds to a sub time period in a corresponding time period corresponding to the test data source, and carries the status of the signaling; and the state carried by each extracted signaling and the state of a mobile terminal in a sub time period corresponding to the signaling are compared, andthe network events corresponding to the comparison results obtained by comparison respectively according to the correspondence between the comparison results and network events. When the technical scheme is adopted, the accuracy and efficiency of the results of the inspection of the operation quality of the network are improved, and the operation is convenient, and expansion is easy.

The invention discloses a method and a system for establishing a personnel distribution model based on signal strength and video. Through WLAN and video surveillance system in modern buildings, basedon the two new statistical methods, personnel data is obtained, through comparative processing, data statistics with highest accuracy is obtained and according to environmental changes, updating modelparameters is implemented. Compared with other existing systems, on the one hand, the system can obtain real-time data and obtain high accuracy results after processing, on the other hand, the systemcan automatically update the model according to the actual building environment changes to ensure the accuracy.

The invention discloses a measurement system and a measurement method for optical fibre morphology, which are mainly used for solving the problems of the need of human judgment, low accuracy, the need of consumption for lots of time, and incapacity of rapidly measuring in the existing measurement method. The measurement system comprises an objective table (1), a structure support (3), a displacement table (4), a controller (7), a shooting system (10) and an image processor (11). During a measurement, the measured optical fibre is fixed on the clamping groove of the objective table, the displacement table is controlled to move by the controller, then the shooting system is driven to carry out focusing shooting on the optical fibre for many times to acquire an optical fibre image, and the optical fibre image is conveyed to the image processor to calculate, so as to obtain the morphology parameters of the optical fibre, that is, an axial length, an end surface radial diameter, the diameter of the middle thinnest part, and a surface inclination angle. According to the measurement system and the measurement method disclosed by the invention, a human judgment is not needed, and the measurement accuracy and measurement speed are improved by evaluating the average value by measuring for many times and automatically processing the image by the image processor; and the measurement system and the measurement method can be used for screening qualified optical fibres products and improving the performances of optical fibre devices.

The invention relates to a headlamp illumination detection method. The method is characterized by carrying out automatic alignment adjusting of a luminosity detector; when the luminosity detector anda locomotive headlamp are in an alignment state, through a stepping transmission mechanism, automatically making the luminosity detector carry out X direction and Y direction stepping shifts, whereina device automatically records each stepping position of the luminosity detector and a headlamp illumination value measured at the stepping position; through calculation, finding the position of the highest headlamp illumination point and determining whether the highest headlamp illumination point is at a standard position; and if the highest headlamp illumination point is not at a standard position, finding an offset direction and an offset distance, and carrying out automatic adjusting subsequently. In the invention, a whole detection and adjusting process is completed automatically by the device, manual determination and operation are not needed, labor saving and time saving are achieved, and accuracy and reliability are high.

A drill rod of a core drilling machine for road detection, disclosed in the invention, comprises a drill rod body and a connecting shaft, a plurality of first grooves are formed in the inner wall of the drill rod body, movable blocks are installed in the first grooves, and baffles are fixedly connected to the side faces of the sides, away from the bottoms of the first grooves, of the movable blocks; second grooves with the same number as the first grooves are further formed in the inner wall of the drill rod body, the ends, away from the connecting shaft, of the second grooves communicate with the first grooves correspondingly, extrusion plates are arranged in the second grooves, and the inclined faces of the extrusion plates face the bottoms, away from the second grooves, of the second grooves; the side face of the side, close to the connecting shaft, of each movable block is provided with a slope section and a plane section and attached to the inclined faces of the extrusion plates, a plurality of driving mechanisms used for driving the extrusion plates to extrude the movable blocks are further arranged in the drill rod body, and a plurality of reset assemblies used for resetting the movable blocks are installed on the inner wall of the drill rod body; and the drill rod body is provided with a moving assembly used for driving the extrusion plate to slide in the first groove. The drill rod has the effect of improving the convenience of taking out the core sample.

The invention discloses an air quality monitoring device based on artificial intelligence, and relates to the field of environment monitoring. The air quality monitoring device comprises an air quality detection box, a control box, a vertical rod, an LED display screen, a storage battery and a tripod. The air quality detection box, the control box and the LED display screen are all fixedly mounted on the outer wall of the vertical rod; the air quality detection box, the control box and the LED display screen are all powered by the storage battery, and the tripod is fixedly mounted at the bottom end of the vertical rod. According to the invention, the driving motor drives the grooved wheel to intermittently move in a moving and stopping manner, the air inlet pipe is driven to rotate by 90 degrees every time, and the air inlet cover is arranged at the top end of the air inlet pipe, so that air in different directions can be extracted every time for quality monitoring, the universality of gas samples is enriched, and air monitoring data in different directions can be comprehensively counted; the accuracy of monitoring data is greatly improved; meanwhile, early warning is more intelligent, manual judgment is not needed, remote control is carried out through mobile terminal equipment, manual operation is not needed, and use is more convenient.