169results about How to "High-precision identification" patented technology

A form identification method which can identify forms without errors even if a form is inputted from a scanner in different direction or the form is enlarged / shrunk. A form registration method which enables change / modification of forms when form types are pre-registered. Inputted points as features are extracted from the form, modification of enlargement / shrinkage, rotation, or skew is made to the inputted points. Form types are identified based on minimum distances between the inputted point and pre-registered dictionary points. When registering a similar form type, warning can be displayed on the screen. In a system a dictionary registration station, a form identification station, and a form edit station are connected via a communication network. Each station operates in interlocking manner, identifies inputted forms, and registers them into a dictionary.

According to the standard solution and the determination method of the present invention, in a case where a voltage is applied by a drive voltage of a measurement apparatus to an electrode portion of a biosensor comprising an electrode portion including a counter electrode and a measuring electrode formed on an insulating substrate, and a reagent layer which reacts with a sample solution supplied to the electrode portion, and a current value which flows at the application is measured, thereby determining a substrate contained in the sample solution, a reducing substance is contained in the standard solution used for controlling a precision of measurement of the measurement apparatus. Therefore, when the standard solution is measured, a large change occurs in a current waveform between time t0 and t1 shown in FIG. 6 due to the reducing substance, thereby discriminating whether the analyte liquid being measured is the standard solution or the sample solution and easily identifying the kind of analyte liquid.

A face detecting unit detects a person's face from input image data, and a parameter setting unit sets parameters for generating a gradient histogram indicating the gradient direction and gradient magnitude of a pixel value based on the detected face. Further, a generating unit sets a region (a cell) from which to generate a gradient histogram in the region of the detected face, and generates a gradient histogram for each such region to generate feature vectors. An expression identifying unit identifies an expression exhibited by the detected face based on the feature vectors. Thereby, the facial expression of a person included in an image is identified with high precision.

The invention discloses a driver road rage real-time identification and warning method based on facial features. The method comprises the following steps: first, using Kinect to track the face of a driver from real-time video data, and extracting a RGB image and a depth image of the driver face; and then, carrying out graying, image size normalization, median filtering for noise removing and other preprocessing operations on the two types of image information, inputting the image information into a trained CNN model to extract features, inputting the extracted features in a Softmax classifier for classification, fusing the classification results of 30 continuous frames of the RGB image and the depth image of the face, calculating the probability values of two channel confidences, comparing the two confidences, taking the result of highest confidence as a driver facial expression final identification result, and warning the driver according to the identification result. According to the invention, RGB-D driver face feature information is fused, and angry feelings of drivers can still be identified in real time and with high accuracy and warning can be carried out in a night or bad driving environment.

The fingerprint identification system includes a fingerprint identification terminal having a function of receiving input of a plurality of fingerprint images taken from one target finger to select a fingerprint image suited for identification processing based on image quality determination and a fingerprint identification device having a function of outputting an identification result by selectively combining, for each image, one-to-N matching, one-to-one matching, fusion score calculation from a plurality of scores and condition determination of an identification score and a fusion score.

A first switch is arranged between a judgment unit and a signal line. The first switch is turned on during a period in which the judgment unit judges a device, after which the first switch is turned off. A second switch is turned off during a period in which the judgment unit judges the device, after which the second switch is turned on. The signal line connects a communication interface port, which allows various different kinds of devices to be connected, and a communication terminal of a processor configured to perform data communication with the device. The judgment unit monitors the electrical state of the signal line, and judges the device connected to the port.

The embodiment of the invention discloses an identity recognition method and device and related equipment, and the method comprises the steps: obtaining target behavior state information when receiving input identity verification information based on a target user, wherein the target behavior state information is behavior state information generated according to an operation behavior of inputtingthe identity verification information by a target user; obtaining an identity authentication model corresponding to the registered user, wherein the identity authentication model is obtained by training according to behavior state information of an input operation behavior of a registered user; identifying an identity matching relationship between the target user and the registered user in the identity authentication model according to the target behavior state information; and identifying the security type of the target user according to the identity matching relationship and the input identity verification information. According to the invention, the reliability of system security authentication can be improved.

An electronic marker locator with a digital architecture for providing accurate and consistent estimation of the signal strength is presented. The marker locator includes a Digital Phase-Locked Loop (DPLL) structure. The electronic marker locator transmits known and adjustable frequency bursts corresponding to the markers to be located while synchronously capturing the signals returned from the markers. Because of the convergence properties of the DPLL, very consistent measurements of the reflected marker signal field strength are possible, resulting in both an improvement of maximum detection depth and depth accuracy. Further, the analog front-end hardware can be reduced, offering wider resistance to component tolerances, lower calibration and test times, and flexible frequency selectivity.

The invention discloses a method and a system for recognizing a fraud short message by using deep learning, used for solving the problem that an algorithm is not high in recognition accuracy due to the fact that the features of the fraud short message is ever-changing. The method comprises the steps of acquiring text data of a short message sample and performing word segmentation treatment; usingWord2Vec to convert the text data that is subjected to the word segmentation into word vectors; using an LSTM algorithm to convert the word vectors into sentence vectors; using the sentence vectors asinput vectors of a softmax classifier to train a deep learning model; and recognizing the fraud short message according to an output result of the trained and deeply learned softmax classifier. According to the method and system for recognizing the fraud short message by using deep learning provided by the invention, the capacity of accurately recognizing the fraud short message is improved.

The invention discloses a method and a device for identifying characters in a card and certificate and computing equipment. The identification method comprises the steps of obtaining a card and certificate image, wherein the card and certificate image comprises at least one type of character sequence; Obtaining various types of character sequence images from the card image by using a first targetdetection model to obtain at least one first character sequence image; Performing block binarization processing on each first character sequence image to obtain a corresponding second character sequence image; Obtaining a single character image from the second character sequence image by using a second target detection model; And for each single-word image obtained from the second character sequence image, identifying character information in the single-word image.

The invention relates to an abandoned land remote sensing recognition method based on texture and vegetation indexes. The characteristics of multiple sensors, multiple resolutions, multiple spectra and multidate remote sensing data are utilized, the spectra and the textural features of remote sensing images, and the vegetation index features are combined, and the abandoned land remote sensing recognition method based on the texture and the vegetation indexes is established by structuring a remote sensing image registration and fusion framework, a spectrum and texture feature space and a vegetation index time sequence. Compared with an existing abandoned land remote sensing recognition method, the abandoned land remote sensing recognition method based on the texture and the vegetation indexes solves the precision problem brought by broken cultivated land, reduces artificial involvement subjectivity and improves abandoned land recognition efficiency. The abandoned land remote sensing recognition method based on the texture and the vegetation indexes is characterized in that a remote image registration and fusion unit, a typical spectrum and texture feature unit, a vegetation index time sequence feature unit and an abandoned land recognition and information extraction unit are included. Through the abandoned land remote sensing recognition method with the four units, the distribution information and data of abandoned land in a large range area can be recognized fast and precisely, and auxiliary decision-making data support can be provided for land utilization, grain guarantee and the like.

A form identification method which can identify forms without errors even if a form is inputted from a scanner in different direction or the form is enlarged / shrunk. A form registration method which enables change / modification of forms when form types are pre-registered. Inputted points as features are extracted from the form, modification of enlargement / shrinkage, rotation, or skew is made to the inputted points. Form types are identified based on minimum distances between the inputted point and pre-registered dictionary points. When registering a similar form type, warning can be displayed on the screen. In a system a dictionary registration station, a form identification station, and a form edit station are connected via a communication network. Each station operates in interlocking manner, identifies inputted forms, and registers them into a dictionary.

The invention discloses a method for identifying electromagnetic ground impulse response. The method comprises the following steps: designing an electromagnetic ground impulse response identification system according to a Wiener-Hopf equation; coding and transmitting an artificial source electromagnetic method transmitting waveform by using a pseudorandom sequence; synchronously acquiring the coded transmitting waveform and a response observed signal during observation; eliminating the complex influence of the autocorrelation side-lobe of a transmitted signal in received and transmitted signals by adopting a method based on a cross-correlation identification principle to realize high-accuracy identification of ground impulse response. Compared with the conventional step source excitation way, the method disclosed by the invention has the advantage that the influence of the autocorrelation side-lobe of a transmitted waveform is eliminated through receiving and transmission cross-correlation, so that the identification accuracy of the electromagnetic ground impulse response is remarkably increased.

A system of the present invention allows to determine the likelihood that a certain block of a web page is an advertising, such as a banner or a native advertising. The system uses combined and holistic approach to the analysis of units for the purpose of finding the advertising features (links, images and HTML code). The system does not impose any restrictions on possible ways of using it--it only provides tools for determining the probabilities of the unit belonging to the advertising or non-advertising class.

A headlight control device automatically switches a light distribution pattern of a headlight for illumination ahead of a vehicle. The headlight control device including an imager that sequentially takes images of a view ahead of the vehicle in a predetermined frame rate, an another-vehicle recognizer that recognizes a light source of another vehicle from each of the images sequentially taken by the imager, a light distribution controller that switches the light distribution pattern according to a recognition result of the another-vehicle recognizer, and a signal provider that provides a signal to illumination light of the headlight by causing the luminance of the headlight to change in a predetermined cycle. The another-vehicle recognizer recognizes, as reflected light of the headlight of the vehicle, a light source which is present within the image and of which luminance change is substantially similar to a luminance change of the signal.

Provided is a bidirectional infrared household appliance control system. The system comprises a voice recognition module, an infrared emission module, a feedback receiving module and a local processing module; the voice recognition module comprises a voice receiving device and a voice text conversion system and is used for recognizing a human voice in ambient sounds, converting the human voice into text information and transmitting the text information to the local processing module; the infrared emission module receives a command request sent by the local processing module, converts the command request into infrared signals and emits out the infrared signals; the feedback receiving module receives the command request sent by the local processing module, is started after the infrared emission module emits out the infrared signals, monitors signals of a voice receiver, a video receiver and an infrared receiver and feeds monitoring results back to the local processing module; the local processing module receives the signals transmitted by the voice recognition module, simply processes the signals, transmits information which cannot be processed to a cloud processing module, receives a command transmitted by the cloud processing module and sends executable commands to the infrared emission module and the feedback receiving module.

The invention provides an intelligent perception and identification method and device for a warehouse, and electronic equipment. The method comprises the following steps that: obtaining a warehouse laser point cloud panorama obtained when a laser radar scans the warehouse, wherein the warehouse laser point cloud panorama is a point cloud stitched image under a pre-constructed warehouse uniform coordinate system; in the warehouse uniform coordinate system, independently carrying out accurate mapping on the spatial region of the warehouse with area attributes and a business perception and identification rule; carrying out point cloud analysis and automatic resolving on the above warehouse laser point cloud panorama to obtain a target and an event to be monitored; according to the corresponding coordinates of the target and the event in the warehouse uniform coordinate system, calling a corresponding pick-up device to carry out video collection and image analysis on the target and the event, and judging the event type of the event; and reporting the coordinate, the event type and the video image of the above target and event. By use of the method, on the basis of the intelligent linkage of the laser radar and video monitoring, the intelligent target and event monitoring and identification of the spatial region of the warehouse can be realized.

The invention discloses an SAR vehicle target recognition method based on an improved convolutional neural network. The method mainly solves the problem that according to the prior art, accuracy ratesof recognition on SAR vehicle targets are low, and networks are prone to generate over-fitting. The solution of the method includes: removing background clutter of each image in training samples, andcropping each SAR image; constructing an improved convolutional neural network structure based on a caffe architecture, namely setting a classifier of a target recognition part of the convolutional neural network to mixed maximum boundary softmax; inputting the cropped training samples into the improved convolutional neural network to carry out training to obtain a trained network model; carryingout background clutter removal and cropping operations on test samples; and inputting the processed test samples into the trained improved convolutional neural network model to carry out a test to obtain a recognition rate thereof. According to the method, an accuracy rate of SAR vehicle target recognition is increased, network convergence speed is increased, and generalization performance of thenetwork is improved.

There is provided a method of identifying a plurality of moving bodies that are present nearby, wherein each moving bodies includes means for transmitting, as advance information, a movement state including an operation instruction received by the moving body, together with identification information of the moving body and means for monitoring the behavior of other moving bodies, the method including: a step of causing at least one of the plurality of moving bodies to carry out, as a target moving body, a reveal operation that enables the behavior of a target moving body to be differentiated from other moving bodies; and a step of detecting the reveal operation by the means for monitoring of the other moving bodies and identifying the plurality of moving bodies by grasping positions relative to the target moving body. By using this method of identifying, it is possible to precisely identify vehicles moving in the periphery and to control vehicles safely while predicting the behavior of other vehicles.

The invention discloses a temperature modeling method for a proton exchange membrane fuel cell (PEMFC) system based on a variation particle swarm and differential evolution hybrid algorithm. The PEMFC pile output performance is influenced by the temperature controlled by each module of an operation system, the method is combined with a PEMFC pile model based on the variation particle swarm and differential evolution hybrid algorithm, and the steps of resolving the optimal temperature operating parameter algorithm for each module of the pile when the electrochemical performance is optimal are proposed. The hybrid algorithm has good global and local searching and optimizing capacities, and the temperature control parameter for each module of the PEMFC system can be identified at high precision and controlled in real time, so that the electrochemical performance of the fuel cell system is improved.

This disclosure provides a method of detecting and deterring a target animal from a target area. A target area is positioned within the field of vision of a video camera connected to a computer processing system. An animal identification computer program using convolution neural networks and deep learning computer programs and camera images rapidly detects a target animal. The animal identification computer program is trained to identify target animals accurately using a learning algorithm and related machine learning technology. The time to deploy a deterrent against a target animal from the instant of detection is 2 seconds or less so that little or no time is available to the target animal to damage the target area.

The invention discloses a method for identifying and disposing a junk call content. The method specifically comprises the following steps: S1, a conversation content is collected; S2, the conversationcontent is converted into text information; S3, word segmentation processing is performed on the text information by means of a word segmentation tool; S4, a telephone discriminating model is obtained through the words processed through word segmentation according to an LSTM algorithm and a DNN algorithm; S5, according to a softmax classifier, a phone category obtained through the telephone discriminating model is output; S6, a call, which is output to be a junk call by the softmax classifier, is interrupted. According to the method for identifying and disposing the junk call, content in thecall can be analyzed in real time and can be blocked at the same time, so that real-time interception of the junk call is realized.