20 results about "Imaging problem" patented technology

The application is suitable for the technical field of radar signal processing, and provides a deep unfolding ISAR super-resolution imaging method based on sparse-neighborhood joint constraint, comprising: firstly, constructing an ISAR image degradation model, constructing an ISAR super-resolution imaging problem based on the ISAR degradation model and the compressed sensing theory, using ADMM to solve the ISAR super-resolution imaging problem, unfolding the process of solving the ISAR super-resolution imaging problem by ADMM into a multi-level neural network, and finally inputting low-resolution ISAR echoes into the trained neural network to obtain an ISAR super-resolution imaging result. The application combines sparse constraint and neighborhood amplitude constraint, deduces a corresponding signal model and an ADMM solving algorithm, effectively improves the reconstruction ability of the algorithm to complex target details and the representation ability of the algorithm to structural information, and can realize effective super-resolution imaging based on narrowband short-aperture echoes.

The invention discloses an industrial machine vision-oriented strong backlight environment imaging sharpening system and method. The method comprises the following steps: S10, performing optical coding imaging; s20, carrying out improved L-R iterative recovery and first-stage wavelet denoising; s30, carrying out secondary wavelet denoising and fine recovery; and S40, carrying out iterative convergence and output. According to the invention, a specific wavefront coding optical imaging module and an improved L-R image restoration algorithm module are integrated into a complete visual detection system for solving the problem of industrial strong backlight imaging. In the L-R iteration process, a staged denoising step based on wavelet transformation is introduced, and before each iteration, first-level high-threshold wavelet denoising is carried out on a residual error; and in a specific stage in the iteration process, carrying out secondary middle threshold wavelet denoising on the current estimation image.

The application discloses a kind of synthetic aperture assisted multi-channel radar forward-looking learning imaging method, first, the echo data of the region to be imaged is acquired, the data obtained is compressed in distance direction pulse, then the echo data after pulse compression is imaged using the synthetic aperture assisted depth unfolding iterative shrinkage threshold algorithm ISTA network, i.e. in real aperture dimension, the imaging problem is converted into the estimation problem of target scattering coefficient, the target function is solved using the depth unfolding ISTA network, then the azimuth focusing is carried out using back projection BP algorithm in synthetic aperture dimension, then the synthetic aperture dimension result and the ISTA network result are fused to obtain the left-right unambiguous super-resolution imaging result.The method of the application overcomes the problem of left-right ambiguity and low resolution of single-base radar forward-looking imaging, realizes multi-channel radar forward-looking high-resolution imaging, and through depth learning, the parameter setting step is saved, the efficiency is improved, and the problem of signal-to-noise ratio sensitivity is overcome.

The invention discloses a welding turbid environment vision sharpening system and method based on circularly polarized light imaging. In the system, an active circularly polarized lighting module comprises a laser, a beam expander, a linear polarizer and a first quarter-wave plate which are connected in sequence; the polarized image acquisition module comprises a second quarter-wave plate, an analysis mirror and a camera which are connected in sequence; the active circular polarization illumination module is located at the position of a welding workpiece, and the polarization image acquisition module is located at the position of a welding gun; and the image processing and control unit controls the on-off of the active circular polarization illumination module and the image acquisition of the polarization image acquisition module, and performs image recovery. According to the invention, the lighting source is innovatively replaced by the circularly polarized light from the traditional linearly polarized light, and a set of self-adaptive image recovery algorithm is combined, so that the imaging problem of the vision of the welding robot in a turbid and high-reflection environment is successfully solved, and reliable visual perception guarantee is provided for realizing high-precision and intelligent welding operation.

The utility model discloses a dustproof motor, which comprises a motor body and an electrostatic adsorption mechanism arranged on the motor body and used for adsorbing small particles in the motor body. The electrostatic adsorption mechanism comprises a conductive unit and a dust sticking unit covering the surface of the conductive unit. According to the utility model, through the synergistic effect of active electrostatic adsorption and passive dust-sticking fixation, the limitation of the traditional motor dust-proof technology is broken through, the POG imaging problem caused by the fact that small particles fall on a BG is avoided, and the imaging quality and the reliability of the camera module are remarkably improved.

The invention relates to the technical field of 3D printing, in particular to a 3D printing process monitoring system and method based on a high-speed two-dimensional swing mirror. According to the invention, the high-speed two-dimensional oscillating mirror is arranged in front of the object space light path of the imaging lens group, so that rapid optical scanning of the working surface is realized; the controller firstly divides a working face into a target dot matrix, and then calculates the diopter dot matrix of the liquid zoom lens according to the current printing layer number and the object distance of each point position; in the actual detection process, the controller drives the high-speed two-dimensional swing mirror to sequentially point to each target point position, and meanwhile, the controller controls the liquid zoom lens to adjust the focal length in real time according to a preset diopter dot matrix, so that the industrial camera can obtain a clear image at each point position; not only is the problem of high-precision imaging in a large-view range solved, but also adverse effects caused by mechanical vibration are avoided through a pure optical scanning mode; the whole detection process is more efficient, the system structure is simple, and the detection precision and reliability are improved.

The application discloses a radar super-resolution forward-looking imaging method and system, constructs a weight vector and a manifold matrix in the distance and azimuth dimensions according to the waveform parameters and the array structure of the radar, and further constructs a range spread function and an azimuth spread function.On this basis, target features are extracted through mean shift theory, and the constructed range spread function and azimuth spread function are used when calculating a mean vector, so that the diffusion mode of the target in the radar image is more in line with the diffusion mode, an image with a smaller full width at half maximum can be obtained, and super-resolution imaging is realized.The application innovatively considers the super-resolution imaging problem from the perspective of the image domain and in combination with the characteristics of the radar.Compared with the previous super-resolution method based on signal processing, the application is more universal, is not easily limited by factors such as the waveform and the array structure, and has no strict performance upper limit, so that the resolution can be further improved on the basis of the existing super-resolution imaging result.

The invention belongs to the technical field of computer vision and image processing, discloses a nighttime high-quality imaging method for an electronic rearview mirror, and aims to solve the problem that daytime and nighttime imaging needs to be processed at the same time and respectively construct learning samples and solve light source masks for daytime and nighttime images according to a brightness threshold. And a two-dimensional U-net structure is used as a model basis, and attention and a multi-scale mechanism are added to construct a learning model, so that better CMS image high-quality imaging is realized.

The utility model discloses a multi-point dynamic adjustment type tunnel light source which comprises a working frame, a workpiece positioning unit, a light source assembly and a visual assembly are fixedly arranged on the working frame, and the light source assembly and the visual assembly are located at the positions corresponding to the workpiece positioning unit. The height, the angle and the distance of the two reflecting shells can be adjusted relative to the workpiece, so that the irradiation angle of a light source is adjusted according to different detection requirements and object shapes, the optimal illumination effect is achieved, the visual imaging problem of the metal high-reflection rotary body is solved, and the detection accuracy is improved. According to the invention, a plurality of camera modules can be simultaneously used and X-axis moving working scenes can be handled, and light can be subjected to set reflection and heat dissipation in the reflection cavity, so that a uniform diffuse illumination effect is realized, shadows and light reflection are reduced, a uniform and stable illumination environment is provided for a machine vision system, image information of an object can be accurately acquired, and the visual effect of the machine vision system is improved. And powerful guarantee is provided for machine vision detection.

The invention relates to an adaptive radar sparse imaging method based on a near-field physical model, and the method is constructed at least based on the following scheme: imaging modeling: at least constructing an imaging problem of a specific distance unit into a linear inverse problem; designing a physical decoupling network architecture, and at least designing a learnable induction strategy and an adaptive threshold; offline training: updating learnable parameters at least according to the obtained total loss to complete model training; and performing online reasoning, and outputting a target image at least based on forward propagation. The problems of model mismatch and imaging failure caused by the fact that an existing deep learning model LISTA cannot adapt to the change of a physical observation model along with the distance in near-field or high-resolution radar application are at least solved.

The present application relates to the technical field of image processing, and particularly relates to a kind of image light uniformity method, device, equipment and medium, the method comprises: obtaining all microscopic images of the object to be measured, all microscopic images are carried out shadow correction processing, obtain all target microscopic images, all target microscopic images are carried out splicing processing, obtain the target fusion image of preset target size, target fusion image is carried out curve fitting processing, obtain target light uniformity image, based on target light uniformity image, determine the image detection result of the object to be measured. Visible, all microscopic images obtained are carried out shadow correction processing, splicing processing and curve fitting processing to obtain target light uniformity image, and then according to target light uniformity image, determine the image detection result of the object to be measured, so as to solve the imaging problem caused by uneven illumination in the prior art, so that the finally obtained image realizes uniform illumination, eliminates shadow interference, effectively improves the measurement accuracy of image.

According to the three-wavelength and four-angle polarization sensitive holographic imaging system provided by the invention, the imaging system has wavelength and polarization resolution capability by using a mode of increasing information channels and synchronously acquiring; distortion phase and scattering noise, caused by medium refractive index disturbance and smoke multiple scattering in a severe imaging environment, of a target light field are analyzed, and the complex and severe environment imaging problem similar to an energetic particle combustion field can be well solved. Furthermore, four-angle polarization is introduced to increase an independent channel, so that the anti-interference capability of holographic imaging is improved, and the information storage capacity of the system is improved by four times on the premise that the physical size does not need to be increased. The problem of extinction and speckle interference caused by high particle concentration number is solved by means of the thought of synchronous interpretation of multi-dimensional light field information, and polarization characteristic parameters capable of distinguishing particles from background areas are extracted from polarization channels by means of the advantage of more stable transmission of polarization characteristics of light so as to obtain three-dimensional position information of the particles.

The invention relates to the technical field of intelligent visual inspection, and discloses a real-time glue dispensing and overflowing monitoring method and system based on intelligent vision, and the method comprises the steps: collecting image data in a glue dispensing process in real time, and carrying out the regulation and intelligent sampling through a self-adaptive light source; carrying out segmentation and feature extraction on the colloid region; analyzing time sequence change characteristics and a multi-point space association relationship in a dispensing process, and dynamically identifying and predicting excessive glue in advance; a glue overflow detection result is converted into an equipment control instruction, and preventive intervention is carried out through advanced control; local real-time processing and cloud deep analysis are carried out; constructing a digital twin model in the dispensing process, and performing quality prediction and process optimization to obtain a quality prediction strategy and a decision support scheme; according to the invention, a self-adaptive light source regulation and control and reflection interference elimination technology is adopted to solve the problem of strong reflection metal surface imaging.

The application discloses a kind of acoustic three-dimensional imaging method and image sensor of resonant holographic sensing, applied to acoustic imaging field, the imaging method utilizes acoustic lens focusing reception target object reflected ultrasonic echo, forms imaging sound field and is imaged in acoustic image sensor surface, the acoustic image sensor is by Helmholtz type microcavity metasurface, RS type laminated MUT unit array and special integrated circuit sequentially laminated and constitutes, and it is arranged in pixelization array, imaging sound field enters corresponding internal microcavity through each surface through hole of metasurface, and it is carried out resonant interference with the reference sound field radiated by the bottom RS type laminated diaphragm, forms holographic sound field and reacts on RS type laminated diaphragm, generates holographic resonance signal, after the signal is peak detection, denoising processing, addressing readout and phase reconstruction, the three-dimensional image of target object ultrasonic echo is formed.The present application can overcome the existing acoustic three-dimensional imaging technology faces the imaging problem of large field of view, high resolution and fast real-time.

The invention discloses a transfer orbit design method in on-orbit service of a synchronous orbit satellite. The problem of how to effectively image a high-orbit target satellite with low fuel consumption is solved. Belongs to the technical field of space. Comprising the following steps: determining total transfer time according to different initial states of a service satellite and a target satellite; constructing a transfer orbit mathematical model based on the total transfer time to obtain the running time of the service satellite on the initial orbit, the transfer orbit and the approaching orbit in the transfer process and the speed increment for completing the Horman transfer; taking a relative position vector included angle and an illumination angle of two stars at the end of intersection as an objective function, taking total transfer time as a constraint condition, taking the operation time of a service star in an initial orbit and an approaching orbit as an optimization variable, optimizing the objective function by using a genetic algorithm based on an HPOP model, and correcting parameters of a transfer orbit mathematical model to obtain a transfer orbit mathematical model; obtaining a transfer orbit mathematical optimization model; according to the method, effective imaging of the high-orbit target with low burnup is realized through the mathematical optimization model of the transfer orbit.

The invention discloses an image correction method and a computer readable storage medium. The method comprises the following steps: normalizing each color component value of each pixel point in a to-be-corrected image; calculating an average value of each color component according to each color component value after each pixel point is normalized, and determining a to-be-corrected color component according to the average value of each color component; correcting each to-be-corrected color component value of each pixel point; and generating a corrected image according to each to-be-corrected color component value after each pixel point is corrected and other color component values. The method can solve the problem of image imaging which changes at any time under complex conditions, and improves the image quality.

The application discloses a TOF-PET reconstruction method based on implicit neural representation, which has a significant technical advantage in the deep integration of physical modeling and data-driven methods. Unlike traditional deep learning methods that rely only on large-scale training data for end-to-end mapping, the method explicitly embeds the physical forward model of TOF-PET imaging in the INR framework. By introducing a strict data consistency term, the network output always satisfies the system projection constraint and TOF response characteristics. This effectively avoids problems such as excessive smoothing, artifact enhancement, and structural distortion under adverse imaging conditions such as low count and low dose. The method achieves high-precision, quantifiable, and physically interpretable reconstruction results, significantly improving the reliability of the reconstructed image in clinical quantitative analysis. Without changing the overall modeling idea, the method can be flexibly extended to dynamic PET imaging, activity-attenuation joint reconstruction, motion compensation reconstruction, and multi-modal fusion tasks such as PET-CT and PET-MRI, and can handle a variety of complex imaging problems.

The application relates to the technical field of 3D printing, in particular to a 3D printing process monitoring system and method based on a high-speed two-dimensional swing mirror. The application sets a high-speed two-dimensional swing mirror in front of the object light path of an imaging lens group, realizes rapid optical scanning of a working surface, and controls the working surface to be divided into a target point array first, then calculates a diopter point array of a liquid zoom lens according to the current printing layer number and the object distance of each point; in the actual detection process, the controller drives the high-speed two-dimensional swing mirror to point to each target point in turn, and controls the liquid zoom lens to adjust the focal length according to the preset diopter point array in real time, so that the industrial camera can obtain a clear image at each point; the application not only solves the high-precision imaging problem in a large field of view, but also avoids the adverse effects caused by mechanical vibration through pure optical scanning; the whole detection process is more efficient, the system structure is simple, and the detection precision and reliability are improved.

The application provides a kind of polarized optical-photoacoustic quantitative three-dimensional imaging method and system, it is related to data processing technical field, the method comprises: obtaining the anisotropic structure parameter of sample surface layer as verification benchmark;Based on photoacoustic effect, the polarized photoacoustic signal of deep sample is collected;Based on anisotropic light energy absorption theory model, the polarized photoacoustic signal data set is reconstructed to obtain the anisotropic structure reconstruction result of deep sample;Combined with the layered structure information obtained by optical imaging, the photoacoustic reconstruction result is verified and visualized correspondingly.The application realizes the comprehensive innovation of polarized photoacoustic imaging technology from theoretical model to system implementation, to solve the high-precision, high-specificity three-dimensional imaging problem of microstructure in deep living tissue.

The invention provides a polarization optical-photoacoustic quantitative three-dimensional imaging method and system, and relates to the technical field of data processing, and the method comprises the steps: obtaining an anisotropic structure parameter of a surface layer of a sample as a verification reference; based on the photoacoustic effect, a polarized light acoustic signal of the deep sample is collected; based on an anisotropic light energy absorption theoretical model, reconstructing the polarized light acoustic signal data set to obtain an anisotropic structure reconstruction result of the deep sample; and performing corresponding verification and visualization on a photoacoustic reconstruction result in combination with layered structure information obtained by optical imaging. According to the invention, a polarized light acoustic imaging technology which is comprehensively innovated from a theoretical model to system implementation is realized, so that the high-precision and high-specificity three-dimensional imaging problem of a microstructure in a deep living body tissue is solved.