631results about How to "Improve judgment accuracy" patented technology

The invention provides an indoor inertial navigation algorithm based on posture recognition and a step length model and relates to the field of indoor inertial navigation and positioning. An inertial sensor is fixed on a foot of a pedestrian and data is transmitted to a smartphone by utilizing Bluetooth; the step frequency and the step number of the pedestrian are calculated by adopting a multi-condition and zero-speed detection method and a posture angle detection method according to posture characteristics of the foot when the pedestrian walks; an inertial navigation calculation result of each step of the pedestrian is counted and classified and the step length model is established; postures and the step frequency of the foot of the pedestrian are used as reference to calibrate the inertial navigation calculation result in real time; electronic compass data of a dynamic complement algorithm based on slide mean filter is used as a real-time course angle, and whether the pedestrian has an inertial and usual movement posture or not is judged according to changes of the step frequency and the course angle of the pedestrian; the step length model is matched by taking a judging result as the basis. The problems of an indoor positioning and inertial navigation technology that error accumulation is caused by double integration so that positioning is inaccurate and the like are solved, and the accuracy of indoor long-distance navigation of the pedestrian is guaranteed.

The invention discloses a pattern recognition method of a supporting vector machine for fault diagnosis of a device, in particular to a device failure pattern recognition method of a reduction Tikhonov regularization supporting vector machine auto-selected by utilizing parameters. The invention comprises the following steps: (1) achieving a derivation process of Tikhonov regularization supporting models of the vector machine; (2) constructing a subsample set to reject redundant sample information by utilizing a pruning method so as to build models of the reduction Tikhonov regularization supporting vector machine; (3) taking classification accuracy as a fitness function, auto-selecting width parameters and balance parameters of a Gauss kernel function of the reduction Tikhonov regularization supporting vector machine by utilizing a heredity algorithm and by taking classification accuracy as a fitness function so as to build the models of the reduction Tikhonov regularization supporting vector machine auto-selected by utilizing parameters; and (4) verifying the pattern recognition method by utilizing failure samples of a motor device so as to indicate superiority of the method proposed in the invention.

The invention discloses a mid and low-voltage distribution network fault positioning method. The method comprises steps: 1, through a power consumption information acquisition system, a distribution line online monitoring system, a residual current operated protective device monitoring system, a dispatching automatic system, a distribution network automatic system and a production management system, power consumption information of the mid and low-voltage distribution network is acquired, the power consumption information of the mid and low-voltage distribution network is corrected through a power flow distribution algorithm, and the topological relationship of the mid and low-voltage distribution network is perfected; and 2, according to power loss information in the power consumption information of the mid and low-voltage distribution network, based on the topological relationship of the mid and low-voltage distribution network, fault study and judgment are carried out through a corresponding fault study and judgment algorithm, a fault device is determined, devices and users influenced by power failure are analyzed, and a geographic area influenced by power failure is generated when a power grid device and the geographic position information of power grid users are combined. The method of the invention has the advantages of high fault positioning precision and high power failure range judgment precision.

The invention belongs to the technical field of nuclear power station information construction, and discloses a nuclear power station information system operation safety pre-warning method and devicesuitable for the information machine room dynamic environment monitoring, a computer apparatus and a storage medium. The method comprises the steps of obtaining the power distribution and environmentparameters of the related apparatuses of an information machine room information system from a designated apparatus system; selecting the important parameters and the abnormal parameters from the operation parameters; inputting the important parameters and the abnormal parameters into a nuclear power station information system operation risk prediction model, obtaining the risk types and the risklevels outputted by the nuclear power station information system operation risk prediction model, and enabling each risk type to correspond to one risk level; and when the risk level corresponding tothe risk type reaches a preset level, sending the early warning information to a designated terminal. According to the nuclear power station information system operation safety early warning method provided by the invention, the judgment precision of the nuclear power station on the safety risk can be improved, and meanwhile, the timeliness of the systematic risk feedback is ensured.

The present invention provides a button identification device and a method thereof, which increases the precision for determining front surface and back surface of the button, sews the button to the object being sewed with a high quality and realizes the shortening of button state determination time. White irradiation light is irradiated to the upper surface of the button. Blue adjustment light is shot with a state of irradiating to the lower surface of the button. The reflection light of the irradiated light and the transmitted light of the adjustment light are received, the part except for the transmitted light of the adjustment light is received for obtaining the shot data. The light with the color same with the adjustment light is extracted from the shot data for generating the adjustment light brightness data. The light with the color different from the adjustment light is extracted from the shot data for generating the non-adjusting light brightness data. The non-adjusting lightbrightness data is subtracted from the adjustment light brightness data for generating the adjustment light data. The adjustment light data are binarized and inverted for generating the profile data.Logic and calculation are performed for the shot data and the profile data for generating the synthesized image data. The synthesized image data is compared with the respective standard image data ofthe front surface and the back surface of the button for determining the state of the button.

The invention relates to the technical field of lithium battery production equipment, and particularly relates to vacuum injection equipment of a flexibly-packaged lithium battery. The equipment sequentially comprises a material loading device, a weighing and detecting device, a flat fixture for clamping air bags, an injection device, a first standing room, a second standing room, a third standing room, a sealing device, a weighing and detecting device, a blanking mechanical arm and an unloading mechanism from left to right. The equipment provided by the invention can be used for improving the production efficiency, the detection precision, the judgment accuracy of the product quality, the sealing quality and the product quality, reducing the electrolyte waste, effectively avoiding the phenomenon of a loading direction error and reducing the working strength of workers.

The invented method is in particular for eliminating rolled dark fringes occurred in procedure of displaying digital image when parameter setting of pickup camera is not suited to external light source. First, states of pickup camera are switched to different parameter settings in sequence, and whether there are dark fringes appeared on image is determined. Then, based on instance of appearing rolled dark fringes under different parameter settings of pickup camera, type of light source at that time is determined. Finally, parameters of pick-up head are modified to suite to ambient light in order to eliminate dark fringes. In the invention, PC completes most of operation, providing high estimating accuracy and low calculated amount.

An automatic tracing type solar generator belongs to the technical field of solar electric power generation, which comprises a chassis, a supporting structure with a rotating shaft, a light concentrating thermal-collecting device, an accumulator, a driving device and an automatic tracing controlling device, wherein the supporting structure is connected on the chassis through a first driving device, the light concentrating heat-collecting device is mounted on the rotating shaft of the supporting structure and is connected with the supporting structure through a second driving device, the driving motor and the automatic tracing type controlling device of the two driving devices are electrically connected, and the light concentrating heat-collecting device is electrically connected with the accumulator. The invention has good property and long service life. The automatic tracing type lighting system combines input ends of four sensing devices to judge the position, which increases judging accuracy. The circuit is simple and reliable, power consumption is low, and cost is greatly reduced compared with a single-chip processer tracker. The invention effectively utilizes solar, which is nonpolluting, no discharging, saves energy and protects environment. A light collecting cover can automatically turnover upwards, has the function of snow resistance, rain resistance and automatic protection, and effectively protects the damage of a motor.

The invention discloses a multi-feature fusion based infrared forest fire judgment method. The method comprises the following steps: A, inputting an infrared image sequence; B, performing binarization processing; C, performing morphological processing; D, extracting flame features; E, performing SVM classification judgment on the flame features; F, performing classification decision-making by utilizing a Bayes classifier; and G, judging whether a fire occurs or not through data in the step F. The flame features in the step D include an infrared flame height characteristic, an infrared flame gradient characteristic, an infrared flame moving characteristic and a flame spectrum characteristic. According to the multi-feature fusion based infrared forest fire judgment method provided by the invention, the fire can be detected quickly, accurately and safely in the early stage of the forest fire, so that the judgment precision of forest fire prevention is greatly improved and the working efficiency is improved.

The invention provides a surface conduction examination method and a substrate examination device, using the optical system which is the same as its prior art to promote the tilting angle resolution, thereby enhancing the examination accuracy. A substrate examination device having a light projecting section 4 consisting of the light sources which emits respectively the R, G, B coloring light and are installed respectively in different azimuth directions with respect to the examination region including soldering portion., extracts 1 or 2 color ingredients which have intensity higher than that of average value. The tilting surfaces adapting to each of light sources 8, 9, 10 are converted respectively into a light and shade monochrome red, green and blue image. Further, the tilting surface at boundary position, which a adapts to light sources 8, 9 is converted into a light and shade image having a color mixed with red and green, whereas the tilting surface at boundary position, which adapts to the light sources 9, 10 is converted into a light and shade image having a color mixed with green and blue.

The invention relates to the technical field of image processing, and discloses an end-to-end image segmentation processing method and system. The end-to-end image segmentation processing method includes the steps: acquiring an image for testing end-to-end image segmentation processing, and taking the image as a sample image for training; establishing an end-to-end image segmentation network, and using the sample image to perform training learning; and according to an obtained image segmentation network model, segmenting a target image so as to realize segmentation of the target image and outputting the segmentation result of the whole target image for one time. The end-to-end image segmentation processing method has the advantages of being relatively high in accuracy of image segmentation, having no demand for segmenting small images, saving the storage resource of a computer, and reducing the time of preparing images, and can output the segmentation result of the whole image for one time, thus improving the segmentation speed and simplifying the steps of image segmentation.

The invention relates to a convolutional neural network adversarial transfer learning method based on Waserstein distance and application thereof and the method comprises the steps: employing a to-be-migrated convolutional neural network to obtain a source domain feature set and a source domain fault judgment set of a source domain mark sample set and a target feature set of a target domain sampleset; and with maximization of a Wasserstein distance between the source domain feature set and the target feature set and minimization of the sum of the Wasserstein distance and a judgment loss valueof the source domain fault judgment set as a target, realizing adversarial migration learning of the convolutional neural network based on a convergence criterion. According to the invention, the Wasserstein distance is introduced into the transfer learning of the convolutional neural network. The maximum Wasserstein distance is used as a target; the distinguishing sensitivity of the features extracted from the two sample sets is improved; and the minimum sum of the Wasserstein distance and the loss value of the source domain fault judgment set is taken as a target, so that the judgment precision of the convolutional neural network is improved, the requirements on sample data and a network structure are low while the fault diagnosis capability is ensured, and the invention can be suitablefor migration among multiple working conditions and is high in practical applicability.

The invention relates to an excavator capable of improving determination precision, thus preventing unstable state of the machine; the excavator comprises the following structures: a lower driving body (1); an upper rotary body (3) mounted on the lower driving body (1); an excavation attaching device arranged on the upper rotary body (3); an operation content detector (29) detecting operation content; a position finding device (M2) measuring position and direction of the machine; a posture detector (M3) used for detecting the posture of the excavation attaching device; an external operating device (30E) having a landscape gaining portion (33). According to the posture propelling of the excavation attaching device, the information obtained by the landscape information gaining portion (33) and related to the present shape of the ground of a working object, and the location of the excavator, the stable / unstable states of the excavator can be determined, thus providing basis for limiting excavator motions when the machine is unstable.

The invention provides a short message identification method and short message identification equipment. The method comprises the following steps of: receiving a short message and a probability characteristic vector of the short message, which are sent by a server; respectively calculating a first generating probability of the short message under a normal short message model and a second generating probability of the short message under a spam message model by utilizing the probability characteristic vector; obtaining a first joint distribution probability through the first generating probability and a normal message priori probability, and obtaining a second joint distribution probability through the second generating probability and a spam message priori probability; and when the first joint distribution probability is more than or equal to the second joint distribution probability, identifying the short message as a normal message, otherwise, identifying the short message as a spammessage. Once a terminal falsely judges a normal message as a spam message, a user still can retrieve the short message from a trash; therefore, the risk of falsely identifying spam messages is reduced. The probability characteristic vector is calculated by the server and is sent to the terminal, so the requirement on computing power of the terminal is reduced.

The invention discloses a fatigue driving preventing system and method. The system comprises terminal equipment arranged on a vehicle, a cloud side arranged on a server, and a first communication module for receiving information sent by the cloud side and sending information to the cloud side. An acquisition module acquires current physical state parameters of a driver and the preset parameter range of the cloud side, and the cloud side acquires the corresponding parameter preset range relevant to the driver according to the physical state parameters, obtained at the start initial stage of the system, of the driver and sends the corresponding parameter preset range back to the terminal equipment. A judgment module judges whether the current physical state parameters of the driver are within the parameter preset range or not. A first control module generates a corresponding control command according to the judgment result of the judgment module. An alarm module sends corresponding alarm information according to the control command. Physical characteristics of the driver are detected and uploaded to the cloud side, the cloud side obtains the parameter preset range corresponding to the driver in combination with a large amount of stored driver information, and therefore judgment precision is improved, and the misjudgment situation is avoided.

An air conditioner and a refrigerant amount determining method which are able to accurately and simply determine the amount of a refrigerant without impairing the reliability of a compressor. A refrigerant circuit (10) has a compressor (21), an outdoor heat exchanger (23) functioning as a condenser, an indoor expansion valve (41, 51), an indoor heat exchanger (42, 52) functioning as an evaporator, connecting piping (4b, 5b) inside an indoor unit, a liquid refrigerant connecting pipe (6), a gas refrigerant connecting pipe (7), and connecting piping (8) inside an outdoor unit. A controlling section (9) performs liquefying control for causing the refrigerant to be present in a liquid state in a portion extending from the indoor expansion valve (41, 51) to the outdoor heat exchanger (23). Thecontrolling section (9) directly or indirectly controls the flow rate of the refrigerant flowing in a liquid bypass circuit (70) from a liquid pool section (Q) toward the gas refrigerant connecting pipe (7). A liquid surface sensor (39) detects at least either the volume of the liquid refrigerant in a portion where the liquid refrigerant pools or the physical quantity equivalent to the volume.

The invention discloses an intelligent website content capture-preventing method and system based on a content delivery network, which can reduce the problem that a normal client end cannot visit a website due to misjudgment and can automatically cancel a visiting-prohibiting state. According to the technical scheme. The intelligent website content capture-preventing method comprises the steps of: receiving a client request and monitoring a visiting behavior of the client end by a content delivery network system; judging whether a visiting frequency of the client end exceeds a preset threshold or not, if not, directly returning the requested content; otherwise, continuously judging whether the current requesting time of the client end is more than the sum of the time how much the visiting frequency is exceeded last time and the preset grab-preventing state duration or not; if so, returning the requested content; and if not, returning to a client end verification code page; judging whether the client end submits a correct verification code or not, if so, returning the requested content, and if not, judging whether the times the client end submits illegal verification codes exceed a set value or not, if not, returning to the client end verification code page, otherwise, indicating that the client end is a grab machine, prohibiting visiting of the client end, and returning 403 response.

The invention discloses a front human face detection method based on a sensitive area. The method comprises the following steps that a virtual coil is set and the position and the size of the sensitive area are determined; the motion foreground in the sensitive area is detected and tracked through background modeling; the candidate area of the human face is detected through the Adaboost algorithm of Harr-like characteristics; some limitation conditions are set to exclude non-front human face images according to the color characteristics of the human face and simplified shape characteristics. According to the front human face detection method based on the sensitive area, due to the fact that the sensitive area is set and the motion foreground is detected, the efficiency of the follow-up Adaboost detection algorithm is improved, and the detection accuracy is improved through the later skin color and shape verification.

The invention discloses a landslide monitoring method and a monitoring system applying the same. A mountain body movement state is monitored in a mode of creatively combining a tilt sensor and GPS / Beidou differential positioning, and accurate original data is provided for landslide judgment and early warning. The GPS / Beidou-based differential positioning method is applied to monitoring landslide mass, position data of a plurality of monitoring points of a mountain body is monitored, fixed errors of satellite clock, satellite ephemeris, ionized layer time delay and troposphere time delay and the like are eliminated, and the GPS / Beidou positional accuracy is greatly improved.

The invention discloses an intelligent algorithm based electricity-larceny-preventive monitoring method for distributed type photovoltaic power generation. The method is characterized by including step 1, creating a distributed type photovoltaic power generation database, wherein database information includes historical meteorological data of a photovoltaic power station and photovoltaic power generation power data in the corresponding period; step 2, taking the meteorological data as input, taking photovoltaic power generation power values as output, and creating an electricity-larceny prevention model according to an intelligent algorithm; step 3, predicting photovoltaic power by the aid of the electricity-larceny prevention model, and calculating theoretical power generation quantity of the photovoltaic power station by the aid of a calculation model of the theoretical power generation quantity of the photovoltaic power station; step 4, inputting collected real-time power generation quantity of the photovoltaic power station and the theoretical power generation quantity of the photovoltaic power station into a power quantity abnormity judgment module to acquire an power generation electricity-larceny suspicion judgment result of the photovoltaic power station. The intelligent algorithm based electricity-larceny-preventive monitoring method has the advantages that photovoltaic power quantity abnormity judgment accuracy is high, electricity-larceny suspicion coefficients are given, auditing work efficiency is improved, and the basis is provided for effective monitoring of distributed type photovoltaic power generation.

The present invention discloses a determination method and device of long term evolution (LTE) network coverage quality. The method comprises the steps of obtaining the communication sampling measurement data of an LTE network, wherein the communication sampling measurement data comprises the communication characteristic values of a plurality of sampling points; based on the communication characteristic value of each sampling point, utilizing a positioning algorithm and / or field intensity positioning algorithm of a tracking area (TA) and an angle-of-arrival (AoA) to determine the coordinate information of each sampling point; projecting each sampling point to the corresponding grid in a grid area based on the coordinate information; based on the communication characteristic value of the corresponding sampling point in each grid, determining the coverage quality of each grid; based on the coverage quality of each grid, determining the coverage quality of the LTE network. Different from the situation in the prior art that the coordinate of one point, for example a base station or a cell, only can be obtained, so that the coverage quality determination precision is lower, according to the present invention, by combining various positioning algorithms, the coordinate of each sampling point in the base station or the cell can be obtained accurately, thereby obviously improving the coverage quality determination precision.

A refrigeration system comprising a refrigerant circuit including a compressor and a receiver in which the judgment accuracy of a liquid level detecting circuit for judging whether liquid refrigerant is stored up to a specified position of the receiver is enhanced. An air conditioner (1) comprises a main refrigerant circuit (10) and the liquid level detecting circuit (30). The main refrigerant circuit (10) comprises a compressor (21) for compressing gas refrigerant, a heat-source-side heat exchanger (24), the receiver (26) for storing liquid refrigerant, and a utilization-side heat exchanger (52). The liquid level detecting circuit (30) is arranged to take out a part of refrigerant in the receiver (26) from a first fixed position (L1) thereof, to reduce the pressure thereof while heating, and to return the refrigerant back to the suction-side of the compressor (21) after measuring the refrigerant temperature, thus detecting the fact that the liquid level in the receiver has reached the first fixed position (L1).

The invention relates to a fusion method and a fusion system of high dynamic range images. The method comprises the following steps of: performing target detection on each LDR image in a low dynamic range image sequence to obtain an initial detection frame, expanding the initial detection frame according to a preset proportion, extracting a region to be segmented in an alternative detection framefrom the LDR image, and segmenting the region to be segmented to obtain a segmented target; The moving state of the segmented object is judged, and the moving object is merged into a background imageto obtain the segmented image of the moving object. Pixel clustering is performed on the LDR image with the highest exposure quality, a super pixel segmentation image is acquired, and the moving object segmentation image and each LDR image are weighted and fused according to the number of pixels of each super pixel mapped on the moving object in the moving object segmentation image to obtain a high dynamic range image. The invention improves the accuracy of moving object detection and judgment, and makes the moving object in the high dynamic range image clearer.

The invention provides a highway monitoring system and a control method thereof. The system comprises monitoring devices arranged on each road section. Each monitoring device is connected with a control server through wireless communication technology. Each monitoring device is provided with a receiver for obtaining a position signal of each mobile terminal within a preset range, and a first communication module for uploading the signals obtained by the receiver to the control server. The control server is provided with a second communication module matched with the first communication module and used for receiving the signals of each monitoring device. The control server is also provided with an analysis module for obtaining the number of effective mobile terminals at each road section according to the received signals of the second communication module, a judgment module for obtaining a congestion coefficient of the corresponding road section according to the number of the effective mobile terminals, and an output module for calling the congestion coefficient and sending the congestion coefficient to a monitoring center. Through the system and method above, the mobile terminals at each road section of a highway are subjected to statistic detection to accurately judge the congestion coefficient of the corresponding road section.

The invention relates to an alarm current value settable bi-color fault indicator which comprises an alarm current setting switch, a current sensing element, a signal conditioning circuit, a processing unit and an optical signal generating circuit, wherein the current sensing element, the signal conditioning circuit, the processing unit and the optical signal generating circuit are sequentially and electrically connected in the signal transmitting direction. The processing unit comprises an A / D converter for digitalizing voltage signals output by the signal conditioning circuit, a comparator for comparing voltage signals, judging whether the voltage signals exceed a set awakening threshold value and then triggering and awakening a CPU if the voltage signals exceed the set awakening threshold value, and the CPU for controlling work of the A / D converter, calculating a digitalization result of the A / D converter and sending a fault indication control signal if the digitalization result shows that a fault is produced, wherein the control signal triggers the optical signal generating circuit to send an optical signal and give an alarm indication. The alarm current value settable bi-color fault indicator has the LED indication advantages of being high in accuracy, low power consumption, capable of achieving accurate multi-gear setting of alarm current values and various in output mode, and a user can perform flexible setting according to line load current level.

The invention discloses a vehicle lane prediction method and a lane level map. The level of current vehicle running road can be judged by using historical route data and lane level data, particularly under the condition of layering at road space level junctions of flyovers and the like, whether the current vehicle and an object vehicle are on the same road level and a collision risk exists within the three-dimensional spatial range can be judged, so that false alarm is inhibited.

The invention aims to provide a robust general steganalysis method for image content awareness. The method of the invention can better utilize the advantage of the CS technology to process digital images, and further improve the data processing efficiency and the classification judgment precision of the steganalysis. The method comprises the following steps: (1) setting the size N, a block size B and a measured value M of a digital image X; (2) determining a 9 / 7-Haar DWT sparse matrix Psi; (3) determining a Hadamard measurement matrix PhiB of the BCS; (4) dividing a digital image into B*B blocks and calculating a measurement vector yi and the measurement data Y of the entire image X; calculating the measurement vector yi of the i-th image block and the measurement data Y of the entire image X according to the Hadamard measurement matrix PhiB ; and (5) dividing the image into a smooth part and a non-smooth part as a perceived result; (6) a training process; 7) a classification process; 8) a judgment and analysis process; and 9) repeating iterative search until the entire image data is detected.

The invention provides a tight oil fractured horizontal well productivity main control factor evaluation and productivity prediction method. The method comprises the following steps of determining potential influence factors of tight oil fractured horizontal well productivity; establishing a fractured horizontal well productivity comprehensive evaluation system adapting to the research area; acquiring related data; judging productivity master control factors of the fractured horizontal well in the research area; calculating the weight proportion of the main control factor in the category; establishing a productivity level prediction formula of the fractured horizontal well in the research area; and establishing a productivity decline prediction model of the fractured horizontal well in theresearch area, and predicting an oil well productivity decline rule. The tight oil fractured horizontal well productivity predicted by the productivity prediction method provided by the invention hashigh goodness of fit with field production data, the oil production data of one horizontal well is compared with a prediction result, the relative error is less than 7%, the prediction precision is relatively high, and the production requirements can be met.

The invention provides a method and equipment for judging misfire of a diesel engine. The method comprises the steps that instant rotating speeds n of crankshaft teeth corresponding to cylinders are collected in a working cycle; average angular speeds omega of the cylinders are calculated due to the instant rotating speeds n of the crankshaft teeth corresponding to the cylinders; angular acceleration alpha among the cylinders is calculated with the average angular speeds omega of the cylinders according to a formula as described in the specification; omega1 is the average angular speed of the current cylinder; omega2 is the average angular speed of the previous cylinder; delta t is a time sum corresponding to the crankshaft teeth of the two cylinders; and whether the angular acceleration alpha among the cylinders is smaller than a preset misfire threshold value is judged, if so, the current cylinder has the misfire, and otherwise, no cylinder has the misfire. As the angular acceleration among the cylinders is obtained by collecting the instant rotating speeds of the crankshaft teeth in every working cycle, the accuracy of judging the misfire is improved. In addition, as the angular acceleration among the cylinders is obtained according to the difference between average angular speeds of the current cylinder and the previous cylinder, the angular acceleration among the cylinders can resist signal interference of the individual cylinders effectively. The accuracy of judging the cylinders have the misfire is improved.