98results about How to "Easy to identify automatically" patented technology

Disclosed is an information extraction system and method. The method comprises receiving a document and annotation data, the annotation data comprising instances of entities which have been identified in the document, the annotation entity data comprising identifiers of instances of one or more entities which have been identified in the document and data specifying the location of the identified instances of entities within the document, wherein the identifiers of instances of entities comprise references to ontology data; displaying the document to a user, with annotations dependent on the annotation data, highlighting one or more of the instances of entities whose location is specified in the annotation entity data at the location within the document specified by the annotation entity data; preparing revised annotation data from a user and outputting output data derived from the amended annotation data. The output data is typically used to populate a database.

Embodiments of the present invention are directed to a method and system for archiving and compacting data in a data storage array. In one embodiment, a plurality of archivable files are identified, archived, and removed from a data storage array. A first configuration of the data storage array is then determined in which the free space on each of a plurality of data storage devices in the data storage array is calculated. A process is performed for moving a plurality of files from a first of the data storage devices to other data storage devices in the data storage array. The moving process is repeated to create a second configuration of the data storage array. In the second configuration, the first data storage device has a greater amount of free space than in the first configuration.

The invention involves a calibration device for camera calibration, which consists of two vertical grid boards that have distribution grid and the number of signs marked with a unique code point. The invention also provides a method of the camera calibration, including the following steps: (a) Acquisition of images of the calibration device; (2) corresponding coordinates of the signs point to the coordinates of object point; (3) projection the plane grid to get the initial point coordinates of grid point; (4) the use of line extraction intersection to get the precise coordinates of points; (5) The use of 3-D direct linear transformation for the initial value of calibration parameters; (6) iterative solution with beam adjustment to get the camera's internal and external orientation elements. After shooting a picture of calibration device in the invention, the follow-up calibration process can be done automatically by computer. The advantage of the invention is automatic and quickness.

In one aspect of the invention, a method of providing automated assistance in configuring customer premises equipment for communication with another network element includes automatically identifying a virtual channel and / or a protocol valid for configuration with the customer premises equipment, and assisting a user in configuring the customer premises equipment for use with the identified virtual channel and / or protocol.

The invention discloses an intelligent on-line monitoring system aiming at a high-voltage circuit breaker. The intelligent on-line monitoring system comprises an on-line monitoring device consisting of an on-line monitoring module group, a power module and an upper computer interface module, and an upper computer device connected with the on-line monitoring device, wherein the on-line monitoring module group comprises a circuit breaker separating brake loop disconnection monitoring module, a circuit breaker arc extinguish chamber vacuum degree monitoring module, a circuit breaker contact temperature rise monitoring module, a circuit breaker over travel monitoring module and a circuit breaker separating / connecting brake speed monitoring module. The five sub monitoring modules are connected with the upper computer interface module to realize data interaction between the five sub monitoring modules and the upper computer device, and the upper computer device performs monitoring data search and parameter configuration on the five sub monitoring modules by the upper computer interface module. According to the system, the measurement of each piece of state information is mutually separated from a power source power supply module, the internal mechanism of the circuit breaker is analyzed again, and the current running state parameter and state of the high-voltage circuit breaker are obtained accurately and efficiently.

The invention provides a network public opinion emergency automatic identification method based on emotion information extraction and analysis. The method comprises the steps of obtaining massive social media data according to a social media collection method, and performing data preprocessing such as data filtering and cleaning; training an emotion model according to the historical social media data to form an emotion recognition model; detecting and analyzing sudden emotions in the social media according to the emotion recognition model; extracting social media tags corresponding to the sudden emotions, and recognizing event original keywords of the tags corresponding to the sudden emotions through word segmentation processing to form an event original keyword library; and extracting social media event keywords, matching the social media event keywords with the keyword library, and when the keywords are completely matched and the emotion state is not negative, obtaining correspondingevent basic information to form emergency description. The invention is suitable for identifying various social media emergencies, high in expandability, good in reusability and low in development cost, and can be applied in multiple dimensions according to service requirements.

The invention relates to a vanishing point based method for automatically extracting and classifying ground movement measurement image line segments. The method comprises the following steps of: firstly, analyzing the attribute of ground object space parallel characteristic lines of a target and calculating vanishing points of an image of the target on the basis of image sequences acquired by ground movement measurement; secondly, extracting characteristic line segments based on an image edge according to the Hough transform of a polar coordinate parameter area extremum and the continuous change of image greyscale; and then, classifying the extracted plane line segments by using vanishing points; and finally integrating the classified line segments according to a total least square method and conditions of the plane adjacency and overlapping degree of the line segments so as to extract image characteristic line segments which correspond to multiple ground object space parallel lines. The invention can improve the indoor processing capability of a movement measurement system and promote the research on automatic recognition, mapping, three-dimensional reconstruction and the like of multiple ground objects (such as buildings, roads and the like).

The embodiment of the invention discloses a method for cleaning a cache directory / document based on a sandbox technology and a device thereof, and the method comprises the following steps of: during program run, triggering program functions by a simulated user; monitoring the cache directory / document generated in the program run by the sandbox technology; storing information of the monitored cache directory / document into a cache information database by corresponding the program name. In the technical solution of the invention, the cache catalogues or the cache documents generated in run of various programs are automatically collected from a server by technicians via the sandbox technology, compared with the cache catalogues or the cache documents artificially collected, the collection efficiency is greatly improved and the workload of the technicians is reduced, thus the more cache catalogues / documents can be automatically indentified.

The invention discloses a cough monitoring device and system capable of being carried about. The device mainly comprises a main engine, a sound sensor, a surface myoelectricity sensor, an accelerationsensor, a filter and a storage. Cough sounds, cough myoelectricity and acceleration signals are recorded at the same time, the cough times, frequency and strength can be monitored, monitoring is morecomprehensive, and the result is accurate and reliable; a computer readable medium is good in interactivity, individuality and specificity setting can be performed according to different cough patients and sound characteristics, the specificity is high, automatic cough recognition and marking and analysis of the cough frequency, intensity and times are facilitated, a cough report is automaticallygenerated, manpower resources are saved, automatic recognition of long-time cough sounds of a chronic cough patient can be achieved, the cough objective evaluation index is obtained, and the device is used for guiding the disease assessment and treatment effect observation of the clinic cough patient, and is of the important clinic application value.

The present invention discloses an intelligent drinks cabinet and its management method. Said intelligent drinks cabinet includes cabinet body and its internal shelves. On the cabinet body an intelligent management system is set, and said system includes micro-type computer and dot matrix touch screen, infrared scanner, thermal printer, USB interface, network interface, position sensor and blinking warning light. Besides, said invention also provides the concrete steps of its management method.

The invention aims at providing an image recognition device. According to the image recognition device, welding spots for welding silver wire on the surface of a silver foil in an acquired image are recognized. The acquired image comprises at least one first region corresponding to the silver foil and at least one second region corresponding to the silver wire. The image recognition device comprises a gray value?acquisition unit, a feature extraction unit and a welding spot determination unit, wherein the gray value?acquisition unit is used for acquiring the gray?value of each pixel in the first region and the second region; the feature extraction unit is used for acquiring the feature points of the first region according to the gray?value of each pixel in the first region; and the welding spot determination unit is used for seeking the welding spots in the second region according to the gray?value of each pixel in the second region. In comparison with the prior art, recognition is carried out according to the gray difference between the silver foil and the silver wire, the position of the silver wire is sought around mass points of the silver foil, the welding spots are finally determined, and the positions of the welding spots are then conveniently, quickly and automatically recognized.

The invention provides a recognition and cleaning method and a recognition and cleaning system of cache directories. The recognition and cleaning method of the cache directories includes that mobile terminals confirm directories matched with the cache directories recorded in a local database during scanning processes of the mobile terminals for storage space; the mobile terminals recognize brother directories of the confirmed directories, and upload directory information of directories which are not recognized into a server; the server recognizes the cache directories from the directory information uploaded by all the mobile terminals. By adopting the technical scheme, the recognition and cleaning method and the recognition and cleaning system of the cache directories replace manual work with a huge quantity of the mobile terminals to collect the directories which are not recognized into the cloud server, recognize the cache directories through the server, greatly improve collection efficiency of the cache directories, overall improve recognition efficiency of the cache directories, reduce workloads of technical staff, conveniently and automatically recognize a large number of the cache directories, and thereby clean the cache directories according to the cache directories which are automatically and efficiently recognized, and can improve use efficiency of the storage space.

The mold and processing method used in the processing of free-combination and suction-type building components. Nowadays, there are many kinds of molds for the processing of concrete prefabricated components, and the disassembly and assembly remain unchanged. Every time the mold is installed, holes must be drilled on the bottom formwork to connect the fixed mold frame, which makes the processing speed of concrete prefabricated components slow, and the labor intensity of workers increases. The low efficiency is not suitable for industrial production and processing of concrete prefabricated components, and the quality requirements of concrete prefabricated components cannot be guaranteed, so that the processing cost of concrete prefabricated components remains high. The composition of the present invention includes: a unit mold (1), a magnetic frame (2), the unit mold attracts the side of the magnetic frame to form a magnetic mold unit (3), and a group of magnetic mold units attract each other to form a strong magnetic force The one-side moulds (4), the strong magnetic one-side moulds are attracted to each other to form an upper mould (5), and the upper moulds are attracted to a bottom formwork (6) to form a building component processing mould. The invention is used for the production of concrete prefabricated elements.

A wound coil comprises terminal electrodes. Each of the terminal electrodes includes a bottom portion extending over a bottom surface of a flange from a ridge at which the bottom surface and an outer end surface of the flange cross each other, and an outer portion extending over the outer end surface from the ridge. The flange includes a recess between adjacent ones of the terminal electrodes. An outer end portion of the recess that is closer to the outer end surface is located in the outer end surface. An inner end portion of the recess that is closer to the inner end surface is located in the bottom surface, or located at a portion in the inner end surface closer to the bottom surface, than the outer end portion of the recess.

The invention discloses an automatic flight method for a wind turbine generator blade routing inspection blade area route. The method comprises the steps: step1, establishing an unmanned aerial vehicle space positioning coordinate system; step 2, setting inspection path points in a space positioning coordinate system according to the appearance parameters of the wind turbine generator; step 3, setting a flight inspection route of the unmanned aerial vehicle blade area for inspection; step 4, acquiring an infrared image of the wind turbine generator blade; step 5, dividing different temperatureregions of the infrared image, constructing a wind turbine generator contour, and setting the wind turbine generator contour as an infrared region of interest; step 6, acquiring a shot image in the inspection process to perform accurate region segmentation, and performing deep learning on the accurate region image by adopting a deep learning model to acquire confidence; step 7, performing fusioncalculation on the infrared region of interest obtained in the step 5 and the accurate region picture obtained in the step 6; step 8, regenerating an image; step 9, adjusting the position of the unmanned aerial vehicle in real time according to the image to complete automatic flight inspection.

The invention discloses a training method for automatically identifying copied images of an original document. The method comprises: constructing an image database composed of a plurality of images from an original document, the image database comprising a training data subset, and the training data subset being provided with training images comprising initial images and duplicated images and annotation results associated with the training images; obtaining an initial convolutional neural network, wherein the initial convolutional neural network is a convolutional neural network obtained by training an existing image data set; and performing fine tuning training on the initial convolutional neural network by using the training data subset, thereby obtaining a target convolutional neural network capable of outputting a classification result of the training image of the original document. According to the invention, copy detection can be easily carried out on the original document.

The invention relates to a tail door control method, device and system, computer equipment and a storage medium. The tail door control method comprises the steps that a tail door control command is received, and a detection command is sent to a detection assembly; the detection command is used for controlling the detection assembly to collect current scene data of a tail door; the current scene data collected by the detection assembly are obtained; according to the current scene data, the corresponding risk grade is determined; and according to the risk grade, a response is made to the tail door control command. According to the tail door control method, the detection assembly for collecting environment information around the tail door is arranged so that an obstacle located around the tail door can be identified conveniently automatically. Whether the obstacle possibly collides with the tail door can be judged by distinguishing the risk grade of the scene where the tail door is located, then environmental adaptability control is conducted on the tail door according to the judgment result, the intelligent degree is high, injuries to people and damage to a vehicle and objects near the tail door are avoided effectively, and universality and practicability are high.

The invention discloses an automatic identification device and method for bar codes of express items. The device comprises a mechanical arm, a first identification unit, a judgement unit and a secondidentification unit. The automatic identification method applied to the device comprises the steps that the mechanical arm grabs the express items and exposes first surfaces of the express items; thefirst surfaces are subjected to first bar code identification; with the first surfaces as the bottom surfaces of the express items, the express items are placed on a conveying belt through the mechanical arm; whether or not first bar code identification is completed is judged, and if not, the non-bottom surfaces of the express items are subjected to second bar code identification through multi-surface bar code scanning equipment arranged above the conveying belt. According to the automatic identification device, the bottom surfaces and the non-bottom surfaces of the express items are distinguished, based on a two-step identification process, the problem is solved that the bottom surfaces of the express items on the conveying belt cannot be quickly and conveniently identified, the device iseasily applied to existing automatic identification equipment and is low in transformation cost, high in identification speed and high in automatic degree, and the automatic identification process for the bar codes of the express items is optimized.

The invention provides a belt conveyor deviation detection method and system, a medium and a terminal, and the method comprises the steps of: collecting image data of a belt conveyor, and forming an original data set; marking the original data set, establishing a target detection model, and training the target detection model; acquiring a to-be-identified belt conveyor image, inputting the to-be-identified belt conveyor image into the trained target detection model, obtaining a detection result, obtaining belt conveyor edge lines according to the position information of belt conveyor edge identification frames, and judging whether the belt conveyor deviates or not according to the positions of the belt conveyor edge lines. According to the invention, the position information of belt conveyor edge recognition frames is obtained through the target detection model, belt conveyor edge lines are obtained based on the obtained position information, and whether the belt conveyor deviates or not is judged according to the positions of the belt conveyor edge lines, so that the edge lines on the two sides of the belt on the belt conveyor can be well and automatically recognized, the coordinate information of the edge lines on the two sides of the belt is judged, the situation that the belt conveyor deviates abnormally is judged in time, and unnecessary losses are avoided.

The invention discloses a method for identifying base station location area cross coverage. The method comprises the following steps: enabling a base station service area to form a triangular network, and calculating location area code (LAC) of each triangle vertex in the triangular network; and selecting any one triangle in the triangular network and the LAC of one vertex of the selected triangle, traversing all triangles connected directly or indirectly with the selected triangle, and identifying cross coverage base stations according to the selected LAC and the LAC of all triangles connected directly or indirectly with the selected triangle. The invention also discloses a device for identifying the base station location area cross coverage.

Recognition of proper nouns by an automated speech recognition system is improved by augmenting the pronunciation of each proper noun or name in the natural language of the speech recognition system with at least one “native” pronunciation in another natural language. To maximize recognition, preferably the pronunciations are predicted based on information not available to the speech recognition system. Prediction of pronunciation may be based on a location derived from a telephone number or postal address associated with the name and the language or dialect spoken in the country or region of that location. The “native” pronunciation(s) may be added to a dictionary of the speech recognition system or directly to the grammar used for recognizing speech.

An automatic defect detection method for detecting a defect in an object comprises: obtaining ultrasound scan data of the object, dividing echo amplitude values of the data into a plurality of sub-sets; assessing each sub-set to determine whether any echo amplitude values resulted from a structure of interest; computing for each sub-set, using at least a representative position of the structure portion of the sub-set and of a neighboring sub-set, a preselected mathematical function which is directly or inversely proportional to a distance between the representative positions; identifying sub-sets having a computed value that does not meet a predetermined criterion as regions of the ultrasound scan data that require inspection; issuing a notification indicating whether or not a region requiring inspection has been found; and if a region requiring inspection been found, storing data identifying the region in a database.

The invention discloses a segmented dotting and embossing welding strip, a photovoltaic assembly and a manufacturing method. The surface of the welding strip is provided with at least one sawtooth surface and at least one smooth surface, and the starting end of each sawtooth surface is provided with a concave point. According to the invention, series welding equipment is facilitated to automatically recognize the starting end of the sawtooth surface and the smooth surface, and thus the welding efficiency is improved.

The invention discloses a video moving object tampering evidence obtaining method based on a VGG-11 convolutional neural network, and the method comprises the steps: calculating a motion residual error between a forged frame and an unforged frame in a video through employing aggregation operation, and classifying the forged frame and the unforged frame; based on the motion residual error, extracting a motion residual error graph feature; constructing a convolutional neural network based on the VGG-11; training the convolutional neural network based on the VGG-11 by using the motion residual image features; and judging whether the video moving object is tampered or not by using the convolutional neural network based on VGG-11. Compared with the prior art, the method has the advantage that the forged frame in the tampered video can be better and automatically recognized.

The invention provides an imaging scaling method capable of controlling main body size based on content, belonging to the technical field of information. The method comprises six steps of calculating gradient, acquiring saliency, calculating importance degree, initializing grid, deforming the grid and mapping image content, wherein a multi-resolution saliency model based on sparsity characteristic provided in step 2 can more exactly identify the main body area with different sizes in the same image compared with the existing mainstream method; in step 3, the weighted average of gradient and saliency is taken as the importance degree, which can better identify edges and inner areas of the image main body; and in the step 5, gridding deformation energy can carry out optimal control in accordance with the importance degree so as to cause the grid with great importance degree to approximate unified scaling of assigned scaling factor as far as possible and gather the deformation resulting from image size and length and width proportion in isomorphism areas with small importance degree as far as possible, thereby protecting the main body areas in the image from deforming, and meeting demands of controlling the main body size by users in accordance with the content.

The invention relates to a color image holographic watermarking method based on wavelet de-noising. The wavelet de-noising processing is performed on an extracted watermark image so as to achieve that a computer accurately and objectively distinguishes watermark information. Firstly, forward wavelet transformation and optical computer-generated holographic double-random-phase encryption watermark embedding are performed on an original color image to obtain a color image containing the watermark information; secondarily, reverse wavelet transformation and optical computer-generated holographic double-random-phase decryption watermark extraction are performed on the obtained color image containing the watermark information to obtain a holographic watermark image; thirdly, two-dimensional wavelet de-noising and binarization processing are performed on the holographic watermark image to obtain a de-noising watermark image. Compared with an existing digital watermark algorithm, the method facilitates watermark image analysis and processing, brings convenience to the computer for accurate and objective distinguishing on the watermark information, and enables the extracted watermark image to be accurate. The color image holographic watermarking method can have application value on batch watermark authentication.

The invention discloses an anti-channel conflict method for electronic business. The anti-channel conflict method comprises the following steps: S1, ordering by customers and obtaining order information by background; S2, comparing keywords in the order information with a blacklist in a database; S3, if no keywords exist in the blacklist in the comparison in the S2, extracting address characters;S4, combining the characters in the S3 to obtain combined words, and then comparing the combined characters with addresses in an address base; S5, comparing the keywords in the S4 to obtain one part;determining whether suspicion indexes need to be manually reviewed or not when whether the compared part belongs to the sales scope of dealers cannot be determined; S6, if the suspicion indexes in theS5 are not up to standard, packing and delivering by adopting packing containers. By use of the method disclosed by the invention, the channel conflict risk in an electronic business system can be quickly and automatically identified; even if the dealers maliciously change the addresses of recipients, a system can recognize the address change by similarity matching, so that the malicious channelconflict behavior of the dealers can be effectively inhibited.

The invention discloses an LED module and a recognition device and recognition method of the LED module. The LED module comprises a shell, an identification resistor is arranged in the LED module, two detecting points are arranged on the shell, and the two detecting points are connected with the two ends of the identification resistor respectively. The recognition device of the LED module comprises a voltage dividing circuit, and the voltage dividing circuit is provided with two branch voltage connecting points and further comprises a detecting circuit, wherein the two branch voltage connecting points are correspondingly connected with the detecting points to be detected of the LED module, and the detecting circuit detects and outputs the branch voltage value of the voltage dividing circuit. According to the LED module and the recognition device and the recognition method of the LED module, through the fact that the LED module is provided with the identification resistor and the detecting points, the type of the LED module can be automatically recognized by using the recognition device for the LED module conveniently, a controller can automatically select a corresponding driving mode to drive the LED module according to the type of the module, the problems of low efficiency and inconvenient using due to the fact that the driving method of the LED module needs to be set manually are solved, and the intelligent degree of the control over the LED module is improved.

The invention aims at providing an image recognition method. According to the image recognition method, welding spots for welding silver wire on the surface of a silver foil in an acquired image are recognized. The acquired image comprises at least one first region corresponding to the silver foil and at least one second region corresponding to the silver wire. The image recognition method comprises the following steps of acquiring the gray?value of each pixel in the first region and the second region, acquiring characteristic points of the first region according to the gray?value of each pixel in the first region, and seeking the welding points in the second region according to the gray?value of each pixel in the second region and the characteristic points. In comparison with the prior art, recognition is carried out according to the gray difference between the silver foil and the silver wire, the position of the silver wire is sought around mass points of the silver foil, the welding spots are finally determined, and the positions of the welding spots are then conveniently, quickly and automatically recognized.

The invention discloses a thunderstorm strong wind storm body identification method and system based on ground observation, and relates to the field of meteorology and computers. The method comprisesthe following steps: step 1, acquiring a radar puzzle, a thunderstorm strong wind storm body and related attribute information of the thunderstorm strong wind storm body output by a disastrous weathershort-time proximity forecasting and early warning service system, and acquiring ground wind speed observation information received by a ground automatic station; 2, preprocessing the radar puzzle, the related attribute information of the thunderstorm strong wind storm body and the ground wind speed observation information to obtain thunderstorm strong wind storm body data to be identified; and 3, performing fuzzy logic calculation processing on the thunderstorm strong wind storm body data to be identified to obtain a matching result of the thunderstorm strong wind storm body and strong windobserved on the ground in the ground wind speed observation information, and completing identification. The problem of reasonable matching of strong wind observed by a ground station and a storm bodyobserved by a Doppler radar can be solved.