5847results about How to "Improve recognition rate" patented technology

A voice recognition device used as a peripheral device for a game machine including a voice input device, a voice recognition section for recognizing the player's voice by comparing the voice signal output from the voice input device with data from previously defined voice recognition dictionaries and generating control signals relating to the game on the basis of the recognition result. The voice recognition section includes a non-specific speaker voice recognition dictionary which is previously defined for unspecified speakers, and a specific speaker voice recognition dictionary which is defined by the player.

The present invention provides a combined lip reading and voice recognition multimodal interface system, which can issue a navigation operation instruction only by voice and lip movements, thus allowing a driver to look ahead during a navigation operation and reducing vehicle accidents related to navigation operations during driving. The combined lip reading and voice recognition multimodal interface system in accordance with the present invention includes: an audio voice input unit; a voice recognition unit; a voice recognition instruction and estimated probability output unit; a lip video image input unit; a lip reading unit; a lip reading recognition instruction output unit; and a voice recognition and lip reading recognition result combining unit that outputs the voice recognition instruction

A method comprises processing M subband communication signals and N target-cancelled signals in each subband with a set of beamformer coefficients to obtain an inverse target-cancelled covariance matrix of order N in each band; using a target absence signal to obtain an initial estimate of the noise power in a beamformer output signal averaged over recent frames with target absence in each subband; multiplying the initial noise estimate with a noise correction factor to obtain a refined estimate of the power of the beamformer output noise signal component in each subband; processing the refined estimate with the magnitude of the beamformer output to obtain a postfilter gain value in each subband; processing the beamformer output signal with the postfilter gain value to obtain a postfilter output signal in each subband; and processing the postfilter output subband signals to obtain an enhanced beamformed output signal.

The invention discloses a display device. The device comprises a display panel with a plurality of pixel units, a protective cover plate arranged on a light-out surface of the display panel, a plurality of photosensitive devices arranged at one side of the pixel unit opposite to the light-out surface and used for line recognition, an optical collimator arranged between the protective cover plate and the plurality of photosensitive devices is added, the optical collimator is provided with a transmittance area on the top of the photosensitive devices, the remaining area except for the transmittance area is a shading area, part of stray light adjacent to a ridge and display luminescence given out by the pixel units and entering the photosensitive devices can be filtered by the shading area, the transmittance area can ensure that the ridge reflected light enters the corresponding photosensitive devices, the line recognition rate of the photosensitive devices is improved, and therefore the precision and definition of line recognition images are improved.

The invention discloses an image classification method based on a convolution neural network. The method comprises the following steps: constructing a deep convolution neural network; improving the deep convolution neural network; training and testing the deep convolution neural network; and optimizing the network parameter. By using the image classification method disclosed by the invention, the improvement and the optimization are respectively performed on the network structure and multiple parameters of the convolution neural network, the recognition rate of the deep convolution neural network can be effectively improved, and the accuracy of the image classification is improved.

A system and method for correcting data in data fields of financial documents containing unreadable characters is described.

A method for automatically recognizing one or more structures in digitized image date includes determining sub-jets of the image graph at the nodes defined by its structure. The sub-jets correspond to at least part of the determined sub-jets of the specific reference graph. The projection of the net-like structure of the specific reference varies until a graph comparison function, which compares the jets of the image graph with the corresponding jets of the specific graph, becomes optimal. The structure is associated with a reference image corresponding to the reference graph for which the graph comparison function is optimal, with respect to the optimal image graph determined for the reference graph.

This invention provides an infrared illuminator and camera system for imaging of auto vehicle license plates. The system works in ambient light conditions, ranging from bright sunlight, to dim light, to dark, to zero ambient light. It yields high-contrast imaging of the letters and numbers on retro-reflective license plates. The images of the license letter and number combinations can be read manually by a remote operator. They can be converted to text format with optical character recognition computer hardware and software. The text data can then be compared to data files listing license numbers to provide further data about the owner of a licensed vehicle. A decision can be made quickly about whether to allow a vehicle to proceed through a gate, or whether to take other action. The system uses a mono camera that is enhanced for infrared sensitivity and combined with a high power infrared illuminator to maximize range at night, and with shutter speeds set up to capture clear license plate pictures even with fast moving vehicles and even with their headlights on and interfering with human observation of the license plates. Optical filtering to pass infrared in the range of the illuminator and to reduce light outside this range, combines with a lens set up, to avoid vertical smear and sensor overload caused by headlights at night and by highlight reflected glare from the sun in daytime.

A user can automatically change a user interface of a portable terminal into a user interface of an electronic appliance suitable for user's intention. A portable terminal recognizes a circumferentially existing electronic appliance based on a photographic image or a radio signal, and allows one or more applications having a user interface varying by electronic appliance to start and then be resident in a memory of the portable terminal. Then, when the portable terminal recognizes a predetermined electronic appliance, the portable terminal changes a user interface displayed on a display unit and an input unit of the portable terminal into a user interface of an application associated with the predetermined electronic appliance to enable a user to view the user interface. When a user puts the portable terminal to a predetermined direction, the portable terminal recognizes an electronic appliance existing in the direction and displays the corresponding user interface.

The invention discloses a train operation fault automatic detection system and method based on binocular stereoscopic vision, and the method comprises the steps: collecting left and right camera images of different parts of a train based on a binocular stereoscopic vision sensor; achieving the synchronous precise positioning of various types of target regions where faults are liable to happen based on the deep learning theory of a multi-layer convolution neural network or a conventional machine learning method through combining with the left and right image consistency fault (no-fault) constraint of the same part; carrying out the preliminary fault classification and recognition of a positioning region; achieving the synchronous precise positioning of multiple parts in a non-fault region through combining with the priori information of the number of parts in the target regions; carrying out the feature point matching of the left and right images of the same part through employing the technology of binocular stereoscopic vision, achieving the three-dimensional reconstruction, calculating a key size, and carrying out the quantitative description of fine faults and gradually changing hidden faults, such as loosening or playing. The method achieves the synchronous precise detection of the deformation, displacement and falling faults of all big parts of the train, or carries out the three-dimensional quantitative description of the fine and gradually changing hidden troubles, and is more complete, timely and accurate.

A mobile phone which is an example of electronic equipment includes an infrared camera and an infrared LED. The infrared camera is arranged above a display and the infrared LED is arranged below the display. A user, by an eye-controlled input, designates a button image or a predetermined region on a screen. When a line of sight is to be detected, an infrared ray (infrared light) emitted from the infrared LED arranged below the display is irradiated to a lower portion of a pupil. Accordingly, even in a state that the user slightly closes his / her eyelid, the pupil and a reflected light of the infrared light can be imaged.

A method of changing a display mode in a car navigation system, the method including: acquiring distance information from a current vehicle location to a guide point; and gradually changing the display mode into any one of a two-dimensional display mode and a three-dimensional display mode based on the distance information.

The invention claims an electroencephalogram recognizing method combing a convolutional neural network (CNN) with a long and short time memory network (LSTM). The method comprises the following steps:firstly, acquiring electroencephalogram signal data by using an Emotive acquiring instrument, and carrying out pretreatment such as mean removal, filtering and normalization on the acquired electroencephalogram signals; secondly, inputting pretreated data into a convolution layer and a pooling layer to extract space features; and finally, directly connecting the rear of the pooling layer to LSTM,extracting temporal order information of electroencephalogram data, and finishing a classifying task through Dropout and a fully connected layer. The temporal and spatial features of electroencephalogram signals can be fully utilized, the space and temporal order information of the electroencephalogram data are extracted, thus, the classifying accuracy of the electroencephalogram signals is improved, and a new way is provided for research on electroencephalogram recognition.

A method for extracting the venous characteristics from human hand back and finding a match to identify a person includes such steps as obtaining clean venous image by special acquisition apparatus, normalizing its size and grayscale, intensifying, dividing based on threshold, removing noise, fining image, extracting venous characteristics by one of three approaches, classifying by classifier, and finding match by decision fusion.

The present invention relates to a mobile terminal and a method for controlling the same, the mobile terminal comprising: a camera; a display unit for displaying an image inputted through the camera; and a control unit which performs a user authentication on the basis of a received first facial image when the first facial image including facial features necessary for the user authentication is received through the camera, and which performs a user authentication by using at least one facial feature included in a received second facial image when the second facial image which lacks a part of the facial features is received.

A system and method for correcting data in data fields of financial documents containing unreadable characters is described. Data fields include MICR and OCR data files on financial documents such as checks. A controller receives MICR or OCR data from a document processor that is operable to retrieve MICR or OCR data from a plurality of financial documents, and performs data correction functions on the MICR or OCR data. Data corrections functions include comparing an erroneous number in the MICR or OCR data with a plurality of correct numbers and electronically replacing the erroneous number with a number from said plurality of possible numbers.

The invention discloses a character segmentation method and a character segmentation device, which can recognize character unit image blocks containing touching characters and character unit image blocks containing components and radicals, and assure the correctness of the character segmentation result. In the technical proposal of the invention, a plurality of character unit image blocks are obtained by making line segmentation and column segmentation to a text image, character unit image blocks containing touching characters are recognized and continue to be segmented, Chinese character unit image block areas and English character unit image block areas are recognized, character unit image blocks occupied by components and radicals of Chinese characters are recognized in the Chinese character unit image block areas, and character unit image blocks occupied by components and radicals of adjacent Chinese characters are merged into a character unit image block. The invention ensures that the character segmentation result does not depend too much on a character recognition feedback mechanism and further improves the recognition rate of the characters.

The invention provides a vehicle license plate recognition method based on a video. According to the vehicle license plate recognition method based on the video, moving vehicles are detected and separated out with the vehicle video which is obtained through actual photographing by means of a camera serving as input, the accurate position of a vehicle license plate area is determined by conducting vertical edge extraction on a target vehicle image obtained after pre-processing, a vehicle license plate image is separated out, color correction, binaryzation and inclination correction are conducted on a vehicle license plate image, each character in the positioned vehicle license plate area is separated to serve as an independent character, feature extraction is conducted one each character, obtained feature vectors are classified through a classifier which is well trained in advance, a classification result serves as a preliminary recognition result, secondary recognition is conducted on the stained vehicle license plate characters according to a template matching algorithm imitating the visual characteristics of human eyes, and then a final vehicle license plate recognition result is obtained. The vehicle license plate recognition method based on the video has the advantages that hardware cost is reduced, the management efficiency of an intelligent transportation system is improved, the anti-jamming performance and the robustness are high, the recognition efficiency is high, and the recognition speed is high.

The present invention discloses a generative adversarial network-based multi-pose face generation method. According to the generative adversarial network-based multi-pose face generation method, in a training phase, the face data of various poses are collected; two deep neural networks G and D are trained on the basis of a generative adversarial network; and after training is completed, the generative network G is inputted on the basis of random sampling and pose control parameters, so that face images of various poses can be obtained. With the method of the invention adopted, a large quantity of different face images of a plurality of poses can be generated, and the problem of data shortage in the multi-pose face recognition field can be solved; the newly generated face images of various poses are adopted as training data to train an encoder for extracting the identity information of the images; in a final testing process, an image of a random pose is inputted, and identity information features are obtained through the trained encoder; and the face images of various poses of the same person are obtained through the trained generative network.

The invention discloses a psychological stress assessment method based on multi-physiological-parameter integration. The method includes: designing a reasonable stimulation program, acquiring four types of electrophysiological signals, namely electrocardiogram signals, electromyographic signals, pulse wave signals and electroencephalogram signals from people suffering psychological stress; extracting affective features of the four types of electrophysiological signals; subjecting the extracted features to feature selection by means of Relief algorithm, genetic algorithm optimization and the like; acquiring related integration functions on the basis of basic probability assignment mass. According to the method, multi-parameter signals are subjected to acquisition, preprocessed, feature selection and psychological stress affective recognition and are integrated; compared to single-parameter classified recognition or multi-parameter data-level or feature-level integration, the method allows data information to be more fully utilized and psychological stress emotions to be more accurately recognized.

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 discloses a method for recognizing voice. The method comprises the following steps of: acquiring audio data; acquiring a Lattice result of the audio data, wherein the Lattice result comprises time point information, a plurality of pieces of candidate information and matching likelihood scoring information; acquiring confidence scoring information according to the plurality of piecesof candidate information and the matching likelihood scoring information; rearranging the plurality of pieces of candidate information by using a stronger voice model and providing the optimal recognition result; positioning a voicing position corresponding to the audio data and simultaneously displaying other candidate words; selecting or inputting a correct text to finish amendment and freezingthe amended text; and searching a related text training language model by using a search engine according to the amended text serving as a key word, interpolating to acquire an adaptive language model, and returning and newly recognizing the rest part of audio data by using the adaptive voice model. By using the technical scheme, the voice recognition rate can be improved, and the workload of manual checking can be reduced.

The invention relates to a face recognition method based on Gabor wavelet transform and local binary pattern (LBP) optimization. Two-dimensional Gabor wavelet transform can associate pixels of adjacent areas so as to reflect the change conditions of image pixel gray values in a local range from different frequency scales and directions. The feature extraction and the classification recognition are carried out on the basis of a face image two-dimensional Gabor wavelet transform coefficient. For a high-dimensional Gabor wavelet transform coefficient, overall histogram features are extracted by adopting the LBP, and then the image is blocked by utilizing priori knowledge to extract the features of each piece of LBP local histogram. The method has better recognition rate, better robustness to illumination and wide using prospect in the fields of biometric recognition and public security monitoring.

The invention discloses a binocular visible light camera and thermal infrared camera-based target identification method. The method comprises the steps of calibrating internal and external parametersof two cameras of a binocular visible light camera through a position relationship between an image collected by the binocular visible light camera and a pseudo-random array stereoscopic target in a world coordinate system, and obtaining a rotation and translation matrix position relationship, between world coordinate systems, of the two cameras; according to an image collected by a thermal infrared camera, calibrating internal and external parameters of the thermal infrared camera; calibrating a position relationship between the binocular visible light camera and the thermal infrared camera;performing binocular stereoscopic visual matching on the images collected by the two cameras of the binocular visible light camera by adopting a sift feature detection algorithm, and calculating a visible light binocular three-dimensional point cloud according to a matching result; performing information fusion on temperature information of the thermal infrared camera and the three-dimensional point cloud of the binocular visible light camera; and inputting an information fusion result to a trained deep neural network for performing target identification.

The system comprises: a terminal with the human-face image detecting function and the voice data acquiring function, a communication network and a personal identification server. Wherein, the terminal with the human-face image detecting function and the voice data acquiring function embeds the acquired voice data into the detected human-face image, and sends the human-face image embedded with the voice data to the personal identification server; the identification authentication server separates the voice data from the human-face image, and respectively authenticate the human-face image and the voice data, and then combines them; the authentication result is sent to the terminal.

The invention discloses a coal-rock interface identifying method and system based on an image. The method comprises the following steps of: acquiring multiple color images of coal and rock on a coal mining working face; extracting a vector based on an image characteristic serving as a sample characteristic vector specific to each color image to obtain a known sample set of coal and rock; and establishing a coal-rock classifier model by adopting a Fisher linear judging method and taking the known sample set of the coal and rock as a training sample set. In the working process of a coal mining machine, a color image of the coal and rock which is cut by using a drum is acquired in real time, and the extracted characteristic vector is input into the coal-rock classifier model to identify a coal-rock type. The system consists of a light source module, an imaging module, a processing module and an anti-explosion shell. The coal-rock interface identifying method and the system provided by the invention have the characteristics of simple structure, easiness for distributing, high suitability and the like, the coal-rock type cut by using the drum can be automatically identified in real time, and reliable coal-rock interface information is provided for automatic heightening of the drum of the coal mining machine.

The embodiment of the invention provides a face recognition method and device. The face recognition method comprises the following steps: detecting whether a video streaming image contains the information of the local characteristics of a face, determining the face contained in the video streaming image according to a detection result, determining the face which is effectively recognized, and carrying out living body detection recognition; when the face which is effectively recognized meets a living body detection recognition condition, extracting a single frame of image or picture of the video streaming image, and generating an individual characteristic head portrait; according to the individual characteristic head portrait, extracting individual characteristic information; comparing a similarity between the individual characteristic information with sample plate characteristic information in an individual characteristic library; and starting a relevant application program when the similarity is greater than a preset similarity threshold value. Correspondingly, the embodiment of the invention also provides a face recognition device. According to the technical scheme provided by the embodiment of the invention, influence on a face recognition operation by external environment can be favorably eliminated, accuracy is higher when the identity of a user is judged through the face, and a recognition rate is improved.

The invention discloses a system and a method for smiling face recognition in a video sequence. The system comprises a pre-processing module, a feature extraction module, and a classification recognition module. According to the pre-processing module, through video collection, face detection and mouth detection, a face image region capable of directly extracting optical flow features or PHOG features can be acquired; according to the feature extraction module, Optical-PHOG algorithm is adopted to extract smiling face features, and information most facilitating smiling face recognition is obtained; and according to the classification recognition module, random forest algorithm is adopted, and classification standards on a smiling face type and a non-smiling face type are obtained according to feature vectors of a large number of training samples obtained by the feature extraction module in a machine learning method. Comparison or matching or other operation is carried out between feature vectors of a to-be-recognized image and the classifier, and the smiling face type or the non-smiling face type to which the to-be-recognized image belongs can be recognized, and the purpose of classification recognition can be achieved. Thus, according to the system and the method for smiling face recognition in the video sequence, accuracy of smiling face recognition can be improved.

The invention relates to a face feature extraction method based on a face feature point shape drive depth model. The method comprises the steps that the face feature point shape drive depth model is set up, N depth convolution neural networks are utilized for extracting features of N face regions divided according to the positions of face feature points to obtain the discrimination feature and the attributive feature of each region, and then all the discrimination features and the attributive features are fused to obtain features higher in descriptive ability. According to the face feature extraction method based on the face feature point shape drive depth model, the problem of robustness under change conditions of illumination, angles, expressions, shielding and the like can be well solved, and the recognition rate of face recognition under these conditions is increased.

The invention provides a multi-task deep learning network-based training method, a multi-task deep learning network-based training system, a multi-task deep learning network-based identification method and a multi-task deep learning network-based identification system. The training method includes the following steps that: the face region of a face image in a training set is obtained; key point detection is performed on the face region, so that key feature point positions are obtained; affine transformation is performed on the face image according to the key feature positions, so that an aligned face image can be obtained; and the aligned face image is inputted into a multi-task deep learning network, so that training can be carried out, and therefore, a multi-task deep learning network model can be obtained. The identification method includes the following steps that: affine transformation is performed on a face image to be identified according to the key feature positions of the face image to be identified, so that an aligned face image can be obtained; the aligned face image is inputted into a trained multi-task deep learning network model, so that feature extraction can be carried out, and feature information can be obtained; and the feature information of the face image to be identified is matched with feature information corresponding to each face image in a registration set, so that identification results can be obtained. With the methods and systems adopted, the training and identification efficiency of the multi-task deep learning network can be improved.