80results about How to "Solve unrecognized" patented technology

The invention discloses a database masking system based on big data. The system comprises a system management module for constructing a role for the database masking system, and performing function authorization and management on the role, a data source management module for providing data source management for the database masking system, a sensitive data discovery module for automatically discovering sensitive data in the database and submitting the sensitive data to a user to adjust and confirm, a masking task management module for configuring and managing the masking task of the database masking system; a masking configuration management module for managing and defining the sensitive data and performing masking algorithm and masking strategy management, a masking verification module for verifying the masking by comparing data before the masking with data after the masking, a multi-stage management module for realizing the status monitoring and strategy issuing and data synchronization of multi-stage masking equipment, and a security auditing module for analyzing the acquired masking log data in real time to guarantee the security of the database masking system. The invention further discloses a database masking method based on the big data.

The invention provides a virtual touch screen system based on image recognition, which is characterized in that the system is composed of at least two one-dimensional image acquisition devices, a display screen and a computing device, wherein the one-dimensional image acquisition devices are arranged at the periphery of the virtual touch screen; each one-dimensional image acquired by the one-dimensional image acquisition devices is positioned on the same plane as the virtual touch screen; the one-dimensional image acquisition devices convert the positions of the fingers on the virtual touch screen and the clicking motion into a plurality of points dots in the one-dimensional image and dot existence / nonexistence; and the computing device recognizes two-dimensional rectangular coordinates of the fingers in the virtual touch screen according to the dot information in the one-dimensional image, and executes the corresponding operation, thereby performing the function of human-computer interaction. The virtual touch screen system solves the problem that the fingers are screened by the human body in the touch screen utilizing the image technology, and is particularly suitable for large-screen touch systems.

The invention discloses a new word automatic searching system and a new word automatic searching method based on a query log. The new work automatic searching system mainly comprises a query log preprocessing module, a new word discovering module and a new word generating module, wherein the query log preprocessing module is used for regularly acquiring a query string, query frequency and the like from the query log in a timed interval according to set timed new word discovering time; the new word discovering module counts the frequency of the same n-gram strings according to a word segmentation result of the query string, computes the concurrence rate of the n-gram strings, and mergers primary and secondary strings with the similar frequency in a candidate new word assembly; and the new word generating module performs filtering and pruning strategies for the candidate new word assembly, and removes rubbish strings in the candidate new word assembly, so that a final new word assembly is obtained. By the new word automatic searching system and the new word automatic searching method, problems that a corpus is difficult to acquire by an existing statistic method, a rule-based method is difficult to extend and the like are solved, new words can be automatically discovered from the query log simply and easily by the aid of the concurrence rate of the word strings and the auxiliary filtering strategy, and the final new word assembly is introduced into a word segmentation bank, so that incremental new word discovery is realized.

The invention relates to a sequence marking system, which comprises a model marking module, an adjustment module, and a strategy base; an output end of the model marking module and an input end of theadjustment module. The model annotation module is used to annotate the input text data serially. One or more policies are stored in the policy library, and the adjustment module is used for retrieving the policies from the policy library, and adjusting the annotation results output by the model annotation module according to the policies and the input text data. As that system or the method of the invention label the text sequence, the accuracy and the applicability of the original model labeling can be enhance.

The embodiment of the invention discloses a method and device for judging the login abnormality of an account, and the method comprises the steps: obtaining the login correlation data of a current login user; determining the login abnormality score of the current login user based on the login correlation data of the current login user and a login abnormality database, wherein the login abnormality database comprises login abnormality data and the corresponding login abnormality score; judging whether the login of the account of the current login user is abnormal or not according to the determined login abnormality score. According to the scheme of the invention, the method achieves the automatic judgment whether the current login of the account is phishing login or not, effectively avoids the loss caused by the phishing login to a user, and improves the user experience.

The invention discloses a servo control system applied to precise numerical-control machine tool processing equipment. The servo control system comprises a programmable control unit with a CAN (controller area network) open communication interface and a servo driving unit with a CAN open communication interface, and the programmable control unit is in communication with the servo driving unit via a CAN open bus. A positioning module of a programmable controller is omitted, the CAN open bus replaces a pulse signal wire and a control wire, consumption of wires is reduced by two-third, cost is saved, and simultaneously, failure rate of circuits is greatly reduced. Digital signals transmitted by the CAN open bus replace pulse signals of the prior art, problems that speed is limited due to over high pulse frequency and individual pulse signals cannot be identified by servo drivers in the prior art are resolved, the speed can be 1.5 times faster than the original speed, processing efficiency is improved, and precision is increased to 0.3mm from 0.5mm of the original precision.

The invention discloses a device identification method and apparatus, a storage medium and an electronic device. The method can be applied to the electronic device, the electronic device provides first detection voltage for a preset pin of its Type-C interface, the preset pin is a non-configured channel pin, and when it is detected that a Type-C interface of an external device is accessed to the Type-C interface of the electronic device, the configured channel pin of the Type-C interface of the electronic device has no detection voltage. The method comprises the steps that when an instructionfor indicating identification of the external device is received from a user, the preset pin is grounded; when it is detected that the preset pin is grounded, a trigger signal is generated; accordingto the trigger signal, second detection voltage is provided for the configured channel pin and is greater than the first detection voltage; after the second detection voltage is provided for the configured channel pin, the external device is identified through the configured channel pin. The problem can be effectively solved that the electronic device cannot identify the external device by adopting the device identification method.

The invention discloses a system for identifying license plates of black smoke vehicles by front and rear snapshotting. A front-end snapshotting unit is respectively connected to a black smoke vehicle identification unit, a motor vehicle license plate identification unit and a motor vehicle license plate verification unit, the black smoke vehicle identification unit, the motor vehicle license plate identification unit and the motor vehicle license plate verification unit are connected to a data processing unit, and the data processing unit is connected to a data storage unit. The system monitors a plurality of lanes by means of a forward camera and a backward camera, consequently, the problem that license plates cannot be identified when detection is carried out on black smoke vehicles is solved, and the black smoke vehicle license plate identification rate is effectively increased.

The invention relates to an automatic correction system for simple answers, which is characterized by comprising a text analysis and labeling module, a similarity comparison module and a similarity scoring module, and a client comprises a user input answer text, a user labeling word pair and a display result. According to the method, deep semantic analysis can be realized without depending on a knowledge base, and the similarity of answer texts is obtained by using an optimized word comparison method.

The invention relates to the field of speech recognition, in particular to an automobile failure solving method and device, a storage medium and an automobile control terminal. The method comprises the steps that speech produced by a user is received; personal speech features of the user are acquired, and the personal speech features of the user are used for performing word segmentation processingon the speech to obtain recognition words contained in the speech; based on the recognition words, matching is carried out in a personal word library of the user to obtain standard words corresponding to the recognition words; the personal word library is used for storing a corresponding relationship between the recognition words and the standard words; according to the standard words, semanticsof the speech of the user is recognized. The automobile failure solving method and device, the storage medium and the automobile control terminal solve the problem that the nonstandard speech of the user cannot be recognized, and improve the accuracy rate and efficiency of speech recognition.

The invention discloses a head area identification method, device and equipment and belongs to the field of machine learning. The method comprises the following steps: obtaining an input image; inputting the input image into cascaded n neural network layers to obtain n groups of candidate identification results of a head area, wherein the neural network layers are used for carrying out identification on the head areas according to preset extraction frames, and at least two neural network layers adopt different sizes of extraction frames; and carrying out aggregation on the n groups of candidate identification results to obtain a final identification result of the head image in the input image. Since the sizes of the extraction frames adopted by at least two neural network layers in the n neural network layers are different, the problem that the head area cannot be identified based on the fixed-size extraction frame when a human face occupies a small area in the monitoring image is solved, so that different sizes of head areas in the input image can be identified, and identification accuracy is improved.

The invention discloses a fast recognition method for multiple optical labels. The method comprises the steps as follows (1) all visible optical labels within a recognition range of a camera are obtained, all visible optical labels are displayed into one display region of a handheld device, the handheld device recognizes regional center positions of the visible optical labels and divides the display region of the handheld device into multiple display regions by employing the regional center positions of the visible optical labels as base points, wherein one visible optical label corresponds to one display region; (2) the display regions are distinguished, realistic pictures of the positions at which the visible optical labels are located are inserted into the display regions corresponding to the visible optical labels, and meanwhile, the visible optical labels are inserted into the corresponding realistic pictures of the positions at which the visible optical labels are located; and (3) a user clicks any display region and the handheld device recognizes visible optical label content corresponding to the display region. According to the method disclosed by the invention, recognition of the multiple visible optical labels can be achieved and the phenomenon of disordered visible optical labels can be avoided.

The invention provides a capacitance value self-calibration device and method as well as a touch display device. The capacitance value self-calibration device comprises a capacitance value detection unit and a capacitance value calibration unit, wherein the capacitance value detection unit is used for detecting an actual capacitance value between an X-axis touch electrode and a Y-axis touch electrode on the surface of a touch display panel, and the X-axis touch electrode and the Y-axis touch electrode are opposite to each other; and the capacitance value calibration unit is connected with the capacitance value detection unit and is used for setting the actual capacitance value into a basic capacitance value. According to the capacitance value self-calibration device and method as well as the touch display device, the problem of false alarm points caused by change of the basic capacitance value or the problem that the touch display panel cannot be recognized when being touched in the prior art can be solved.

A household member identification system is provided. A member identity corresponding to member information collection is obtained by face recognition, The method can realize family member recognitionby using pedestrian detection and pedestrian re-recognition technology, and can realize multi-person recognition in a video frame. The algorithm can extract the features of all the people to be identified in the picture only once, and the searching speed is fast. In addition, the information database of family members can be updated in time so as not to be unrecognizable due to the change of clothes, and the actual demand of smart home can be met; At that same time, the method only require a plurality of ordinary USB cameras, and the device is simple, low in cost and strong in practicability.

The invention relates to an identification method and device between equipment, computer equipment and a storage medium. The method comprises the following steps: under the condition that second electronic equipment sends pairing broadcast information containing second electronic equipment identification information and scans an ultra wide band label in a preset range, first electronic equipment scans the second electronic equipment identification information in the pairing broadcast information, and establishes communication connection with the second electronic equipment; a control command is sent to the second electronic equipment through the established communication connection, wherein the control command is used for requesting to acquire scanned ultra-wideband tag information; candidate ultra-wideband tag information returned by the second electronic equipment is received; and target ultra wide band label information is identified from the candidate ultra wide band label information. By adopting the method, rapid and effective identification between equipment can be realized.

The invention discloses a connection body and template-based character identification method. By obtaining a character image, converting a grayscale image, performing image binarization, realizing character segmentation by a connection body algorithm, loading character templates, performing template matching and outputting an identification result, number and image identification is realized. According to the connection body and template-based character identification method, while conventional character identification can be supported, the problem that non-conventional, namely, super-large orsuper-small or irregular font characters cannot be identified is solved; self-defined templates become simple and easy to operate; the character templates are added directly on an original template;and combination of multiple templates can be conveniently realized.

The invention provides a flow analysis method and device and a computer system. The method comprises the steps of obtaining plaintext features and ciphertext features of a message in traffic, whereinthe ciphertext features comprise length features of an encrypted field in the message; and performing analysis on the flow according to the plaintext features and the ciphertext features to identify aservice or an application to which the flow belongs. The method can be used for service identification and can also be applied to application identification. By introducing ciphertext characteristicsinto traffic analysis, the accuracy of traffic identification is improved in a message encryption scene. On the other hand, the invention further provides a public service flow attribution method anddevice and a computer system, the flow of the public service is attributed to the corresponding application by identifying the initial service and / or the exclusive service of the application, the accuracy of application flow attribution is improved, and a foundation is provided for subsequent application flow charging and other services.

The invention discloses a power quality disturbance identification method based on an improved HHT algorithm, which comprises the steps of firstly, collecting signals, and segmenting the signals by using an SAX algorithm; secondly, performing HHT analysis on the different segments of segmented signals, and extracting feature values of the different segments of signals; then identifying the power quality disturbance condition by using a decision-making tree according to the extracted feature values; and finally performing iterative HHT analysis on identified normal power signals so as to further judge whether there is minor disturbance or not, and thus achieving complete identification for the power quality disturbance. The power quality disturbance identification method disclosed by the invention based on the improved HHT algorithm solves a problem that it is impossible to identity non-stationary signals the prior art. The signals are segmented by adopting the SAX algorithm, and then different segments of signals are analyzed to further judge whether there is disturbance or not, thereby well solving a problem of mode aliasing occurred with high frequency and small amplitude.

The present invention discloses a method for identifying an out-of-zone fault in a power transmission line protection zone of an electric power system. The method comprises the following steps: in step 1, an initial traveling wave voltage and a pre-fault voltage at each point of a protection line are obtained; in step 2, correlation between the initial traveling wave voltage and the pre-fault voltage at each point of the protection line is calculated, absolute values of correlation coefficients are obtained and a curve of the absolute values of the correlation coefficients is formed; in step 3, according to the curve of the absolute values of the correlation coefficients, whether the fault occurs within a scope of protection or outside the scope of protection is determined; if the curve of the absolute values of the correlation coefficients has an extreme point, that the fault occurs within the scope of protection is determined; otherwise, the fault occurs outside the scope of protection. Via the method disclosed in the technical solution of an embodiment of the present invention, an in-the-zone fault and the out-of-zone fault in traveling wave distance protection can be identified, a problem that the in-the-zone fault and the out-of-zone fault cannot be identified in a single-ended traveling wave protection can be solved, great pioneering effects are exerted, and reliability of traveling wave distance protection can be ensured.

The invention provides a method and system for extracting key information from an identity card image picture, and the method comprises the steps: obtaining an identity card picture through recognition, detecting the identity card picture through a face detection model and a national emblem detection model to obtain front and back screenshots of an identity card, carrying out the textbox recognition of the screenshots with shading removed, obtaining a textbox screenshot, and performing character recognition on the textbox screenshot, and calibrating a recognition result to obtain a final recognition result. By adopting a multi-model series processing structure, the problem of identity card picture information extraction under various picture quality conditions is solved; the problem that skewed and inverted images cannot be recognized is solved through rotation recognition. The identity card number training data annotation problem is solved by utilizing the verification rule of the identity card number and combining with a plurality of machine learning models for common identification. By using the image morphology to cooperate with the CTPN model, the problem of CTPN model training set data annotation in the textbox recognition process is solved.

The invention provides an identity card quality inspection method and system based on deep learning. The method comprises the steps: acquiring a first image containing an identity card image and a background image, preprocessing the first image to obtain a second image only containing the identity card image, inputting the second image into a set reflection quality inspection model for reflectionquality inspection to obtain a third image, and inputting the third image into a set fuzzy quality inspection model for fuzzy quality inspection to obtain a fourth image, wherein the set fuzzy qualityinspection model and the set reflective quality inspection model are both obtained according to the basic structure of SqueezeNet. According to the invention, the problem that a plurality of index thresholds need to be adjusted during identity card quality inspection in the traditional image processing technology and equalization is difficult to achieve on positive and negative sample recall indexes is solved, and a traditional image processing method for separating a reflective area from a non-reflective area by using brightness, gradient and color cast can not distinguish reflective areas (characters and head portraits) of an identity card from reflective areas of the non-critical area and can not identify Great Wall watermarks similar to reflective areas.

The invention discloses a release agent, a release film, a composite film and an ink-jet printing medium, preparation methods of the release agent, the release film, the composite film and the ink-jetprinting medium, and an application of the ink-jet printing medium. The release agent comprises the following components in parts by weight: 5-7 parts of polyvinyl alcohol, 1-2 parts of silicon dioxide, 4-8 parts of titanium dioxide, 5-10 parts of waterborne polyurethane, 5-10 parts of emulsion wax, 0.5-1.5 parts of a wetting agent, and water, wherein the solid content of the release agent is 15-30%. The preparation method of the release agent comprises the following steps: preparing 8-12% of polyvinyl alcohol water solution, 10-15% of a silicon dioxide dispersion liquid and 20-40% of a titanium dioxide dispersion liquid, and then performing uniform mixing with other components to obtain the release agent. The release film is obtained by coating the release agent on a base material, and performing drying, cooling and stripping, wherein the usage amount of the release agent is 100-130 g / m<2>. The composite film comprises a polymer base film and a release film. The ink-jet printing medium comprises a composite film and a transparent ink-absorbing coating. The release film and the polymer base film have moderate adhesive force, the release effect is good, and the ink-jet printing medium can be identified during printing.

The present invention discloses an FBG reflectance spectrum abnormal mode rapid discrimination method. The method comprises the following steps of 1) obtaining an FBG reflectance spectrum via a fibergrating demodulator, and converting the FBG reflectance spectrum into a digital sampling signal to send to an upper PC; 2) setting an alarm parameter on the PC; 3) rapidly discriminating the normal mode and the abnormal mode of the FBG reflectance spectrum within the FBG wavelength range [lambda min, lambda max] and according to the set alarm parameter. The method solves the problems that under the atmospheric ambient temperature environment and the high vacuum thermal environment, the FBG reflectance spectrum abnormal mode is identified incompletely, and a fault mode is low in identificationaccuracy, has the characteristics of being simple and rapid in algorithm, little in mode discrimination participation data quantity and high in abnormal mode identification accuracy, and being not influenced by the low spectrum signal to noise ratio, and enables the fault diagnosis efficiency of an FBG under the regular usage environment and the extreme environment to be improved substantially.

The invention relates to the technical field of customer service systems, and particularly discloses a multi-round dialogue intelligent customer service system based on proprietary word correction andcold start. The system comprises a scene and intention recognition module, an entity grooving recognition module, a multi-round session management module and a model training management module, wherein the scene and intention recognition module, the entity grooving recognition module and the multi-round session management module are sequentially connected through a protocol, and the model training management module is separately connected with the scene and intention recognition module and the entity grooving recognition module through protocols. A proprietary word correction function is applied during the entity grooving process, so that the problem that the user does not definitely input the proprietary nouns set by the system one by one, so that the system cannot identify, is solved. The whole system not only can perform intention switching, but also can perform intention inheritance, and after the BERT is introduced, the manual annotation containing universal nouns can be reduced.

The method includes: subscriber equipment receives the system message from radio network controller; the system message includes the convergence deviation message of frequency layer; when session begins, the radio network controller sends the multi media broadcast and multicast message including the convergence message of the said layer to the subscriber equipment. The convergence message of the said layer includes the frequency information of firstly selected layer in current location and the ID of cell in the firstly selected layer cell; the subscriber equipment decides if itself is at the firstly selected layer in term of the frequency information in the layer convergence message in the received broadcast and multicast message; if not, executes the reselection of cell in term of the ID of cell in the convergence message of the said layer and the convergence deviation information of the frequency layer.

The invention belongs to the technical field of picture recognition, and particularly relates to a table recognition method adaptive to multiple types of OCR recognition interfaces and related equipment. The method comprises the following steps: receiving an extraction request, wherein the extraction request comprises a document and an identification mode; according to the recognition mode, calling a preset external OCR interface, performing recognition processing on the document through the external OCR interface, and receiving recognition data returned by the external OCR interface; and generating table data from the identification data, and returning the table data. According to the invention, through the adaptation of a plurality of identification modes, most of the encountered OCR tables in the disclosed bond announcement can be identified basically, the coverage range is wider, and the identification rate is higher.

The invention discloses a tabletop microphone supporting a PTT hardware detection function. The tabletop microphone comprises a PTT press key, a single key keyboard module, a movable ring microphone,an audio frequency wire and a USB data wire, wherein the PTT press key uses a button switch; the button switch has two contacts; the contacts are used for detecting states of the PTT press key; when the press key is pushed down by external force; the two contacts are in a conduction state; the single key keyboard module is used for generating press key information capable of being recognized by the computer; the movable ring microphone is used for being in charge of sound collection and converting voice input signals into electric signals; the audio wire is used for transmitting sound signalsconverted by the movable ring microphone to a sound card of the computer. The PTT hardware []detection mode with high reliability in wireless communication is applied to the tabletop microphone; the USB single key keyboard module is arranged inside the tabletop microphone; a user can realize high reliability during the real-time talkback of voice conference; the PTT function triggering by a mistake due to sound in ambient environment is avoided.

The invention provides an output-voltage-adjustable electric vehicle charging pile and a charging voltage adjusting method. The charging voltage can be adjusted according to the voltage specificationof an electric vehicle battery pack. The electric vehicle charging pile comprises a machine case and a charging circuit board arranged in the machine case. The charging circuit board comprises a charging circuit, a main control MCU, a voltage signal collecting circuit, a current signal collecting circuit and a PWM signal feedback circuit. The charging circuit comprises a transformer. The chargingvoltage adjusting method includes the steps that the voltage signal collecting circuit collects the voltage of electric vehicle batteries and sends the voltage to the main control MCU; the main control MCU judges the range of the collected voltage according to the collected voltage signals of the battery pack; and the main control MCU controls the voltage of the transformer T2 to be adjusted to the matched charging voltage of the battery pack through the PWM signal feedback circuit according to the obtained voltage specification of the battery pack. The matched charging voltage is adjusted according to the battery voltage of the electric vehicle, and the voltage specification of the battery pack can be automatically recognized within a cross range.

The invention provides an asset and fingerprint identification method based on a feedback mechanism and related equipment, and the method comprises the steps: carrying out the analysis and processing of preset data information through a detection engine, obtaining at least one suspected protocol detection data packet, and preliminarily reducing the detection range; according to a feedback detection method, a detection engine selects version detection data packets from at least one suspected protocol detection data packet in sequence, and sends the version detection data packets to a target application, and the target application analyzes the protocol type and the protocol version number supported by the target application and the protocol type and the protocol version number in the version detection data packet to obtain an analysis result, sends the analysis result as feedback information to the detection engine, and stops matching until the detection engine confirms that the first successfully matched target version detection data packet appears. The problem that the non-standard port application protocol type and the protocol version number cannot be identified is solved, the identification accuracy of the protocol type and the protocol version number is improved, and then the asset type and the fingerprint information are accurately identified.