306results about How to "Efficient and accurate identification" patented technology

In general, the present invention is directed to systems and methods for finding the position and shape of an object using video. The invention includes a system with a video camera coupled to a computer in which the computer is configured to automatically provide object segmentation and identification, object motion tracking (for moving objects), object position classification, and behavior identification. In a preferred embodiment, the present invention may use background subtraction for object identification and tracking, probabilistic approach with expectation-maximization for tracking the motion detection and object classification, and decision tree classification for behavior identification. Thus, the present invention is capable of automatically monitoring a video image to identify, track and classify the actions of various objects and the object's movements within the image. The image may be provided in real time or from storage. The invention is particularly useful for monitoring and classifying animal behavior for testing drugs and genetic mutations, but may be used in any of a number of other surveillance applications.

A system for identifying test tube types and properties in a sample handling machine using visual information automatically obtained by an optical imager and then processed using vision processing methods. The system includes an optical imager positioned to capture images containing one or more test tubes in a rack and a microcontroller programmed to extract predetermined regions of interest and interpret the optical information in the image to decipher the dimension of the test tubes, determine the presence or absence of caps on the test tubes, decode any encoded data, and interpret custom symbologies. The system may then determine the nature of the test tubes or other containers presented before the image and provide that information to the sample handling machine to assist with processing of samples.

A method and system for altering a presentation of broadcast content is disclosed. Embodiments provide an accurate and efficient mechanism for suppressing advertisements by using downloaded templates to identify advertisements before suppression for a duration indicated by information downloaded about the advertisement, where the suppression is effectuated by controlling a display or set-top box using remote-control functionality. In one embodiment, a digital signal comprising broadcast content is accessed. A plurality of templates are received and stored, wherein each template represents a known portion of the broadcast content. The digital signal is then compared against a portion of the plurality of stored templates. Upon a match, a control signal operable to alter a component of a presentation of the known portion of the broadcast content is generated.

A computer may identify an individual according to one or more biometrics based on various physiological aspects of the individual, such as metrics of various features of the face, gait, fingerprint, or voice of the individual. However, biometrics are often computationally intensive to compute, inaccurate, and unable to scale to identify an individual among a large set of known individuals. Therefore, the biometric identification of an individual may be supplemented by identifying one or more devices associated with the individual (e.g., a mobile phone, a vehicle driven by the individual, or an implanted medical device). When an individual is registered for identification, various device identifiers of devices associated with the individual may be stored along with the biometrics of the individual. Individuals may then be identified using both biometrics and detected device identifiers, thereby improving the efficiency, speed, accuracy, and scalability of the identification.

A method of screening complementary metal-oxide-semiconductor CMOS integrated circuits, such as integrated circuits including CMOS static random access memory (SRAM) cells, for transistors susceptible to transistor characteristic shifts over operating time. For the example of SRAM cells formed of cross-coupled CMOS inverters, separate ground voltage levels can be applied to the source nodes of the driver transistors, or separate power supply voltage levels can be applied to the source nodes of the load transistors (or both). Asymmetric bias voltages applied to the transistors in this manner will reduce the transistor drive current, and can thus mimic the effects of bias temperature instability (BTI). Cells that are vulnerable to threshold voltage shift over time can thus be identified.

In general, the present invention is directed to systems and methods for finding the position and shape of an animal using video. The invention includes a system with a video camera coupled to a computer in which the computer is configured to automatically provide animal segmentation and identification, animal motion tracking (for moving animals), animal-posture classification, and behavior identification. In a preferred embodiment, the present invention may use background subtraction for animal identification and tracking, and a combination of decision tree classification and rule-based classification for posture and behavior identification. Thus, the present invention is capable of automatically monitoring a video image to identify, track and classify the actions of various animals and the animal's movements within the image. The image may be provided in real time or from storage. The invention is particularly useful for monitoring and classifying animal behavior for testing drugs and genetic mutations, but may be used in any of a number of other surveillance applications.

The present invention provides a novel system and method for identifying individuals and for face recognition utilizing facial features for face identification. The system and method of the invention comprise creating facial features or face patterns called face pattern words and face pattern bytes for face identification. The invention also provides for pattern recognitions for identification other than face recognition. The invention further provides a means for identifying individuals based on visible and / or thermal images of those individuals by utilizing computer software implemented by instructions on a computer or computer system and a computer readable medium containing instructions on a computer system for face recognition and identification.

The invention provides a ship target identification system and method based on artificial intelligence image processing. The method comprises the following steps of collecting Video information samples a by adopting ship video equipment in a high-dynamic water surface environment; Labeling the sample by adopting a manual and automatic tracking algorithm; Under the condition of low-quality video; identifying a sea surface weak target on the basis of the synthesized sample; Target identification method, In the prior art, only a shore-based system is used for identifying a marine weak target; According to the technical scheme, the technical effects that double-dynamic identification is carried out under various platforms such as a shore-based platform and a water-surface platform, a weak target on the sea is identified, and a ship target in the severe environment on the sea is accurately and efficiently identified are achieved through one set of system.

The invention discloses a pavement marker information processing method and apparatus. The processing method comprises the steps of obtaining road point cloud data of a target road based on track point information of a laser collection device and laser point cloud data of the target road, wherein the laser point cloud data is obtained by performing point cloud measurement on the target road through the laser collection device, and the pavement point cloud data is used for describing pavement information of the target road; according to a preset range, dividing a region which the pavement point cloud data belongs to into a plurality of pavement regions; obtaining reflectivity screening thresholds of the pavement regions, and screening pavement marker data in the pavement point cloud data of the pavement regions according to the reflectivity screening thresholds of the pavement regions and a preset extraction window, wherein the reflectivity screening thresholds of at least two pavement regions in the pavement regions are different; and identifying pavement markers in the pavement marker data. According to the method and the apparatus, the technical problem of low identification accuracy of the pavement markers is solved.

A system for identifying test tube types and properties in a sample handling machine using visual information automatically obtained by an optical imager and then processed using vision processing methods. The system includes an optical imager positioned to capture images containing one or more test tubes in a rack and a microcontroller programmed to extract predetermined regions of interest and interpret the optical information in the image to decipher the dimension of the test tubes, determine the presence or absence of caps on the test tubes, decode any encoded data, and interpret custom symbologies. The system may then determine the nature of the test tubes or other containers presented before the image and provide that information to the sample handling machine to assist with processing of samples.

A system providing precise determination of resource use within a software system such as a database system. A user, such as a database administrator, operates a user interface to select and analyze portions of an application for resource use. In one embodiment in a DB2 database system individual SQL statement resource use can be determined. The invention is capable of uniquely identifying even dynamic SQL statements (i.e., Equalized I.D). Prediction information is presented for when a query, or other database task, will end based on selectable parameters such as optimizer cost, estimated cardinality, and delta delivery rates of data. A user can create or select event monitors to acquire statistical data on a database task. The results of activating the event monitor are presented to the user such that the user can progressively examine portions of an application, package, section and even individual statements. The user interface allows a user to efficiently and precisely identify processing shortcomings so that the performance of a database task can be improved. The interface also allows a user to sort, or otherwise organize, the statistical data so that analysis of the data is improved and maximum value is derived from the SQL cost evaluated data.

To detect a slip state in a contact region of a rotating tire on a road surface, first, the first radical direction data and the second width direction data of measurement data of acceleration at a tread portion of the rotating tire for a duration corresponding to one round of tire rotation are acquired. Second, time series data of acceleration due to tire deformation is extracted from the first radical direction data and displacement data is obtained by subjecting the time series data of acceleration due to tire deformation to a time integration of second order, thereby a deformation at the tread portion is calculated and a contact region of the rotating tire is determined from the calculated deformation. Next, from the second width direction data, a slip region within the determined contact region is specified based on vibration level of the second data.

The invention discloses a speech recognition method, a device, equipment and a computer-readable storage medium. The method comprises the following steps: extracting preselected feature vectors of real speech in a pre-established corpus; extracting preselected feature vectors of real speech in the pre-established corpus; extracting preselected feature vectors of real speech in the pre-establishedcorpus; extracting preselected feature vectors of real speech in the pre-established corpus; according to the preset speech synthesis algorithm and the text information corresponding to the real speech in the corpus, extracting the forged speech is synthesized and the preselected feature vector of the forged speech; training a true / false speech classifier by using a preselected feature vector of the real speech and a preselected feature vector of the forged speech; recognizing the speech to be detected by the trained true-false speech separator so as to judge whether the speech to be detectedis forged speech or not. The method, the device, equipment and the computer-readable storage medium provided by the invention can accurately and efficiently recognize the counterfeit speech in the actual scene and effectively prevent the pseudo sound attack.

The invention relates to a bill processing method, a bill processing device and a computer readable storage medium. The bill processing method comprises the following steps: collecting a bill image; extracting position information of the bill image to obtain an overall positioning result of the bill image; based on the overall positioning result, extracting frame information of the bill image to obtain a frame positioning result of the bill image; restoring the original shape of the bill image through affine transformation based on the frame positioning result; according to the original shape,carrying out region division on the bill image to obtain a partitioning result of the bill image; performing interference information removal on the bill image based on the partitioning result; and detecting and identifying text information in the bill image to obtain bill content. According to the invention, the accuracy of bill identification can be improved, the working efficiency of bill statistics personnel is improved, and the labor cost is reduced.

The invention provides a mental state assessment and regulation system based on transcranial stimulation. The mental state assessment and regulation system is characterized by comprising the followingparts: a physiological signal acquisition module for acquiring a physiological signal of a tested subject; a mental state assessment module used for performing physiological signal preprocessing, feature extraction and mental state analysis to obtain a mental state assessment result; control equipment used for judging the mental state evaluation result and adjusting stimulation parameters according to the mental state evaluation result; and transcranial stimulation equipment used for performing transcranial stimulation on the tested subject according to the stimulation parameters. The mentalstate assessment and regulation system can carry out personalized targeted mental state regulation on the tested subject, has a good mental state adjustment effect, is convenient and effective to use,and has the advantages of noninvasiveness, effectiveness, safety, easiness in operation and the like. The invention further provides a working method of the mental state assessment and regulation system.

A method of screening complementary metal-oxide-semiconductor CMOS integrated circuits, such as integrated circuits including CMOS static random access memory (SRAM) cells, for n-channel transistors susceptible to transistor characteristic shifts over operating time. For the example of SRAM cells formed of cross-coupled CMOS inverters, static noise margin and writeability (Vtrip) screens are provided. Each of the n-channel transistors in the CMOS SRAM cells are formed within p-wells that are isolated from p-type semiconductor material in peripheral circuitry of the memory and other functions in the integrated circuit. Forward and reverse body node bias voltages are applied to the isolated p-wells of the SRAM cells under test to determine whether such operations as read disturb, or write cycles, disrupt the cells under such bias. Cells that are vulnerable to threshold voltage shift over time can thus be identified.

The invention discloses a novel Feulgen staining method-based abnormal cervical cell automatic identification method. According to the automatic identification method disclosed by the invention, the features of the cervical cells are extracted, and the cervical cell classifier is trained, so that the identification of abnormal cervical cells is realized; the process of producing the cervical cellclassifier mainly comprises the following four steps of: 1, dyeing a cervical cell slide by using a Feulgen dyeing method, and automatically scanning the slide by using a microscope to generate a digital view; 2, segmenting cervical cell nuclei in the view map by using a Surf algorithm in combination with a RegionGrowing algorithm; 3, extracting DNA content information of the cell nucleus, cell nucleus morphological characteristics, cervical cell image texture characteristics and the like, and constructing a characteristic vector for representing the abnormal degree of each cervical cell; andstep 4, constructing and training a neural network classification model based on the feature vectors to obtain the cervical cell classifier; and finally, predicting a new cervical cell feature vectorby using the trained cervical cell classifier, thereby realizing the purpose of identifying abnormal cervical cells. Experiments show that the abnormal cervical cell automatic identification method based on the Feulgen staining method can complete the identification task of the abnormal cervical cell with high precision and efficiency, and the automatic identification method has high practical value when being applied to a real product.

The present invention relates to a shielded face identification method and device. The method comprises: obtaining a target shielded face image; performing preprocessing of the target shielded face image, obtaining the sample data of the target shielded face image, and taking the sample data as the test sample data; determining the estimated value of the first expression coefficient of the test sample data on a training sample set; determining a shielding mark according to the training sample set, the test sample data and the estimated value of the first expression coefficient; determining the estimated value of the second expression coefficient of the test sample data on the training sample set according to the shielding mask; constructing an identity identification model according to the training sample set, the shielding mask and the estimated value of the second expression coefficient; and inputting the test sample data to the identity identification model, and obtaining the identity information of a figure expressed by the target shielded face image. The shielded face identification method and device can simply, efficiently and accurately identify the shielded face.

The invention discloses a self-service vending method of a container on the basis of binocular vision. The method comprises the following steps that a first planar image and a second planar image of atarget area of a vending machine or a vending cabinet are shot through binocular cameras fixed at the opposite positions in the container; a binocular vision processing algorithm is adopted for processing the first planar image and the second planar image, and a three-dimensional coordinate space of a target object is established; on the basis of deep leaning cognitive features of images in the three-dimensional coordinate space, the goods type of the object is identified.

The embodiment of the invention discloses a reinforcing steel bar end face recognition method based on a deep convolutional neural network. The method comprises the steps of cutting a reinforcing steel bar end face area image and a non-reinforcing steel bar end face area image through traversal of a sliding window; establishing an image library of the area images and dividing the images in the image library into training samples and test samples; applying the training sample to the training of the deep convolutional neural network, and determining learning parameters in the deep convolutionalneural network; After the convolutional neural network is trained, traversing is conducted on the original image of the end face of the reinforcing steel bar used for testing through a sliding window,and a result obtained after traversing at each time is transmitted to the trained convolutional neural network to be recognized; And marking red points on the identified end faces of the reinforcingsteel bars, then clustering the marked red points, finding out the centers of the identified end faces, and marking the centers on the original image for testing. According to the method, the strong feature learning capability of the deep convolutional neural network is fully utilized, so that the steel bar end face is efficiently and accurately identified.

The invention provides a method for generating an image fingerprint and a method for searching similar images and recognizing image embezzlement. The method is characterized in that an image is provided with M rows and N columns of pixels, and comprises the following steps of a) for each pixel point Xij in the image, i=2, 3,...M, j=2, 3,...N, comparing a gray value of the pixel point with that of a left pixel point and an upper pixel point of the pixel point, wherein the left pixel refers to the left pixel in the four pixels which are adjacent to and arranged above, below, on the left and on the right of the pixel, and the upper pixel refers to the upper pixel in the four pixels which are adjacent to and arranged above, below, on the left and on the right of the pixel; b) giving a fingerprint value Zij to the pixel pint Xij according to different comparison result; c) combining the fingerprint values Z of each pixel point to obtain a fingerprint of the image with the length of (M-1)*(N-1). By utilizing the image embezzlement recognizing method, the similar image can be precisely and high efficiently recognized, and the image embezzlement behavior can be prevented.

The invention provides a vehicle recognition method based on image detection. The vehicle recognition method includes the steps that a vehicle image of a vehicle is obtained by means of a monitoring device, feature filtering processing and normalization processing can be carried out on the vehicle image, and accordingly the vehicle image undergoing normalization processing can be obtained; a feature vector of the vehicle image obtained after normalization processing is obtained; the feature vector of the vehicle image undergoing normalization processing is compared with a corresponding feature vector of vehicle samples in a database, and accordingly the vehicle sample most similar to the vehicle image undergoing normalization processing in the database is obtained; the type of the obtained vehicle sample in the database serves as the type of the vehicle. According to the vehicle recognition method based on image detection, by means of normalization processing and comparison processing, the type of the vehicle can be quickly, efficiently and accurately recognized; the technical problems that an existing vehicle recognition method is low in recognition efficiency and poor in recognition accuracy are solved.

The invention discloses an intelligent CAD drawing recognition method used for solving the technical problem that an existing CAD drawing recognition method is poor in practicability. According to thetechnical scheme, a large number of imported drawings are classified, and whether an electrical device, components and other devices exist or not is automatically distinguished through a convolutional neural network based on a region. For a drawing containing the electrical device or the components, the components and the text information in the drawing are respectively positioned and recognizedby using a region-based convolutional neural network method. Each component matches with the corresponding text information, and generating a BOM list. A CAD drawing is processed and analyzed throughan intelligent recognition and matching method. Manpower consumption is reduced to a great extent. High recognition speed and accuracy are achieved and practicability is good.

The invention relates to the technical field of horseshoe kiln energy consumption anomaly detection, in particular to a horseshoe kiln energy consumption anomaly detection method based on improved density peak clustering. An artificial bee colony algorithm is adopted to achieve an adaptive optimization process of a truncation distance, an outlier coefficient strategy is set to achieve a function of automatically dividing cluster centers and outliers, and intelligent, rapid and accurate kiln energy consumption anomaly detection is achieved. According to the method, the adaptive density peak clustering algorithm is modified, and clustering analysis is performed on the energy consumption data in the production process of the glass melting furnace by using the adaptive density peak clusteringalgorithm, so that abnormal energy consumption samples can be efficiently and accurately identified.

The invention relates to a differential cloud cluster extrapolation rainfall forecasting algorithm based on Hu moments, and belongs to the technical field of meteorological forecasting. In the algorithm, cloud growth regions and vanishing regions are the objects matched by a cloud cluster contour matching algorithm based on Hu moments; and the cloud contour matching algorithm based on Hu moments and a differential cloud cluster extrapolation algorithm are adopted, finally, that the data representativeness of a high reflectance region is always higher than that of a low reflectance region is comprehensively considered, an extrapolated image is mainly compared with a real image, and the similarity is weighted and summed to obtain the comprehensive similarity. The algorithm is high in speed and high in recognition, provides a good description of the shape of a cloud cluster, and has a high accuracy and efficiency in the shape recognition of the cloud cluster.

The present invention relates to compositions and methods for identifying, monitoring, and sorting specific genetically-modified plant portions from other genetically-modified plant portions. The present invention also relates to compositions and methods for identifying, monitoring, and sorting specific genetically-modified plant portions from non-genetically modified plant portions where both are present in a mixture. Either or both of the genetically modified plant portions or the non-genetically modified plant portions can comprise a distinguishable marker which is identified and used for sorting such mixtures of plant portions. The present invention is also directed toward kits useful in the methods disclosed herein. The compositions, methods, and kits of the present invention are used inter alia in high-throughput, sorting systems for identity preservation of a seed stock, to provide seed populations that are free of genetically-modified seeds, to isolate hybrid seed uncontaminated with selfed seed, and to isolate one type of genetically-modified plant portion from a mixture of genetically-modified plant portions.

The invention discloses a training and recognition method and device for a visible light fault model of a photovoltaic module. The method comprises the steps of in response to an obtained to-be-detected image, cutting the to-be-detected image; marking pixel coordinates of each piece of image data in the to-be-detected image; in response to the acquired multiple pieces of image data, outputting target object image data in the multiple pieces of image data by a Faster-RCNN training model; judging whether the pixel coordinates of the image data of a certain obstacle are within the range of the pixel coordinates of all photovoltaic module image data or not; and if the pixel coordinates of the image data of a certain obstacle are within the range of the pixel coordinates of all photovoltaic module image data , outputting the pixel coordinates of the image data of the obstacle. By extracting the image features of the photovoltaic module and the bird droppings, the coordinate range of the photovoltaic module and the coordinates of the bird droppings in the detected image are identified, and the overlapped part of the coordinate range and the coordinates of the bird droppings is marked asthe bird droppings covering fault point of the photovoltaic module, so that the identification of the bird droppings shielding fault point of the photovoltaic module is realized.

The invention provides an insurance policy risk evaluating method and device, terminal equipment and a storage medium. The method comprises that an insurance policy relation network composed of insurance policies associated with one another according the same attribute information is established; case information of historical insurance policies is obtained and serves as sample data; a scoring model which comprises a preset number of scoring items and scores of each scoring item in different scoring conditions is constructed according to the sample data; the scoring model is used to evaluate the insurance policy relation network, and a total evaluation score is obtained; and an insurance policy fraud risk in the insurance policy relation network is analyzed according to the total evaluation score. Insurance policy data is dug and analyzed in the associated way, the evaluation model is constructed, related staff are helped identify the risk of group insurance fraud precisely and efficiently, and the insurance policy identification efficiency and group insurance fraud identification rate are improved.

The invention relates to a shielding face identification method and a shielding face identification device. The method comprises steps that a target shielding face image is acquired; the target shielding face image is pre-processed to acquire the sample data of the target shielding face image, and the sample data is taken as the test sample data; an estimate of a first expression coefficient of the test sample data on a training sample set is determined; according to the training sample set, the test sample data and the estimate of the first expression coefficient, a shielding mask and a recovery mask are determined; the no-shielding estimation data of the test sample data is determined according to the estimate of the first expression coefficient, the shielding mask and the recovery mask; an estimate of a second expression coefficient of the no-shielding estimation data on the training sample set is determined; an identity identification model is constructed according to the training sample set and the estimate of the second expression coefficient; the no-shielding estimation data is inputted to the identity identification model, and the identity information of the target shielding face image is acquired. Through the method, a shielding face can be simply, efficiently and accurately identified.

The invention mainly relates to a method for identifying authenticity of a medicago sativa seed and special primers thereof, in particular to a gene detection kit of medicago sativa and melilotus albus and a detection method thereof. The kit for identifying the authenticity of the medicago sativa seed is mainly characterized in that: the total volume of a medicago sativa PCR (Polymerase Chain Reaction) detection system is 20 mul; and the kit comprises 2*PCR Master, the special primer MsITS-F (5'-3'): TTTTTGAACGCAAGTTGCGC and the special primer MsITS-R (5'-3'): CAACCAACCACGAGACAGA and ddH2O. Detection result shows that the technology can be used for identifying the authenticity of the medicago sativa seed, has the characteristics of high sensitivity, high accuracy, high repeatability, quickness, convenience and the like, has broad application prospect in medicago sativa industrial promotion of China, can guarantee that the production of livestock husbandry of China is served by the medicago sativa well and lays a solid foundation for guaranteeing grain safety of China.