193 results about "Phase retrieval" patented technology

Fast retrieval and progressive retransmission of content, including the steps of obtaining erasure-coded fragments by an assembling server from fractional-storage CDN servers; reconstructing a streaming content file from the obtained fragments; and streaming the reconstructed content or a transcoded version of the content by the assembling server to a client device. Wherein most of the fractional-storage CDN servers are located close to or on the Internet backbone, and streaming the reconstructed content takes at least an order of magnitude longer than obtaining the fragments and reconstructing the content.

A novel phase-coded aperture, associated imaging system, and design method is disclosed. The optical imaging system includes a coded-aperture followed optically by a detector array and includes an image processor. A diffraction pattern in the form of a band-limited uniformly redundant array is formed on the detector array when focusable radiation from a point source in object space is modulation by the transmission function of the coded-aperture. Since diffraction effects cannot be ignored in the optical regime, an iterative phase retrieval method is used to calculate the phase-coded aperture transmission function. Correlation type processing can be applied for the image recovery.

A phase diversity wavefront sensor includes an optical system including at least one optical element for receiving a light beam; a diffractive optical element having a diffractive pattern defining a filter function, the diffractive optical element being arranged to produce, in conjunction with the optical system, images from the light beam associated with at least two diffraction orders; and a detector for detecting the images and outputting image data corresponding to the detected images. In one embodiment, the optical system, diffractive optical element, and detector are arranged to provide telecentric, pupil plane images of the light beam. A processor receives the image data from the detector, and executes a Gerchberg-Saxton phase retrieval algorithm to measure the wavefront of the light beam.

The invention discloses a self-correlation optical fiber communication system based on mode division multiplexing. The self-correlation optical fiber communication system based on mode division multiplexing comprises an optical carrier input unit, optical signal modulation units, a wavelength division multiplexing unit, a mode division multiplexing and demultiplexing unit, a wavelength division multiplexing unit and coherent reception units. The mode division multiplexing and reconciliation multiplexing unit comprises a mode multiplexer and mode demultiplexer, and the mode multiplexer and the mode demultiplexer are connected through few-mode optical fibers. The optical carrier input unit is connected with each optical signal modulation unit, the wavelength division multiplexing unit and the mode multiplexer through single-mode optical fibers. The mode multiplexer is connected with the mode demultiplexer through few-mode optical fibers. The mode demultiplexer is connected with the wavelength division multiplexing unit and each coherent reception unit through single-mode optical fibers. The optical carrier input unit is further connected with the wavelength division multiplexing unit. The self-correlation optical fiber communication system based on mode division multiplexing eliminates an expensive narrow line width adjustable local oscillator light source which is arranged at a reception end, enables a laser to be conveniently managed and maintained, is free from using frequency offset estimation in a digital signal processor (DSP) and phase retrieval algorithm, reduces complexity of the DSP and has the advantages of being high in spectrum effectiveness and big in nonlinearity tolerance.

An X-ray imaging apparatus includes a phase grating, an absorption grating, a detector, and an arithmetic unit. The arithmetic unit executes a Fourier transform step of performing Fourier transform for an intensity distribution of a Moiré acquired by the detector, and acquiring a spatial frequency spectrum. Also, the arithmetic unit executes a phase retrieval step of separating a spectrum corresponding to a carrier frequency from a spatial frequency spectrum acquired in the Fourier transform step, performing inverse Fourier transform for the separated spectrum, and acquiring a differential phase image.

The invention discloses a generation method for a structured query sentence used for the precise retrieval of a natural language. The generation method comprises the following steps: obtaining a natural language query sentence; carrying out word segmentation on the natural language query sentence by a preset dictionary and a first rule to obtain a word segmentation scheme; generating an analytic tree on the word segmentation scheme by a preset second rule; according to the root node of the analytic tree, generating a local query parameter, and substituting the global query parameter of a first layer of subnodes of the analytic tree, and obtaining the global query parameter of the root node of the analytic tree; according to the global query parameter of the root node of the analytic tree, generating the structured query sentence. The invention also provides a generation system for the structured query sentence used for the precise retrieval of the natural language. According to the generation method and the generation system, the natural language query sentence can be automatically converted into the structured query sentence.

The invention relates to a single channel mixed speech time domain separation method which is the single channel mixed speech time domain separation method based on a Convolutional Neural Network. The method comprises the following steps of 1, constructing a speech data set for training, 2, preprocessing speech data, 3, obtaining hybrid speech data, 4, constructing a neural network structure, 5, using data organized to train a neural network in a monitoring mode, and 6, using the trained neural network to carry out a separation test. According to the single channel mixed speech time domain separation method based on the Convolutional Neural Network, time domain speech signals serve as input and output of the Convolutional Neural Network, single channel hybrid speeches are separated, and therefore estimation for two source signals is obtained. The method does not need to deal with the problem of phase retrieval, and the separation quality of a single channel speech is improved.

A method of retrieving phase information from input intensity information, representative of a target image, in which a Fourier transform is performed on data and the result used in forming a phase estimate, the phase estimate being inverse Fourier transformed, thereby producing magnitude and phase replay, and wherein not only is the phase reply component but also data derived from the magnitude replay component, iteratively fed back.

A phase retrieval method for differential phase contrast imaging includes receiving data corresponding to a differential phase image generated from a measured signal. The measured signal corresponds to an X-ray signal detected by a detector after passing through a subject located with a grating arrangement between an X-ray source and the detector. The method further includes generating a phase image corresponding to the integration of the differential phase image. Generating the phase image includes performing an iterative total variation regularized integration in the Fourier domain.

A single-channel color image encryption method based on chaos and phase retrieval process comprises a color image chaos scrambling and decomposition step, a pure phase extraction and modulation step and a chaos diffusion step. Combining fractional Fourier transform and paired Logistic chaotic mapping, an encrypted image has nonlinearity and randomness, security and convergence speed of an encryption system are raised, and the problem of slow convergence speed existing in a present encryption method is solved. Meanwhile, complexity of the color image encryption system is reduced, the system is simple, and operation is convenient.

The invention provides a programmable optical component based on a scattering medium, an optical field regulation system, and an optical field regulation method, and solves technical problems that a conventional optical component has single function, is high in optical wave modulation cost, has complex apparatus and low integration degree. The system in the invention is successively provided with a light source beam expanding module, a modulation module, a scattering medium microscopic module and a detection module, a data processing module arranged between the modulation module and the detection module. A computer circularly controls a space light modulator to modulate an input optical field, and a detector detects an output optical field; then data shaping is carried out and an optical transmission matrix of the scattering medium is detected; and then a programmable optical component is constructed so as to achieve optical field regulation with methods such as optical phase conjugation, phase retrieval, speckle reconstruction and the like, wherein the optical field regulation includes imaging, chromatic dispersion, abnormal chromatic dispersion, focusing, shaping, etc. The programmable optical component and the system are intelligent, multi-functional, low-cost, easy to operate, highly integrated and strong in operability, and can be applied to the scientific research fields such as military science, bio-medicines, integrated optic, nano photonics, etc.

Embodiments of methods and apparatus are disclosed for obtaining differential phase contrast imaging system and methods for same. Method and apparatus embodiments can provide regularized phase contrast retrieval that can address noise reduction and / or edge enhancement. Certain exemplary embodiments can suppress stripe artifacts occurring in the process of integration of noisy differential phase data. Further, certain exemplary embodiments can use transmission images and / or dark-field images to improve or restore phase contrast images affected by noise edges.

The invention relates to a method for detecting the surface shape of a large-aperture telescope by using phase diversity phase retrieval. The method comprises the following steps of: representing a 2-bit variable by using a 1-bit variable in order to representing an imaging relation of an optical system; expanding a wave aberration into a series, solving a transfer function by using self-correlation operation of a pupil function and changing a Zernike coefficient so as to obtain different transfer function calculating results, comparing the transfer function calculating results with a transfer function result acquired from the Fourier transform of a measured point spread function, using a minimum mean square error result as a solution; and solving an evaluation value E. The method for detecting the surface shape of a spherical mirror on the basis of phase diversity phase retrieval in a focal plane detector is low in development cost, simple in hardware, susceptible of influences of environment (especially vibration), and capable of dynamically detecting optical elements and systems.

A system for detecting piston diversity between mirror segments. The system includes a pupil plane mask, a transform optical element, and an image detector. The pupil plane mask includes two or more open mask areas and two or more polarizers. Each polarizer is disposed within a respective one of the open mask areas. A first one of the two or more polarizers has a first polarization orientation, and a second of the two or more polarizers has a second polarization orientation.

Spatial mode conversion modules are described, with the capability of efficiently transforming a given optical beam profile, at one plane in space into another well-defined optical beam profile at a different plane in space, whose detailed spatial features and symmetry properties can, in general, differ significantly. The modules are comprised of passive, high-efficiency, low-loss diffractive optical elements, combined with Fourier transform optics. Design rules are described that employ phase retrieval techniques and associated algorithms to determine the necessary profiles of the diffractive optical components. System augmentations are described that utilize real-time adaptive optical techniques for enhanced performance as well as power scaling.

The invention discloses a phase retrieval based 4f mirror surface detection imaging system and a phase retrieval based 4f mirror surface detection imaging method. The system comprises a laser, a neutral density filter, a microobjective, a pinhole, a measured mirror surface, a 4f imaging unit and a computer, wherein the 4f imaging unit comprises a lens 1, a space light modulator, a lens 2 and a charge coupled device (CCD) camera. The light emitted by the laser irradiates the measured mirror surface after passing through the neutral density filter, the microobjective and the pinhole. The CCD camera arranged in the 4f imaging unit is used for acquiring a plurality of times of a light wave modulation image, and then the image is sent into the computer for sparse constraint phase recovery treatment. Based on the acquired light wave intensity image of the measured mirror surface, the method utilizes the sparse constraint phase recovery treatment to obtain the phase position of the light wave on the measured mirror surface, thus realizing the error detection for the measured mirror surface. The invention has the advantages of being high in accuracy, good in stability, simple in operation and good in noise robustness.

The invention discloses a multi-target evaluation optimization method based on a group decision making retrieval strategy and a multi-target evaluation structure for the optimization calculation of set values. The method comprises the following steps of: defining an encoding function; randomly assigning weights for initialization; performing case retrieval after decoding; calculating individual adaptability; judging whether the accuracy meets requirements or not (outputting optimized weights if the accuracy meets the requirements, otherwise, performing the case retrieval again); obtaining final weights by adopting group decision making; calculating case similarity to obtain new retrieval results; judging whether new cases to be retrieved exist or not; calculating product quality, product yield and an energy consumption index within a unit time; evaluating the set values; judging whether to perform compensation and correction on the set values or not; and outputting final set values. The method and the structure are applied to control over the roasting process of a shaft furnace in the metallurgical industry; the rationality of the set values can be evaluated; the quality of roasted ore is improved by 2 percent, the yield of the roasted ore is improved by 4 percent, and energy consumption is reduced by 5 percent; and optimal target values of production indexes are realized, the quality of the roasted ore is improved, gas consumption in a production process is reduced, and the aims of energy saving and consumption reduction are fulfilled.

Provided is an intelligent bad video detection method based on face retrieval. The method involves three parts, and the three parts comprise six modules in total. The video face extraction module M1 is used for extracting faces emerging in a video sample. The face gender classification module M2 is used for conducting gender classification on results obtained by the module M1. The face feature extraction module M3 is used for conducting feature extraction on obtained female faces to generate an inverted list. The database building module M4 is used for building a retrieval database for querying through output results of the module M3. The database updating module M5 is used for registering newly-found samples which are labeled as bad video and are not recorded in the database and adding registering results into the retrieval database. The face retrieval module M6 is used for comparing faces in the sample to be detected with faces in the retrieval database. The intelligent bad video detection method based on face retrieval can be used for conducting intelligent identification on video samples including common video and bad video, manpower and time are greatly saved, and meanwhile the detection accuracy is improved.

A real-time FROG system provides ultra fast pulse measurement and characterization. The system includes direct, integrated feedback that measures how well the system is retrieving pulses and tracking changes in the pulse train. This feedback is provided in real time and may be in the form of the FROG trace error, the display of the measured and retrieved FROG trace, accuracy of background subtraction or other quality measurement. The system includes preprocessing options that can be used to adjust the dynamic range of the measured signal or to perform different types of filtering. The preprocessing of the FROG trace precedes phase retrieval processing and improves the quality of pulse retrieval.

The invention provides an identity uniformity check method and device based on a spectrogram and phoneme retrieval. The method includes the following steps: acquiring a spectrogram corresponding to asample audio file; acquiring the speech feature parameters of the sample audio file; constructing a phoneme recognition model, inputting the speech feature parameters to the phoneme recognition modeland carrying out phoneme retrieval to get qualified phonemes; and marking the qualified phonemes on the spectrogram, checking the uniformity of vowels or vowel combinations with the same identifier, and judging whether a to-be-identified person corresponding to the sample audio file passes identity verification. The technical problem on phoneme searching and finding in actual voiceprint authentication is solved. Phonemes can be displayed visually. The identification efficiency of investigators is improved.

The invention relates to a face retrieval system and method based on deep learning. The method includes the steps: retrieval model training: training a model by a caffe deep learning framework and optimizing model parameters in a gradient descent mode to obtain a retrieval model; face feature extraction: feeding pictures of all people in a training set into a neural network for a feed-forward process and storing output of a first fully connected layer and a second fully connected layer; face registration: sufficiently extracting feature vectors and Hash codes of registrants under different illumination and different postures; face retrieval: extracting Hash codes and feature vectors of face images to be retrieved, comparing the Hash codes and the feature vectors with the pictures in the training set. The most similar face images are obtained by combining coarse retrieval of the Hash codes and fine retrieval of the feature vectors, and corresponding similarity values are outputted.

The invention discloses a single focal plane high-precision testing method for optical wavefront of an optical imaging system, relates to the technical field of optical testing, solves the problems that exit pupil amplitudes are not distributed uniformly and calculation errors are introduced by fast Fourier transform in the conventional phase retrieval algorithm, and provides the scheme for eliminating the influence of vibration in the process of image acquisition on detection accuracy. The method comprises the following steps of: establishing a detection platform of the optical imaging system; detecting the position of the focal plane of a lens to be detected by using a detection device in the detection platform and acquiring an out-of-focus stellar image of the lens to be detected by the detection device; selecting effective data according to the acquired out-of-focus stellar image and calculating a pupil function of an optical system; and extracting the phase of the acquired pupil function to obtain the optical wavefront of the optical imaging system. The pupil function of the optical system is calculated by a Zernike multinomial, an extended Nijboer-Zernike multinomial, and a generalized inverse matrix. The single focal plane high-precision testing method is low in cost, and high in accuracy and is suitable for manufacturing enterprises, scientific research and detection units of the optical imaging system.

The invention belongs to a database reading and writing method for aerospace, and particularly relates to a memory map-based mass data preprocessing method. The method relates to a data table generation module, a mass data fast import module, a mass data fast preprocessing module and a mass data fast retrieval module, wherein the mass data fast import module comprises a memory map sub-module. The method has the advantages that a rational efficient read-write mode and a database structure are established for mass data preprocessing of a spacecraft, a data table with an identifier is generated for telemetry information each time according to a test state and an index is created to facilitate fast retrieval and reading of specific parameters in databases; the gigabit information of the spacecraft is classified and sequentially and quickly inserted into the corresponding databases according to subsystems, parameter names, parameter types and the like, and the insertion completion time frame is second.

Systems and methods for X-ray phase-contrast imaging (PCI) are provided. A quasi-periodic phase grating can be positioned between an object being imaged and a detector. An analyzer grating can be disposed between the phase grating and the detector. Second-order approximation models for X-ray phase retrieval using paraxial Fresnel-Kirchhoff diffraction theory are also provided. An iterative method can be used to reconstruct a phase-contrast image or a dark-field image.

The invention discloses a convolutional neural network transcendentally enhanced phase retrieval detection system for intermediate frequency errors of optical elements. According to the system, a calibrated spatial light modulator is utilized to generate phase modulation of intermediate frequency errors; modulation light is projected on an imaging camera for reception, so as to obtain multiple data pairs of strength patterns and intermediate frequency error description items to serve as a training data set of a neural network; and a trained model is used for detecting real intermediate frequency errors. Compared with data obtained by adoption of simulation, the model trained by the data provided by the system is more suitable for recovering practical intermediate frequency errors. According to the system, initial de-optimization of a phase retrieval intermediate frequency error detection technology is realized; and by utilizing a convolutional neural network model in deep learning, a relationship between the strength patterns after intermediate frequency error modulation and error distribution is established to predict phase distribution of the intermediate frequency errors, and the result serves as an initial solution of a phase recovery algorithm, so that the convergence performance of the algorithm is effectively improved and the convergence speed is improved.

A multi-image encryption method based on the chaos and FrFT domain phase retrieval process includes the chaos scrambling step, the pure phase extraction step, the phase modulation step and the chaos diffusion step. The method includes the steps of firstly, scrambling multiple original gray level images to be encrypted through paired logistic mapping to obtain scrambled images; secondly, conducting the iterated phase retrieval process on each scrambled image to extract a pure phase function of the scrambled image; thirdly, conducting phase modulation on the pure phase functions on the multiple original gray level images to be encrypted; fourthly, conducting chaos diffusion on the amplitudes of the results obtained through modulation to obtain the final cipher text. According to the multi-image encryption method, nonlinearity and randomness of plaintext images in the spatial domain and the conversion domain are enhanced, the convergence speed of an encryption system is increased, and security of the encryption system is enhanced.

The invention belongs to the technical field of optical fiber communication, in particular to a multimode blind equalization digital signal processing algorithm applied to a quadrature duobinary (QDB) frequency-spectrum-compressed polarization multiplexing signal. According to the invention, in a coherent optical communication system, the multimode-blind-equalization-based digital signal processing algorithm is adopted to realize equalization compensation and demodulation of an optical signal which is subjected to QDB frequency spectrum compression; suitable filters are adopted in an electric field or an optical field to realize the QDB frequency spectrum compression of a polarization multiplexing signal; the polarization demultiplexing of a transmitted optical signal is realized by using a cascade multimode algorithm (CMMA); frequency offset estimation is realized by using a multimode quadrature phase shift keying (QPSK) partition method; and phase retrieval is realized by using the multimode QPSK partition method and maximum likelihood estimation. By the algorithm, the frequency spectrum efficiency of the signal can be effectively improved, the transmission quality of the signal can be improved, and high-spectral-efficiency coherent optical transmission can be realized.

The invention relates to a multi-dimensional object robust high-speed retrieval and positioning method for some feature images, which comprises the steps of pretreatment, feature point treatment, image robust high-speed accurate matching, and retrieval and positioning of matched images, and comprehensively solves the problems that an image retrieval and positioning system is unstable, low corresponding rate between the accuracy rate and speed, and poor positioning accuracy due to the size, rotation, viewing angle conversion, partial shielding, and close coupling of the images and other factors during the image object retrieval and positioning. The retrieval and positioning method provided by the invention belongs to the retrieval and positioning technical field of some feature images and has the effects of better flexibility, easiness in realization, stability, accuracy, high speed and high precision.

The invention relates to the technical field of optical fiber communication, in particular to a spatial multiplexing-based non-local oscillation coherent receiving optical fiber communication system. The spatial multiplexing-based non-local oscillation coherent receiving optical fiber communication system comprises a light source input unit, a light signal modulation unit, a spatial multiplexing and demultiplexing unit, a light signal conversion unit and an electric signal processing unit; the light source input unit is connected with the spatial multiplexing and demultiplexing unit through the light signal modulation unit; the light source input unit is also connected with the spatial multiplexing and demultiplexing unit; and the spatial multiplexing and demultiplexing unit is connected with the electric signal processing unit through the light signal conversion unit. By the spatial multiplexing-based non-local oscillation coherent receiving optical fiber communication system provided by the invention, an expensive narrow-linewidth adjustable local oscillation light source at a receiving end is eliminated, and the complexity degrees of the frequency estimation and the phase retrieval algorithm of the receiving end are reduced.