2415results about How to "Fast recognition" patented technology

The invention discloses a system and method for processing an unmanned aerial vehicle polling image. The system comprises an unmanned aerial vehicle polling image processing system body which comprises a human-computer interaction module. After the human-computer interaction module obtains a user instruction, a main control panel analyzes the user instruction and transmits the analyzed instruction to a data receiving and transmitting controller, the data receiving and transmitting controller sends a parameter configuration instruction to corresponding peripheral equipment according to the instruction and transmits an instruction execution result back to the main control panel, and the main control panel processes the execution result of the parameter configuration instruction and then sends processed information to the human-computer interaction module so that the information can be displayed to a user. The main control panel is respectively communicated with a secondary controller, a power supply module, a wireless communication module, a data storage module, a locating module, a distance measuring meter and a height gauge. The system and the method improve the automation degree and the informationization degree of polling image processing, improve the efficiency of searching for and processing the polling result, and lower the cost of processing the polling result.

A method of recognizing characters in a document image comprises examining the intensity of pixels in the document image and identifying a peak intensity deemed to represent foreground in the document image. A threshold level for distinguishing the foreground from background in the document image as a function of the identified peak intensity is determined. The document image is thresholded using the threshold level to identify the foreground. Character recognition is performed on the foreground of the document image.

The embodiment of the invention discloses self-adapting face identification method and device relating to the field of face identification. The self-adapting face identification method comprises the following steps of: typing in the basic information of a face user; acquiring a plurality of face images of the face user and respectively extracting the face features from the face images to serve as a face feature template; saving the base information of the face user and the face feature template to a face user information base; extracting the face features from the acquired face images, and carrying out comparison identification with the face feature template in the face user information base; if the identification passes, further judging whether the face features meet the preset self-adapting demands or not; and if yes, updating the face feature template in the face user information base to complete the face feature self adapting. The method effectively eliminates the influence generated on the face identification effect because the face features slowly change with the changes of ages, beards and the like, thereby improving the identification speed and accuracy.

A real-time visual detection and identification method for high speed rail surface defects comprises the following steps of: (1) image acquisition; (2) image preprocessing; (3) defect preliminary detection, that is, performing logic or operation combination of detection results based on gray scale compensation with detection results based on top-hat operation, and detecting whether an abnormal area exists in an image, if not, finishing the detection, or else continuing the processing; (4) defect accurate positioning, that is, accurately positioning the defect by an algorithm of bonding single defect, an algorithm of filling holes in the defect area, and an algorithm of selecting the main defect, and extracting the defect area through marks; (5) defect classification, that is, extracting and selecting characteristics of the defect area, designing and training a BP neural network, and classifying the defects by the BP neural network. The invention has the advantages of simple principle, high automation degree, high detection speed, and high detection precision.

There is provided a method of providing Augmented Reality (AR) using the relationship between objects in a server that is accessible to at least one terminal through a wired / wireless communication network, including: recognizing a first object-of-interest from first object information received from the terminal; detecting identification information and AR information about related objects associated with the first object-of-interest, and storing the identification information and AR information about the related objects; recognizing, when receiving second object information from the terminal, a second object-of-interest using the identification information about the related objects; and detecting AR information corresponding to the second object-of-interest from the AR information about the related objects, and transmitting the detected AR information to the terminal.

The invention relates to an intelligent terminal network teaching method. The intelligent terminal network teaching method comprises the steps that communication connection among the interior of a teaching terminal, the interior of a learning terminal, the interior of a parent management terminal, the interior of a server and the interior of a service platform is set up; the teaching terminal is used for recording the course of a teacher and uploading and storing the course of the teacher in the server; the learning terminal feeds information back to the teaching terminal through the server, and the teaching terminal answers the fed back information correspondingly; the parent management terminal is used for maintaining parent information; data are stored in the server and used for being called by the teaching terminal, the learning terminal and the parent management terminal, and the server is used for carrying out data interaction with other external devices; the teaching terminal and the learning terminal both are in information interaction with the service platform, a teacher management module, a student management module and an intelligent search module are arranged in the service platform, and the intelligent search module comprises an image retrieval module and an image recognition module; the learning module and a sign-in management module carry out information transmission. The intelligent terminal network teaching method can be used for developing a pre-class lesson preparing application, a classroom teaching application, an after school care application, a remote education application and the like based on the Internet.

The invention discloses a method and a system for recognizing images, which relate to a method and a system for recognizing static target images by adopting the image recognition technology. The invention solves the problem that the recognition speed is relatively slower in the prior image recognition technology. The method and the system for recognizing the images are characterized in that image information in the identification area and characteristic information in the identification area are stored as template information, then the area to be recognized in the images to be recognized is confirmed by adopting the identification area, then the characteristic information in the identification area is compared with that in the area to be recognized, when the similarity of both characteristic information in the identification area and that in the area to be recognized is in the error range, users consider that the images to be recognized and the target images are mutually matched, so as to realize image recognition, in the recognition process, the characteristic information in the area to be recognized is just needed to compare, the data quantity is less, and the logical reasoning and the mathematical operation of the characteristic information are not required, thereby quickening the speed of image recognition. The invention is mainly used for searching matched images, for example, bill recognition, seal recognition and the like.

The invention relates to a method and system for recognizing a video program. The method comprises the steps that S1, video pictures played by a video play terminal are shot; S2, the overall or local feature information of video frames in the video pictures is extracted to generate digital fingerprints describing the video frames; S3, search matching is carried out on a video frame fingerprint database through the generated digital fingerprints, and program information corresponding to the video frames is acquired through a matching result. The video frame digital fingerprints are generated through the shot video pictures of the video play terminal, the search matching is carried out in the video frame fingerprint database to obtain the corresponding program information, and therefore recognition speed is high, the precision degree is high, and real-time requirements are met. Related information resources on a network are searched for and obtained through the recognized program information and are returned to a mobile communication terminal, a user can acquire the related information of a television program being watched by the user conveniently and fast when watching the video program, the operation is simple, real-time performance is strong, and the user experience effect is good.

The invention provides a multi-modality medical image identification method based on deep learning. The method includes the steps that on the basis of multi-modality medical images of a patient's part to be detected, a registration method is adopted to display the multi-modality medical images in the same three-dimensional space; on the basis of the multi-modality medical images, R-CNN is adopted to identify lesion areas in the multi-modality medical images; according to coordinates of the lesion areas in the multi-modality medical images, lesion bodies are displayed in the same three-dimensional space, and according to image blocks corresponding to diagnosed lesion areas, a dense sampling method and a CNN are adopted to obtain occurrence probabilities of various preset disease types. The invention provides a multi-modality medical image identification device based on the deep learning. The device includes a multi-modality medical image display module, a lesion area detection module, a lesion body display module and a preset disease-type occurrence probability module. According to the multi-modality medical image identification method and device, the automatic identification of the lesion areas in the medical images is achieved, and effective reference data is provided for further diagnosis by doctors.

The invention discloses a bone age mark identification assessment method and system based on deep learning and image omics. The bone age mark identification method includes the steps: performing preprocessing of window adjusting, alignment and standardization on the wrist bone image; using a bounding box to mark the bone age characteristic areas and mark the coordinates, wherein the bone age characteristic areas include a metacarphphalangeal group and a brachidium group according with a TW3 method; according to the requirement, performing augmentation processing, and inputting the wrist bone image data to a convolutional neural network of the area based on ResNet-101 to perform multi-task (positioning, classification and assessment) training at the same time; and based on the bone age characteristic areas, combining with the clinic information (demographic characteristics and inspection reports) to further train and improve the bone age assessment speed and accuracy. The bone age markidentification assessment method and system based on deep learning and image omics firstly utilize a small number of marked samples to perform preliminary training on the bone age model, and utilize the model with relatively higher positioning detection accuracy to automatically mark a large number of samples so as to realize automatic positioning, classification and bone age assessment of the bone age characteristic areas.

The invention discloses an Internet video face recognition method based on deep learning, comprising the following steps: (a) marking face data, acquiring an image with a face from the Internet, marking a face location box and the name of the person, and establishing a face image library; and (b) using the face data marked in step (a) to train a convolutional neural network, wherein the input of the convolutional neural network is set as the face location box, and the output of the convolutional neural network is set as a name tag and the confidence level of the name tag. The beneficial effects are as follows: in face track analysis, first, the quality of each face image on the track is evaluated to discard low-quality face images and retain only high-quality images in order to guarantee the reliability of recognition, and then, statistical analysis is made on the tags of face recognition results of single frames on the track, face image quality filtering is carried out, and the face tags of the overall track are decided based on the statistical parameters of the tags, which can effectively avoid the influence of quality of video on the accuracy of recognition.

The invention provides a security protection early-warning system and method based on radar detection and video image behavior analysis. The method comprises: S1, carrying out pairing and calibrationon a radar detecting device and a cloud platform camera; S2, monitoring whether a moving target exists in a monitoring range in real time by a radar and sending position information of all moving targets to an intrusion event detecting module; S3, determining positions of the moving targets relative to the cloud platform camera; S4, determining an adjustment angle of the cloud platform camera anda needed magnification value; S5, moving the visual field center of the cloud platform camera to the positions of the moving targets and controlling the cloud platform camera to carry out magnification; and S6, acquiring a current image of the cloud platform camera, carrying out a behavior analysis on the moving targets, determining a real intrusion target, and sending out alarm information. The security protection early-warning method has advantages of simple algorithm, fast recognition speed of the intrusion target, high recognition accuracy, and flexible equipment installation position arrangement and the like; the missed detection rate and the false alarm rate are reduced substantially; and the response speed of the system is increased.

The invention discloses a computer vision-based dynamic gesture recognition method, and aims at solving the gesture recognition problems under complicated backgrounds. The method is realized through the following steps of: acquiring a gesture data set and carrying out artificial labelling; clustering a labelled image set real frame to obtain a trained prior frame; constructing an end-to-end convolutional neural network which is capable of predicting a target position, a size and a category at the same time; training the network to obtain a weight; loading the weight to the network; inputting agesture image to carrying out recognition; processing an obtained position coordinate and category information via a non-maximum suppression method so as to obtain a final recognition result image; and recording recognition information in real time to obtain a dynamic gesture interpretation result. According to the method, the defect that hand detection and category recognition in gesture recognition are carried out in different steps in the prior art is overcome, the gesture recognition process is greatly simplified, the recognition correctness and speed are improved, the recognition systemrobustness is strengthened, and a dynamic gesture interpretation function is realized.

The invention discloses a partially covered face identification method based on data enhancement. By means of the artificial intelligence depth convolution neural network, the face alignment, facial detection, face feature extraction, and face matching are carried out for many times; through the principle of data enhancement, the pre-judgment of the covering type of the object covering a standard face, and the pre-judgment of the identity of the covered face, the two pre-judgment results are integrated to carry out the data enhancement, a covered face database is reconstructed by taking the standard face as the basis, and at the end, the original covered face and the reconstructed covered face are subjected to matching. The method solves the technical problems that the existing covered face identification method is easy to loss characteristic information due to the covering, is difficult to carry out the face identification, and low in identification accuracy, thereby achieving the technical effects of improving the face identification accuracy and efficiency.

A medical blood pressure transducer that provides an identifier to a monitor that conveys characteristics of the transducer. The monitor can use the information to decide whether to function, or in calibration of the system. The transducer may be part of a disposable blood pressure monitoring system, and may include two transducers closely-spaced to provide two separate but identical outputs. In this way, the transducer may be connected to both a patient monitor and a cardiac output monitor at the same time measurements from a single line can be simultaneously supplied to two separate monitoring devices (for example a patient monitor and a cardiac output monitor). The identifier for the transducer may be circuitry, specifically a resistance / capacitance (RC) combination that possesses a characteristic time constant.

The invention relates to a human-computer interaction system based on an electro-ocular signal, which is characterized in that an electro-ocular signal acquisition and amplification module acquires the electro-ocular signal through an electro-ocular signal sensor and amplifies the signal; an online electro-ocular signal processing module realizes the online mode identification of the signal from the electro-ocular signal acquisition and amplification module to judge the eye action of an operator in real time; and a controlled unit is used to realize the coding of an eye movement mode and the generation of a controlled command. The human-computer interaction system has strong applicability, high measurement precision, high identification speed and simple and convenient operation and realizes the special human-computer interaction based on the electro-ocular signal. The human-computer interaction system can help physically-disabled realize computer manipulation, text writing, webpage browsing and other operations like normal people or wireless control on other electronic equipment (such as home appliances) and other human-computer interactions; and simultaneously, the human-computer interaction system can be also used in occasions where a computer is inconvenient to operate by normal people with hands, such as a mine, rescue operation, a spacecraft, underwater and other severe or narrow environments.

The invention relates to a distributed semantic and sentence meaning characteristic fusion-based character relation extraction method, and belongs to the field of natural language processing. The method comprises the steps of firstly performing training in a small amount of marked corpora and a large amount of unmarked corpora by utilizing statistic word frequency features and a Bootstrapping algorithm to obtain a relational feature dictionary; secondly constructing a triple instance of a statement through an element distance optimization rule, and constructing a triple feature space by fusing distributed semantic information and semantic information; and finally performing true-false binary decision on a triple, and obtaining a character relation type by utilizing a confidence degree maximization rule. According to the method, automatic generation of the feature relation dictionary is realized; a conventional relational multi-class problem is converted into a triple true-false binary decision problem, so that a conventional machine learning classification algorithm is better adapted; and by utilizing the distributed semantic information, the accuracy of relational classification is improved.

The invention discloses a rapid authentication method based on handshape and palmprint recognition. The method comprises: collecting a hand image and separating handshape and palmprint features from the hand image, using the extracted handshape and palmprint features of a registered user as templates and storing into registration template libraries together with personal identifications of the user; setting a threshold value, and comparing the handshape feature extracted from the shot hand image with the handshape feature in the handshape feature template library; then determining whether identification names are the same, if yes, determining that the user is legal, otherwise, determining by recognizing the palmprint; setting a threshold value, and comparing the palmprint feature extracted from the shot hand image with the palmprint feature in the palmprint feature template library, if comparison of the palmprint features is successful, determining that whether the identification names are the same, if yes, determining that the user is legal, otherwise, determining that the user is counterfeit, and refusing the user. The rapid authentication method based on handshape and palmprint recognition solves the problems that the existing hand multi-feature fusion recognition technology is slow in recognition speed and low in recognition rate, and is rapid in recognition speed, accurate in authentication effect and good in practicability.

A system and method for fast synchronization of an incoming signal with a UWB receiver rapidly. The present invention synchronizes with a UWB receiver with an incoming signal. The present invention correlates a local pulse generated at the receiver with the incoming signal, finds a phase angle in the correlation function that would correspond to a high signal to noise ratio, thereby matching the receiver to the incoming signal phase, and operates the receiver at that phase. Exemplary options of fast synchronization include using multiple detection arms to compare one parameter of the correlation function to a predetermined threshold.

The invention relates to a method for identifying multi-class facial expressions at a high precision based on Haar-like features, which belongs to the technical field of computer science and graphic image process. Firstly, the high-accuracy face detection is achieved by using the Haar-like features and a series-wound face detection classifier; further, the feature selection is carried out on the high-dimension Haar-like feature by using the Ada Boost. MH algorithm; and finally, the expression classifier training is carried out by using the Random Forest algorithm to complete the expression identification. Compared with the prior art, the method can reduce training and identifying time while increasing the multi-class expression identification rate, and can implement the parallelization conveniently to increase the identification rate and meet the requirement of real-time processing and mobile computing. The method can identify the static image and the dynamic image at a high precision, is not only applicable to the desktop computer but also to the mobile computing platforms, such as cellphone, tablet personal computer and the like.

The invention provides a method for extracting video subtitle information, which comprises: preprocessing video images of which subtitles are required to be extracted; performing multi-dimensional subtitle region detection on the preprocessed video images, and obtaining rectangular blocks which represent the dimension and the position of subtitle regions; utilizing the rectangular blocks to split detected subtitle regions; enhancing the contrast of split subtitle regions; and adopting a K mean cluster method to realize extraction of the subtitles. The method has commonality on different language types, different fonts and different dimensions of the subtitles and different video program types, has the advantages of high accuracy of subtitle region detection and character recognition and high speed, does not need to perform model training under the condition of subtitle region detection, and reduces the possibility of influence of poor model training on prediction results.

The invention discloses a method for extracting crown word numbers from scanned banknote images in a banknote detection module. The method comprises the following steps that the green ray of an image sensor is used for banknote collecting to form source images, the boundaries, in the source images, of the banknote images are found by adopting the mode of scanning boundary points, and then the banknote images are turned over and corrected to the horizontal state; feature points are extracted, and face value and orientation of banknotes are identified according to the obtained feature points; crown word number areas are determined, binarization processing is adopted for the crown word number areas, the obtained binarization images of the crown word number areas are searched for the boundaries containing all crown word numbers, then each single crowd word number is segmented, and a support vector machine identifies and processes the segmented and normalized single word images. The method for extracting the crown word numbers from the scanned banknote images provides a powerful support for identifying the authenticity of banknotes, crown word numbers can be identified accurately, the ability to identify the authenticity of banknotes is greatly improved, and the property safety of the financial department is effectively protected.

A UWB communication system and method for fast synchronization of one transceiver with another using the incoming UWB signal, where synchronization is achieved in less than a full code wheel spin. An exemplary embodiment includes a UWB waveform correlator, a timing generator, and a controller wherein the controller examines the correlator outputs as the code-wheel spins, and generates control signals to cause the timing generator to stop and track the incoming UWB signal whenever the incoming signal is received with sufficient SNR to provide a predetermined quality of service such as bit-error rate (BER). This embodiment will in any case determine when the receiver has been substantially synchronized with an incoming signal, yet without an exhaustive search of the entire code-wheel.

The invention discloses a real-time positioning supervisory system of a prison. The system is a real-time positioning system based on an RFID technology. The system comprises an RFID radio frequency electronic tag, an RFID reader, an RFID antenna, an electronic map display screen and a background monitoring center. The RFID radio frequency electronic tag includes a prisoner tag and a prison police tag; and the RFID reader includes a fixed type reader and a hand-held reader. According to the system provided in the invention, an RFID antenna at a specific position is utilized to read electronic tag information including a signal intensity and arrival time and the like; a background monitoring center uses the obtained information and applies an RSSI algorithm or a TDOA algorithm to calculate a position of an electronic tag; and the position is displayed on an electronic map display screen. Therefore, the system enables real-time localization on positions of a prisoner and a prison police to be realized as well as processing can be carried out timely in case of an emergency, so that security of the prisoner and the prison police can be effectively ensured.

The invention relates to a face recognize system, comprising a network interface module, a dual-core chip SOC module connected with the network interface module, a CPLD address code, a DDR2 high-speed operation memory, and an external program memory, or even with a hard disk with an IDE. The dual-core chip module is connected with the high-speed operation memory via a data bus. The CPLD address code via an EMIF bus is connected with the external program memory NAND Flash or connected with the IDE hard disc too. And the whole system is powered by a power module. The recognize method comprises video code compression, pretreatment, motion check, face positioning, organ positioning, normalizing, character extraction, face comparison, and data communication. The invention has the advantages in high recognize rate and high recognize speed, strong data processing ability, better compatibility with prior detect network, stable operation, easy upgrade and maintenance, with low cost.

The invention provides a video-based vehicle overspeed monitoring method and a video-based vehicle overspeed monitoring system. The method comprises the following steps: firstly, performing standardization and enhancement on originally acquired images through image pretreatment; then, acquiring the vehicle speed of a dynamic target through dynamic target segmentation, the dynamic target tracking and the vehicle speed measurement, and judging whether the dynamic target is overspeed through the vehicle speed; and finally, finishing the identification of the plate number of the dynamic target and transmitting the data to a monitoring center through the plate number positioning and character recognition. The video-based vehicle overspeed monitoring system realizes full-automatic monitoring and networking, and has convenient erection, low cost and higher practical value. The video-based vehicle overspeed monitoring system has strong universality, strong openness, and strong extensibility. In addition, the video-based vehicle overspeed monitoring system is unnecessarily embedded with a wayside induction coil, adopts video to detect the overspeed state in a non-contact way, integrates the vehicle speed measurement and the plate number identification at the front end of the system, and has low requirement on the communication bandwidth.

A system and method for fast synchronization of an incoming signal with a UWB receiver. The present invention synchronizes a UWB receiver with an incoming signal. The present invention correlates a local pulse generated at the receiver with the incoming signal, finds the value in the correlation function that would correspond to a high signal to noise ratio, thereby matching the receiver to the incoming signal phase, and operates the receiver at that phase. Exemplary options of fast synchronization include a maximum peak detector.

The invention discloses a low-rank partitioning sparse representation human face identifying method, which adopts low-rank matrix decomposition, introduces a reference item, adopts a DCT (discrete cosine transform) algorithm to realize the normalization of images and effectively solves the problem of uneven lighting in a human face image. At a classifying stage, a clustering thought is used, and the identifying speed is effectively improved. The algorithm is used on a standard human face database to perform multiple times of tests, and the test result shows that compared with the existing human face identifying method, the low-rank partitioning sparse representation human dace identifying method has the advantage that the identifying accuracy and computing efficiency of the algorithm are all consistently improved. The precision and stability on human face identifying are improved under the complex conditions, such as shielding, disguising and illumination varying, of the human face image.