1053 results about "Picture recognition" patented technology

A method and device for authorizing the use of an electronic device by picture identification, comprising: retrieving a number of pictures from a picture passkey library, displaying a subset of the pictures on a display of the electronic device, receiving a user input by which at least one of the displayed pictures is selected, comparing the at least one selected picture with predetermined identification data, and performing an authorizing operation, if the at least one selected picture matches with the predetermined identification data, wherein the pictures in the picture passkey library comprise user generated pictures.

Individuals asked to present two forms of picture identification face a problem as individuals usually only have one form of picture ID, that form of identification generally being a state-issued driver's license. The disclosed invention is a system and method for a party, generally a Customer or Patron, to establish proof of identification to another party, generally a Merchant or Agent, via a mobile phone using the mobile communications networks. The invention thereby provides the Merchant or Agent with an additional capability to verify the identity of Customers, including verification of the date of birth of the Customer. In a preferred embodiment, the invention uses the Short Message Service (“SMS”) and Multimedia Messaging Service (“MMS”) features available on mobile phones, or other wireless communication devices, to establish the identity of a Customer after communicating a request for such identification to a government administered identification database system.

The invention provides an auto-tracking recognition system and a method. The system comprises a fixed viewing field camera, a suspicious object determination module, a trail tracking module, a control module, a variable viewing field camera, a picture recognition module, a contrast module and an alarm module, wherein, the suspicious object determination module is used for analyzing a video picture collected by the fixed viewing field camera, determining a suspicious object, and acquiring the position of the suspicious object; the trail tracking module is used for determining the motion trail of the suspicious object; the control module is used for outputting a control signal according to the position and the motion trail of the suspicious object; the variable viewing field camera is used for acquiring the video picture in accordance with the feature recognition requirement based on an adjustment parameter of the control signal; the picture recognition module is used for acquiring a feature parameter of the suspicious object; the contrast module is used for contrasting the acquired feature parameter with the feature information of the contrasted object in a feature database; and the alarm module is used for giving an alarm to the suspicious object in accordance with the set alarm condition. The invention can automatically track and collect detail pictures of the suspicious object, and efficiently realize the recognition of the suspicious object, thereby giving an early and reliable alarm.

The invention discloses a method of monitoring image mark quality and an apparatus thereof. The method comprises the following steps of executing an acquisition step and acquiring accuracy of mark data of a single task; issuing an image to be marked to a mark account, wherein the image to be marked includes a sample image of a predetermined proportion; acquiring manual mark data of the image to be marked, wherein the manual mark data includes current mark data of the sample image; acquiring an accurate specific value of the current mark data relative to standard mark data of the sample image; responding to the specific value which is greater than a pre-determined value; inputting the image to be marked into a picture identification model so as to acquire automatic mark data; comparing the manual mark data and the automatic mark data so as to acquire accuracy of the mark data of the single task; calculating a variance of the accuracy of mark data of multiple tasks and preset average mark accuracy; and based on the variance, determining the number of the sample images issued to the mark account and / or determining an image type suitable for marking of the mark account. Efficiency and accuracy of mark result detection are high.

The invention discloses an electronic device, a bill information identification method and a computer readable storage medium. The method includes performing identification on bill type of a bill in abill picture by utilizing a pre-trained bill picture identification model after receiving the bill picture to be processed; performing inclination correction on the bill picture by utilizing a predetermined correction rule; determining to-be-identified fields corresponding to the identified bill type; determining a first identification model corresponding to each to-be-identified field, performing area identification on line field areas of the bill picture subjected to inclination correction by calling the corresponding first identification models so as to identify the target line character zone containing the character information of each to-be-identified field; determining a second identification model corresponding to each to-be-identified field and performing character identificationby calling the corresponding second identification models so as to identify the character information contained by the target line character zones of each to-be-identified field. The technical schemeachieves accurate and high-efficiency automatic identification on text information in bill pictures.

The embodiment of the invention discloses a video identifying method, device and system. After video information needing identification is obtained, correlated information of a picture, a voice, a subtitle, and a video in the video content is identified respectively to obtain multi-dimensional identification results including a picture identification result, a voice identification result, and a text identification result and the like; and fusion calculation is carried out on the identification results to obtain a video identification result. According to the scheme, the identification accuracy is enhanced and the identification result is improved.

The invention belongs to the technical field of mobile terminals, and provides a photographing method and device of a mobile terminal. The photographing method comprises the steps that picture recognition is conducted on a preview picture of a camera of the mobile terminal, and the position of a face in the preview picture is determined; the rotation angle is calculated based on the position relation between the position of the face and the golden section ratio of the preview picture; the camera of the mobile terminal is controlled to rotate by the rotation angle so as to accomplish photographing. By the adoption of the photographing method of the mobile terminal, when a user conducts self photographing operation through the mobile terminal, the camera of the mobile terminal can rotate according to the position relation between the position of the face and the golden section ratio in the framing process, and therefore the good picture composition effect is automatically generated, the phenomenon that due to the fact that photographing operation is repeatedly conducted when taken pictures are poor in composition, meaningless energy consumption of the mobile terminal is caused is avoided, and energy consumption is reduced.

The present invention discloses a video surveillance system and method for face tracking and capturing. The system includes a data acquisition unit, a data analysis unit, a data processing unit, a data transmission unit, a rear end main control unit and an alarm unit. The data acquisition unit consists of cloud platform cameras, the data analysis unit analyzes a moving target in a video image, and captures the face of the moving target, the data processing unit controls the video acquisition unit to perform tracking, video acquisition and capturing of the face of the moving target, the data transmission unit sends the video image and a high definition picture of the face of the moving target acquired by the video acquisition unit to the rear-end main control unit for the rear-end main control unit to perform picture identifying analysis of the target face, and add an alarm signal to the corresponding face while sending the position information of the video image where the face picture exists to the alarm unit, and the alarm unit sends the alarm signal, which effectively solves the problems such as the small size of the captured picture of the face of the moving target, unclear image, and increases the degree of user experience.

An device for diagnosis and control of honeybee varroatosis (Varroa mite infection) comprises one or several cameras (310), connected to image processor (311). The software of the device is capable of processing the picture of honeybee, recognition the presence or absence of varroa mite on the body of the bee on the field-of-view of the cameras or in the chamber for observation, counting the infected and non-infected bees, separating the infected bees from non-infected bees or killing the recognized varroa mites. Controlled gates (314, 315, 316), valves, positioning actuators and / or heaters (329, 330) are used as output devices for realization of the said actions. The method of image processing for recognition of varroa mite or other parasites comprises of searching reflections in the image and analysis of the surroundings of the found reflections. Also the method for determination of location of body elements of insect in an image is disclosed.

The invention discloses an intelligent car insurance loss assessment method and system based on image recognition. The method comprises the following steps: receiving picture information captured by a client; identifying the picture information to obtain the loss part and the loss degree of a vehicle; according to an incidence relation between the loss part and the loss degree of the vehicle and fitting man-hour, calculating corresponding loss price and forming a loss result. The method can quickly, efficiently and automatically determine the loss part, the loss fitting, and the loss degree of the vehicle by identifying and processing the vehicle picture captured and uploaded by the client and required to be subjected to loss assessment so as to automatically create loss items and to realize automatic claim settlement. In cooperation with a pocket claim client, the method and system can achieve a whole process without the audit staff and can greatly improve loss assessment efficiency.

The invention relates to an intelligent lung cancer early cell pathological image recognition processing method, which comprises image pretreatment, image segmentation, laminate cell separation and reconstruction, cell character extraction and cell classification. The invention has the advantages that the image segmentation based on reinforcement learning uses incremental learning and continuous interaction with environment to search for optimized segmentation threshold value to obtain the segmentation effect which average value is 91%, the laminate cell separation and reconstruction uses B spline and modified deBoor-Cox method to simulate true cell edge better, the classifier uses general vote method to avoid low classifying accuracy of single classifier and improve total classifying accuracy, the application of two-stage classifier can reduce the possibility of false positive and false negative. Tests prove that the classifying accuracy of cancer or no cancer can average reach 93.8%, the classifying accuracy of squamous cell carcinoma, adenocarcinoma and small cell carcinoma average reaches 75%, and the false positive and negative average reach 4-6%.

A stereo picture recognition device has a time-sharing picture display shutter disposed between a display surface of a display device and a viewer. The time-sharing picture display shutter has a function of shutting off only an area covered by a viewing angle which corresponds to a display area of the display device.

The invention discloses a method, a device and a system for testing an application program. The method comprises steps as follows: an interface where testing operation of the tested application program is executed, operation execution coordinates during execution of testing operation on the interface and a recorded picture corresponding to the operation execution coordinates are acquired on a first terminal; an interface picture acquired when the interface is operated on a second terminal is intercepted; the interface picture and the recorded picture are subjected to picture recognition, and operation playback coordinates corresponding to the operation execution coordinates in the interface picture are determined according to a picture recognition result; the testing operation is executed on the tested application program on an interface of the second terminal according to the operation playback coordinates. With the adoption of the method, the device and the system for testing the application program, the problem that a software testing result is inaccurate due to the fact that a record and playback method used for testing software in the prior art cannot adapt to terminals with different resolutions is solved, and the operation execution coordinates can be acquired accurately, so that the testing operation can be accurately performed on the second terminal, and an accurate testing result of the application program can be acquired.

The invention provides a training method and a device for a sensitive picture classifier; the regional shape characteristic of the training picture set is extracted; the distribution characteristics of the regional shape characteristic in the positive example sample set, the first negative example sample set and the second negative example sample set are measured; the divisibility of the regional shape characteristic is determined according to the distribution characteristics; the regional shape characteristic with the divisibility relative to the first negative example sample set is marked as the first characteristic group; the regional shape characteristic with the divisibility relative to the second negative example sample set is marked as the second characteristic group; the first classifier is obtained through the characteristic training of the first characteristic group; the second classifier is obtained through the characteristic training of the second characteristic group; the invention also provides the method and the device which adopt the sensitive picture classifier to recognize the sensitive pictures; the invention improves the accuracy of the sensitive picture recognition.

The invention discloses a car insurance intelligent loss assessment method and system. The method comprises the steps: building a vehicle claim settlement loss assessment database; receiving a car insurance claim settlement instruction and a vehicle loss image which are transmitted by a client, and recognizing a loss part and loss degree of the vehicle through combining with the vehicle claim settlement loss assessment database; searching a loss assessment scheme from the vehicle claim settlement loss assessment database according to the vehicle type, the loss part, the loss degree and a maintenance factory selected at a client, and calculating the loss assessment price, thereby improving the loss assessment efficiency, reducing the loss assessment bias of a case, and reducing the manual operation risks. The method carries out the matching with the price in the database on the basis of image recognition, completes the systematic automatic loss assessment, and improves the intelligentization degree of loss assessment.

The invention discloses a method and instrument for counting the people coming in and going out, composed of CCD camera and PC (personal computer), where the CCD camera is fixed on the top of a passageway and connected with the PC by lines, and the PC is installed with picture collection card and picture recognition algorithm software that comprises five modules, i.e. target-background separating module, background reconstructing module, target picture post-processing and recognizing module, target detecting-tracking module and data accounting-outputting module; and the counting method uses the picture collection card to collect the pictures, circularly, real-timely recognizes each collected picture frame, then makes target-background separation, reconstruction, connectivity analysis and false target elimination on the target pictures, after separating the bonded target pictures, compares them with the standard target pictures of human bodies to judge their behavior, and finally counts the people coming in and going out and outputs the count. The invention can be adapted to count the people coming in and going out in various orders.

The present invention relates to a brain-machine interface system based on face recognition specific wave N170 component, which implements sorting of facial pictures and object pictures, and comprises a picture stimulator, a cortical electric signal collector, a face recognition specific wave N170 detector, and a picture sorter. Pictures to be sorted are presented by the visual picture stimulator as visual stimulations to a user, and the cortical electric signals of the user generated in response are recorded and subjected to amplification and A / D conversion, and then processed and analyzed by the face recognition specific wave N170 detector, whereby a determination is made whether the collected cortical electric signals contain specific wave N170 component related to facial picture stimulation, and accordingly, the cortical electric signals are transformed into picture sorting control commands for identifying facial pictures and object pictures. The advantages of the present invention are: the present invention utilizes the specific cortical electric signal N170 component generated in the facial picture recognition process in response to facial picture stimulations, and employs an effective online feature extraction and sorting algorithm in the cortical electric signal analyzer; therefore, the discrimination ratio of the system is increased. The present invention provides a novel means for persons who suffer from dyskinesia but can think normally to communicate with and control the external environment.

The invention provides a picture translation method and system. The method comprises the steps that a picture is obtained, and a target object is determined from the picture; the picture of the target object is sent to a picture recognition system so that the picture recognition system can recognize type information which the target object belongs to from the picture, and the type information is sent to the translation system so that the translation system can translate the type information; the type information and translations corresponding to the type information are received and displayed. According to the picture translation method and system, after the picture input by a user is obtained, the type information and the translations corresponding to the picture can be provided for the user at the same time, so the problem that the user can not know what a current object is when having a translation demand is solved, and the demand of the user for translation based on pictures is met.

The invention belongs to the technical field of image processing and provides a pornographyic picture identification method based on joint detection of skin colors and featured body parts, which comprises the following steps: (1) preparatory stage of sample training for acquiring needed parameters of the identification stage of an image to be detected, the preparatory stage of sample training comprises color skin sample training and featured body part sample training; (2) the stage of reading the image to be detected; (3) the identification stage of the image to be detected for judging whether the image to be detected is a pornographyic picture or not according to the parameters acquired at the preparatory stage of sample training: the skin color and the featured body parts are detected in sequence, and if the area of the skin color region accounts for more than a half of the whole image and at least one featured body part emerges, the picture is judged to be a pornographic picture. The method of the invention has the advantages of being capable of reducing false detection rate of pornographyic pictures, and having wide range of application and high accuracy rate.

A method for reading a picture is provided which is capable of removing adverse effects caused by lighting conditions or a like. A principal color extracting section extracts principal colors from a color image input by an image inputting section based on a local peak value shown in a color histogram obtained from the color image and divides the color image into principal color images each corresponding to each of the extracted principal colors. A combination creating section creates a combined image by combining separated principal color images. A picture region extracting section extracts character region candidates from the principal color image and combined image. A picture recognizing section performs character recognizing processing on all extracted character region candidates to obtain the best character reading result according to a character recognizing confidence of character recognizing processing results.

The objective of the invention provides a picture recognition method and device, a computer device, a computer readable storage medium and a computer program product. The picture recognition device positions the character region and the image region in a picture. Word vectors corresponding to the character contents in the character region are extracted; image characteristic vectors are extracted for the image region; and the word vectors and the image characteristic vectors are integrated, so semantic contents of the picture can be determined. Compared with the prior art, the invention provides a scheme for recognizing the picture through the computer device automatically, so content recognition and checking are performed before the picture including the character contents and the image contents is issued.

The invention discloses a rapid object recognition method for a mobile robot based on deep learning, and the method comprises the following steps: 1), the obtaining of a mobile image; 2), the preprocessing of image data; 3), the extracting of the features of the image; 4), image prediction output; 5), environment constrained optimization; 6), image recognition output. Through the unified integration of detection and recognition results, the method overcomes the shortcomings of the complexity and instability of a conventional object recognition system which needs to employ an object. Through a multilayer residual error network design and the generation of environment gravity constraint conditions, the method irons out the defect that an integrated object recognition system is poor in accuracy. The integration of detection and recognition tasks can guarantee the processing efficiency of a system, and improves the sensing capability of the robot in a moving process.

The invention provides an image recognition method and device, computer equipment, and a computer readable medium. The method comprises the steps: obtaining a to-be-recognized image; obtaining a prediction class label of the to-be-recognized image according to the to-be-recognized image and a pre-trained image recognition model, wherein the pre-trained image recognition model employs a convolution neural network model and the convolution neural network model comprises at least one nuclear pool layer which enables the channel dimension to be changed into many dimensions from one dimension; and recognizing the type of the to-be-recognized image according to the prediction class label of the to-be-recognized image. According to the technical scheme of the invention, the method can achieve the recognition of the type of an image with the finer granularity through employing the image recognition model with the nuclear pool layer when the method is used for the recognition of the to-be-recognized image, and effectively improves the accuracy and efficiency of image recognition.

The invention provides a picture recognition method and system. The method comprises the following steps of: obtaining a to-be-recognized picture; inputting the to-be-recognized picture into a pre-trained picture recognition model so as to obtain a predicted category label of the to-be-recognized picture, wherein the pre-trained picture recognition model adopts a convolutional neural network model, the convolutional neural network model comprises an input layer, a convolution layer, an attention branch, an element-wise operation layer, a pooling layer, a full connection layer and an output layer, the attention branch is used for determining a weight of each area of the picture according to global features of each area of the to-be-recognized picture, and the element-wise operation layer isused for weighting local features, of the picture, output by the convolution layer according to the weight of each area of the picture; and recognizing a category of the to-be-recognized picture according to the predicted category label of the to-be-recognized picture. According to the method and system, the problem that the efficiency of recognizing specific types of pictures such as pictures with too small main body areas or embedded / spliced pictures is relatively low is solved.

The invention discloses an intelligent garbage can capable of automatically identifying unknown garbage types and garbage sorting method. The intelligent garbage has the functions of barcode information verification, image recognition information verification, organic polymer sensor information recognition, garbage value conversion, and data uploading cloud. The garbage sorting method is simple insteps, mature in technology, and high in maneuverability, multiple methods are carried out together, the accuracy is high, moreover, the value generated by the garbage can be accumulated in a cloud account of the user through credits, the enthusiasm of people for garbage collection and recycling is improved, and the recovery utilization rate of the resources is improved.

The invention discloses a picture processing method and device. The picture processing method comprises acquiring the infrared picture of structured light from a target object, computing the infrared picture of the structured light, and synchronically obtaining the deep picture and infrared picture of the target object from the infrared picture of the structured light. Through the way mentioned above, the invention greatly increases the hit rate of recognition and improves the efficiency of picture recognition. Furthermore, the way of obtaining the deep picture and infrared picture from the infrared picture of the structured light can effectively reduce computed amount, accelerate picture processing speed, and decrease the power consumption of picture processing equipment and the whole cost, compared with prior art.

The invention, which belongs to the technical field of image processing, discloses a depth-learning-based Chinese character recognition system realizing method. Preprocessing, segmentation, identification, reorganization are carried out on a picture with characters to form a text, thereby realizing conversion from the picture to the text. The method comprises: (1), picture pretreatment is carried out; to be specific, graying, binarization, and inclination correction are carried out on an inputted original picture to obtain a regular picture; (2), picture segmentation is carried out; to be specific, segmentation is carried out on an inputted picture with lots of characters, the picture is segmented into rows and then each row is segmented into single character; (3), picture identification is carried out; to be specific, the segmented single character is identified and an identification module is invoked for each single character picture to obtain an identification character result; and (4), text reorganization is carried out; to be specific, the obtained single character identification results are combined according to a sequence, correction is carried out, and then an identification result text segment is produced. With the image processing technique, graying and binarization are completed; and the generalization ability and the anti-interference capability are high.

The invention provides a picture annotation data acquiring method and system. The method comprises the steps that the recognition result of a picture to be annotated is acquired; the picture to be annotated and the corresponding recognition result are displayed on an annotation interface; and the annotator's selection of the recognition result on the annotation interface is used to acquire the annotation data of the picture to be annotated. According to the picture annotation data acquiring method and system, an annotator simply needs to click the corresponding recognition result and does notneed to input a name manually; the annotation efficiency is improved; the method and system are especially suitable for the preliminary data preparation of an image vertical class recognition algorithm; the manual picture annotation cost can be greatly reduced; and the development cycle of picture recognition class items can be shortened.

The present invention discloses a method and an apparatus for identifying a pathological picture. The method comprises: obtaining sample data, wherein the sample data comprises a positive sample and anegative sample, the positive sample is a malignant lesion pathological picture, the negative sample is a normal or a benign lesion pathological picture, and a lesion area is marked on the malignantlesion pathological picture; dividing the sample data into a training set and a test set; training a deep neural network model by using the training set; testing the trained deep neural network modelby using the test set; performing parameter adjustment on the trained deep neural network model according to a test result; and identifying the pathological picture by using the trained deep neural network model. According to the method and apparatus disclosed by the present invention, the efficiency and accuracy of pathological picture identification can be improved.

The invention discloses a financial reimbursement all-invoice picture recognition and processing method. According to the method, a specific scene of invoice recognition is optimized to some extent, invoice recognition types are expanded, invoices of all types can be recognized, and recognition is more accurate and higher in efficiency. The method is mainly used for performing recognition and processing after the invoices of all types are scanned into pictures. The method comprises the specific steps that the invoice pictures obtained after scanning are subjected to color preprocessing, and input data is provided for picture contour preprocessing; the pictures obtained after color preprocessing are subjected to contour detection, and pixel interference outside an invoice paper range is eliminated; the pictures obtained after contour detection are subjected to text preprocessing, characters on the invoices are recognized; and a character set is formed and output; the output character set is subjected to picture recognition and processing, a recognition and processing result is organized into a result object with semantics according to semantics of invoice recognition, and the result object is used as a final recognition result.