38results about How to "Achieve super resolution" patented technology

The invention discloses an image super-resolution reconstruction method based on dictionary learning and structure similarity, mainly solving the problem that a reconstructed image based on the prior art has a fuzzy surface and a serious marginal sawtooth phenomenon. The image super-resolution reconstruction method comprises the following implementation steps of: (1) acquiring a training sample pair; (2) learning a pair of high/low-resolution dictionaries by using structural similarity (SSIM) and K-SVD (K-Singular Value Decomposition) methods; (3) working out a sparse expression coefficient of an input low-resolution image block; (4) reestablishing a high-resolution image block Xi by using the high-resolution dictionaries and the sparse coefficient; (5) fusing the high-resolution image block Xi to obtain a high-resolution image X'I subjected to information fusion; (6) obtaining a high-resolution image X according to the high-resolution image X'I; and (7) carrying out high-frequency information enhancement on the high-resolution image X through error compensation to obtain a high-resolution image subjected to high-frequency information enhancement. A simulation experiment shows that the image super-resolution reconstruction method has the advantages of clear image surface and sharpened margin and can be used for image identification and target classification.

The invention discloses a light field multi-view image super-resolution reconstruction method based on deep learning. The method comprises the following steps: constructing a training set of high-resolution and low-resolution image pairs by adopting multi-view images which are obtained from a light field camera or a light field camera array and are distributed in an N * N array shape; constructing a multi-layer feature extraction network from the N * N light field multi-view image array to the N * N light field multi-view feature image; stacking the feature images and constructing a feature fusion and enhanced multi-layer convolutional network to obtain 4D light field structure features capable of being used for reconstructing a light field multi-view image; constructing an up-sampling module to obtain a nonlinear mapping relation from 4D light field structure features to a high-resolution N * N light field multi-view image; constructing a loss function based on the multi-scale feature fusion network, training the loss function, and finely adjusting network parameters; and inputting a low-resolution N * N light field multi-view image into the trained network to obtain a high-resolution N * N light field multi-view image.

The present invention discloses a MIMO (Multiple-Input Multiple-Output) radar waveform separation method based on compressive sensing. The method comprises: (1) emitting orthogonal waveforms; (2) receiving a target echo; (3) constructing a sensing matrix; and (4) performing sparse reconstruction of a target one-dimensional distance image. Based on the sparsity of a radar target on a distance direction, the MIMO radar waveform separation method based on the compressed sensing constructs the sensing matrix related to a transmitted waveform set and employs the joint sparse recovery algorithm to realize the orthogonal separation of different waveform components in the MIMO radar echo. The method provided by the invention replaces a traditional matched filtering process, effectively inhibits the high sidelobe of the distance direction caused by non-orthogonal waveforms, breaks through the restriction of signal bandwidth for the distance resolution, and improves the distance resolution.

The invention discloses a multi-scale multi-aperture optical imaging system. The multi-scale multi-aperture optical imaging system comprises a center optical imaging system and four secondary optical imaging systems having complete identical optical structures, wherein the center optical imaging system is a rotary symmetric system, the optical axis and the system center axis are superposed, the four secondary optical imaging systems are arranged at the back of the center optical imaging system, from the perspective of image cross sections, four optical axes of the secondary optical imaging systems are distributed on tip tops of a rectangle which takes the optical axis of the center optical imaging system as the center, and parallel lights emitted from different perspectives by a target are respectively imaged through the center optical imaging system and the four secondary optical imaging systems on different coordinate points on an image plane of one same detector. Independent multiple target images can be acquired by applying the multi-scale multi-aperture optical imaging system to acquire target object information, the center image resolution is higher than the secondary image resolution, images respectively have certain pixels superposed in the field of view, and the multi-scale multi-aperture optical imaging system can provide excellent support for subsequent data processing.

The invention relates to an improved dimension-reduction space-time adaptive processing-based ship-borne high-frequency ground wave radar sea clutter suppression method and belongs to the clutter suppression field. The objective of the invention is to solve the problems of large calculation amount and poor clutter suppression effect of an adaptive processing algorithm in the prior art. The improved dimension-reduction space-time adaptive processing-based ship-borne high-frequency ground wave radar sea clutter suppression method includes the following specific steps that: step one, distance transformation and Doppler transformation are performed on high-frequency ground wave radar echo baseband signals, so that distance unit and Doppler unit data can be formed; step two, azimuth processing is performed on the data of any one distance unit and any one Doppler unit through utilizing the MUSIC algorithm; and step three, dimension-reduction space-time adaptive processing is performed on the data which have been processed by the MUSIC algorithm, so that sea clutter can be suppressed. The improved dimension-reduction space-time adaptive processing-based ship-borne high-frequency ground wave radar sea clutter suppression method of the invention is applied to the clutter suppression field.

The invention discloses a super-resolution method for reconstructing a potential image based on a dynamic visual sensor, relates to the technical field of event camera application, and solves the problems that edge details of a grayscale image are not sharp enough, the grayscale of the same pixel value is not stable enough, the dynamic range is liable to lose and the like in a reconstruction junction in the prior art. The method is realized through a neural network structure, and comprises two steps of potential grey-scale map reconstruction and multi-image fusion; a high-quality high-resolution grey-scale map is reconstructed through a neural network architecture, and event signals and grey-scale map signals are considered at the same time, a series of potential grey-scale maps are reconstructed on the basis of the grey-scale map, and the pixel value of the fused high-resolution grey-scale map is stable and continuous. A plurality of potential grayscale frames are reconstructed, and the super-resolution of the APS grayscale image is realized by adopting a multi-image super-resolution method, so that the quality of super-resolution reconstruction is greatly superior to the reconstruction effect of the previous related methods. According to the method, the quality of the image super-resolution is improved by using a deep learning method.

The invention belongs to the technical field of image super-resolution, and discloses an image super-resolution method and system, and the method comprises the steps of processing an image SR through employing a 3D convolutional neural network 3DCNN in combination with the non-local self-similarity of an image, and proposing a 3DCNN-based non-local super-resolution method; modeling non-local similarity by directly utilizing 3DCNN, and extracting non-local similarity information of a natural image; constructing a 3DCNN basic model based on an eight-layer full convolutional network; designing a 3D convolutional neural network in 3DCNN, and proposing an RNN-based improved model, so that a basic model becomes a special example of the improved model. By using the non-local operation provided by the invention, the non-local similar information in the image can be effectively captured, and the SR reconstruction performance is improved; compared with an existing CNN model, the invention shows obvious reconstruction advantages, and is prominent in image scenes with rich structural information.

A antimony and bismuth intersection alloy mask read-only super-resolution CD structurally comprises a dielectric layer, a mask layer and a disk base and is characterized in that the dielectric layer consists of silicon nitride; the mask layer consists of antimony and bismuth alloy, that is, SbxBi(1-x) film, wherein, the variation range of x is from 0.7-0.9. The super-resolution CD is characterized by simple structure, low read-out power, high sensibility, good read-out cyclicity, etc.

The present invention relates to a novel optical nano-imaging scheme by using a transparent particle as an optical lens in order to achieve super-resolution nano-imaging. In particular, the present invention relates to a membrane for retaining a microsphere. The present invention relates to the membrane, and its application in super-resolution nano-imaging. In an aspect of the present invention, there is provided a membrane having a first surface and a second surface, the first and second surfaces oppose each other, the membrane comprising: (a) an opening penetrating through the membrane and is sized to retain a microsphere, wherein the opening is tapered, the size of the opening on the first surface is larger than the size of the opening on the second surface.

The invention relates to a quasi-phase super-resolution circuit of thoracoabdominal surface respiratory movement signals, and belongs to the technical fields of precision instruments and thoracoabdominal radiotherapy. The circuit comprises a signal conversion module, a phase difference multiple output module, a resistor chain multiphase generation module, a multiphase sinusoidal square wave conversion module and a multiphase fusion logic gate module; the modules are indispensable as a whole, and convert the respiratory movement signals of one cycle into quasi-phase square wave signals of multiple cycles jointly, and the null cycle of the signals is output by a super-resolution square wave with phase difference of a quasi phase; that is to say, super resolution of the thoracoabdominal surface respiratory movement signals is achieved, and further technical advantages which can be brought by the result are that in a process of multiple respiratory cycles, by judging frequency change of square wave signals, change of the respiratory frequency can be judged, and more importantly, due to the fact that the cycle of the super-resolution square wave is much smaller than the respiratory movement cycle, change of the respiratory frequency can be judged in a time range smaller than one respiratory cycle.

An auto-locating confocal scan microscope using optical fibre is composed of light beam input unit A, optical fibre scan unit B, measurement locating unit C alternative with B, probing-sampling unit D and control data acquisition and processing unit E. Said A, B, D and E constitute an optical fibre confocal scan microscope. Said A, B and C constitute a quasi-confocal small-view-field microscope. That is, it has both the microscan-tomographic imaging function and one-pass imaging function. It is suitable for examining 3D structure with nm depth and submicron width.

The invention relates to a super-resolution method for chest-and-belly surface breathing movement signals and belongs to the technical fields of precise instruments and chest-and-belly radiotherapy. According to the method, firstly, breathing movement signals are converted into periodic signals through a signal conversion module, then, equal-phase-difference III output with the phase difference ofPI/2 is formed through a phase difference multi-output module, then, equal-phase-difference V output with the phase difference of PI/5 is formed through a resistance chain multi-phase generation module, then, square waves are formed through a multi-phase sine square wave conversion module, and quasi-phase super-resolution square wave output is formed through a multi-phase merged logic gate module, so that the breathing movement signals of one period can be changed into quasi-phase square wave signals of multiple periods, and phasic difference is a zeroth period of the quasi-phase super-resolution square wave output signals.

The invention discloses a terahertz super-resolution imaging system and method based on a confocal waveguide. The system comprises a terahertz or millimeter wave source, a pinhole, the confocal waveguide, an imaging target, a two-dimensional scanning mechanism, a second waveguide body and a detector. The terahertz or millimeter wave source is used for giving out a terahertz or millimeter wave; thepinhole is formed in an incident light path of the source and used for obtaining terahertz or millimeter transmission light; the confocal waveguide is composed of two confocal waveguide bodies, wherein the first waveguide body is arranged on an incident light path after the terahertz or millimeter wave transmits the pinhole and used for transmitting and focusing the terahertz or millimeter wave;the imaging target is arranged on the focal position of the first waveguide body; the two-dimensional scanning mechanism is used for controlling the imaging target to conduct scanning imaging; the second waveguide body is arranged on a transmission or reflection light path of the imaging target and used for transmitting and focusing a transmission wave; the detector is arranged on the focal position of the second waveguide body and used for detecting the intensity of the terahertz or millimeter wave, and finally imaging is achieved. According to the system, the confocal waveguide is used for replacing a traditional confocal lens to achieve focal imaging, the transmission loss of the terahertz or millimeter wave is reduced, the diffraction limit can be broken through, and super-resolution imaging is achieved.

The invention discloses a radar distance super-resolution calculation method based on a sparse Bayesian learning algorithm. The method comprises the following steps of S1, carrying out the pulse compression of an echo signal of a radar, and determining a group target radar signal segment, S2, performing frequency dechirp processing on the group target radar signal segment to obtain a single-frequency signal, S3, performing super-resolution processing on the single-frequency signal by using a sparse Bayesian learning algorithm to obtain a frequency point position of a group target radar signal, and S4, determining a radar distance according to the frequency point position of the group target radar signal. Compared with a traditional pulse compression processing algorithm, the resolution effect of the radar distance super-resolution algorithm based on the sparse Bayesian learning algorithm can be improved by more than one time.

The invention relates to a meter-wave radar low-elevation height measurement method based on robust principal component analysis noise reduction, and the method comprises the steps: carrying out the modeling of an array receiving signal of a meter-wave radar into a two-dimensional receiving signal matrix, and obtaining an array receiving signal model; establishing a convex optimization model of a low-rank signal matrix and a sparse noise matrix by adopting a noise reduction method based on robust principal component analysis; solving the convex optimization model by using an improved non-precise augmented Lagrangian multiplier method to obtain an optimal low-rank signal matrix without noise pollution; establishing a covariance matrix by using the optimal low-rank signal matrix, and constructing the maximum likelihood estimation of the direct wave elevation direction of arrival in combination with the covariance matrix and the synthetic steering vector to obtain a target elevation estimation value; and calculating the target height by using the target elevation angle estimated value. According to the method, the problems that in the prior art, under the conditions of non-uniform noise, low signal-to-noise ratio and few snapshots, the angle measurement performance is reduced, and the angle resolution is insufficient are solved, and the method has high angle measurement stability and good angle resolution.