204 results about "Adaptive optics" patented technology

The invention discloses an artificial intelligence machine vision image acquisition system, and the system comprises a multi-mode perception layer which integrates a self-adaptive optical module, inhibits metal reflection, captures a visible light to short wave infrared image, and captures a motion edge; the dynamic adaptive layer adopts an illumination compensation and motion compensation module to dynamically adjust camera parameters and micro displacement compensation, feeds back an illumination trend, outputs a motion vector to the cognitive layer, generates a confrontation sample through a GAN, simulates virtual defects in combination with a physical engine, and expands training data; the cognitive reasoning layer is used for deploying a dynamic routing network, distributing computing resources according to image complexity and optimizing feature extraction efficiency; reducing data deviation through anti-fact analysis, and generating a thermodynamic diagram to explain a detection basis; and the collaborative decision-making layer is used for rapidly screening samples by edge nodes, training a global model by cloud aggregated data, automatically triggering manual rechecking when the confidence coefficient of the model is insufficient, synchronously optimizing a training set and a causal reasoning module by a rechecking result, and improving the labeling efficiency through AR assistance.

The invention provides an image recognition system and method based on deep learning, and the system comprises a self-adaptive optical collection module, a heterogeneous preprocessing pipeline, a hierarchical reconfigurable convolutional network, a multi-dimensional training optimization engine and a cross-modal verification output interface, aperture parameters are dynamically adjusted through deep reinforcement learning; the heterogeneous preprocessing pipeline comprises a quantum noise modeling non-local mean noise reduction unit, a double-discriminator generative adversarial network enhancement unit and a dynamic normalization unit; the hierarchical reconfigurable convolutional network adopts a staged feature distillation structure and comprises a separable convolution module, a mixed pooling layer and a three-dimensional attention fusion module. According to the invention, through a multi-modal data fusion and dynamic optimization mechanism, the image acquisition quality in a complex illumination and noise scene is improved, the adaptability of the model to different environments is enhanced, and all modules work cooperatively to realize an end-to-end efficient identification process.

The invention discloses an intelligent sofa fabric defect detection system and method based on AI image recognition, and relates to the technical field of intelligent fabric defect detection.The system comprises a plurality of modules, an image acquisition module is combined with a multispectral camera, quantum enhanced imaging and adaptive optics are introduced, polarization and terahertz cameras are further arranged, the image quality is improved, and the detection accuracy is improved. Image preprocessing adopts deep learning adaptive filtering and the like; aI identification is combined with meta learning, a knowledge graph and data analysis, time sequence analysis is introduced, a VR / AR technology is adopted for result output, a reinforcement learning algorithm is adopted for feedback control, and fabric flaws can be comprehensively and efficiently detected. The system is high in detection precision, can capture flaws in all directions, greatly improves the detection efficiency, meets the requirements of large-scale production, can deeply analyze data, predicts the quality trend, achieves the real-time and accurate production regulation and control, sample management, remote monitoring and other modules, and brings remarkable benefits to enterprises.

The invention relates to the technical field of industrial nondestructive testing, in particular to a paper drum defect full-inspection method based on machine vision, which comprises the following steps: step 1, constructing a curved surface self-adaptive optical environment: arranging annular low-angle strip-shaped light source arrays along the circumferential direction of a paper drum at equal intervals, and forming a dynamically adjusted set acute angle between the emergent direction of each light source and the surface normal of the paper drum, the acute angle is reduced along with the increase of the diameter of the paper barrel, and near-infrared coaxial diffused light sources are arranged at the two axial ends of the paper barrel; step 2, motion compensation three-dimensional reconstruction; 3, multi-modal defect hierarchical decision making: expanding the three-dimensional point cloud into a two-dimensional image along the circumferential direction, and constructing a dynamic reference grid with the grid cell size adaptively adjusted along with the pattern complexity on the two-dimensional image; cooperatively extracting geometric structure features, surface texture features and gluing seam features; and outputting defect classification through a three-level decision tree. By means of the method, efficient and accurate paper barrel full inspection can be achieved, and the automation level of a production line and the paper barrel quality control capacity are effectively improved.

The invention discloses a model-based adaptive optical dynamic modeling and control method and device, and belongs to the cross field of adaptive optics and model-based reinforcement learning, and the method comprises the steps: defining the state of an AO system as a combination of multi-mode observation, and carrying out the corresponding coding; fusing the multi-modal observation data through a cross attention mechanism to form a unified observation potential state rich in dynamic information; a deep dynamic world model based on the recurrent neural network is constructed, the model maintains a hidden state and deduces a discrete potential state, and the hidden state and the discrete potential state jointly form complete potential state representation of the world model; rebuilding original observation and predictive rewards according to the complete potential state representation of the world model; and training a reinforcement learning agent in the learned world model, learning an optimal strategy through an imaginary trajectory, and outputting key parameters of the AO system in real time. According to the method, the adaptability, the control precision and the robustness of the AO system in a complex dynamic environment are improved.

The invention discloses a die casting burr detection system and method based on machine vision, and relates to the field of machine vision detection.The method comprises the steps that based on a spectral feature vector set, a curved surface self-adaptive optical compensation mechanism is used, a polarization-spectral image set is obtained again through light path optimization, and the optimized polarization-spectral image set is obtained and corrected; generating a high-resolution image set, performing illumination distribution analysis on the surface of the die casting through the high-resolution image set to form a multi-exposure image set, inputting the multi-exposure image set into the constructed U-Net model for semantic segmentation, generating a high-dynamic-range image, and performing shadow elimination on the high-dynamic-range image by applying a gradient Poisson fusion algorithm to obtain a high-dynamic-range image. And after the shadow is eliminated, a pre-trained Mask-R-CNN model is used for detection, and a burr detection result is output. According to the method, the polarization-spectrum image set is optimized through curved surface adaptive optics, and imaging optimization and geometric correction of a complex curved surface area are achieved.

The invention discloses a single crystal turbine blade crystal orientation defect detection system based on adaptive optics, particularly relates to the field of nondestructive testing, and is used for solving the problem that thermal damage and signal consistency are difficult to consider in acoustic imaging of complex structural parts. Through partition energy threshold setting, dynamic pulse planning, photoacoustic processing and unified coordinate cross-subarea data synthesis, local thermal damage caused by excessive concentration of energy is effectively avoided in the material detection process, and the requirement for collecting high-bandwidth signals can be met in a large range. The signal-to-noise ratio and the detection efficiency are improved by regulating and controlling pulse energy and a scanning path, the amplitude comparability and the image coherence among regions are ensured through signal normalization correction and subregion image set splicing, high-resolution and comprehensive-coverage acoustic mapping is output, the crystal orientation defects of the single crystal turbine blade are positioned and evaluated, the detection precision is improved, the measurement period is shortened, and the method is suitable for large-scale popularization and application. And the method has an applicable value for aero-engine parts requiring rapid and high-precision nondestructive detection.

The invention discloses a lubricating oil quality on-line monitoring system, and relates to the technical field of lubricating oil quality detection.The system comprises a photoacoustic spectrum detection module which is composed of a modulation light source, a sample pool, a photoacoustic signal detection component and a signal processing component. The photoacoustic spectrum detection module is used for accurately analyzing components such as impurities and oxidation products in the lubricating oil from the microscopic chemical composition level, and the self-adaptive optical imaging module is used for acquiring high-resolution images of the impurities in the lubricating oil in real time and revealing the quantity, size, shape and spatial distribution condition of the impurities; richer and more accurate data support is provided for the evaluation of the quality of the lubricating oil, and the change of the quality of the lubricating oil and the potential abrasion problem of equipment can be found in time.

The invention provides a differential InSAR system, method and application and a readable storage medium. The system comprises a laser emission module, a radar detection module and a self-adaptive optical compensation module. The laser emission module supplies energy to the system and splits light; the radar detection module cooperatively detects a target by means of two symmetrical branches, and acquires and processes echoes to achieve the purpose of differential detection. The adaptive optical compensation module corrects laser wavefront distortion caused by atmospheric turbulence in real time according to deep learning and a wavefront detection mechanism, and ensures accuracy and stability of long-distance remote sensing. The method is characterized in that the adaptive optical performance is improved by combining deep learning, and phase noise is suppressed. The method has far-reaching significance in the fields of topographic mapping, geological disaster monitoring and the like, and can accurately measure the topographic elevation and the ground surface infinitesimal displacement, for example, the early millimeter-level variation is sensitively captured in the landslide early warning, and the opportunity is robbed for disaster prevention and control.

A method and system for estimating dimensions of vehicle exterior defects using multiple cameras arranged in a predefined configuration. The method comprises receiving images from multiple strategically positioned image sensors including side cameras parallel to a vehicle height axis, diagonal cameras at an inclined angle, and roof top cameras perpendicular to the height axis. An angle to detected defects is computed based on image sensor parameters. Different distance calculations are applied based on vehicle section location, with specialized formulas for windshield, back window, and roof components. Defect sizes are computed by determining multiple defect dimensions, with at least one dimension being adjusted by an angular correction factor derived from the relationship between camera angle and vehicle surface orientation. The system implements comprehensive validation through cross-referencing between cameras, comparison with known specifications, and measurement consistency analysis across multiple frames.

The invention discloses a self-adaptive optical lens curved surface optimization processing method based on machine learning, and the method comprises the steps: generating an optimal target curved surface which is easy to process based on the optical performance and manufacturability criterion through employing a first machine learning model; after the current surface shape error is obtained through machining and measurement, modeling is conducted on the non-linear and space coupling effect in the machining process through a second machine learning model, so that an optimal compensation machining scheme is predicted and determined; compensation processing is executed, and iterative circulation is carried out until the surface shape error is converged to a preset threshold value; according to the method, the processing convergence can be improved from the source, high-precision intelligent error compensation is realized, and the manufacturing efficiency and the yield of high-end optical elements are expected to be improved.

The invention discloses a high-throughput multi-row parallel high-speed scanning high-resolution microscopic imaging device and method based on adaptive optics. A spatial light modulator is used for carrying out pixel-level regulation and control on a light beam phase to generate a strip-shaped light spot, and a special arrangement mode of a detector array is used for realizing a one-to-many relationship between a scanning line and an imaging line and realizing improvement of an imaging speed, a signal-to-noise ratio of an imaging result and a resolution ratio; a full-digital area array wavefront regulation and control method is utilized, a spatial light modulator is used for rapidly and dynamically compensating a laser wave surface according to imaging result feedback, the laser wave surface is free of distortion, and the influence of aberration on an imaging result is reduced; designing a double telecentric Fourier transform structure capable of simultaneously keeping rotational symmetry of object plane and image plane light beams, and realizing distortionless mapping of scanning light beam quality and a linear signal system; a switchable design of two scanning modes is developed, high-flux, high-resolution and high-speed light field scanning is realized, and the advantages of large visual field, high resolution and high frame rate imaging are achieved.

A digital adaptive optics encoder module includes an input mounting flange, a collimating lens, a bandpass filter, digital adaptive optic elements, refocusing lens, an output mounting flange, and a housing. The input mounting flange is capable of attaching to a telescope. The collimating lens is capable of expanding light from a target to fill a plurality of primary apertures. The bandpass filter has a bandwidth ranging from about 40 nm to about 100 nm. The digital adaptive optic elements include the plurality of primary apertures, an optical spreader, a focusing optic, and a detector. The refocusing lens is capable of refocusing an output from the digital adaptive optic elements onto a sensor plane. The output mounting flange is capable of attaching to an output connection. The housing encloses all of the interior components of the digital adaptive optics encoder module.

The invention discloses a compensation device and method for a digital holographic microscope, relates to the technical field of optical imaging, and can dynamically adjust optical response according to wavefront distortion data measured in real time by using a deformable lens group and a real-time wavefront sensing technology. The adaptive optical element accurately controls the light wavefront, effectively expands the imaging depth, optimizes the focus control, and enhances the resolution. In addition, according to the compensation phase algorithm, the required compensation is calculated by introducing a regularization item, so that the adaptability of the algorithm to environment change is improved, and the robustness of the algorithm in the imaging process is enhanced. Reconstruction and post-processing, such as phase recovery and image sharpening, of the digital hologram are carried out through the constructed phase recovery algorithm, a high-quality imaging result can still be generated under the condition that noise is large or imaging conditions change, and it is further guaranteed that the algorithm can maintain high performance under various imaging conditions.

The invention belongs to the technical field of optical calculation, and discloses an error-adaptive optical diffraction neural network in-situ training method, which comprises the following steps: based on the phase of each phase modulation layer, calculating the light field complex amplitude Uk of the input sample complex amplitude on the rear surface of each phase modulation layer and the light field complex amplitude Uo of the input sample complex amplitude on an output plane after forward propagation; calculating an error light field P based on Uo, carrying out three-dimensional central symmetry on a back propagation light path of the error light field P without changing the phase distribution of each phase modulation layer, and calculating the light field complex amplitude of the complex amplitude of the symmetrical error light field on the front surface of each phase modulation layer after forward propagation; the method comprises the steps of obtaining the light field complex amplitude Un + 1-k'of P back propagation to the rear surface of each phase modulation layer, calculating the gradient related to a loss function based on Uk and Un + 1-k 'so as to carry out updating, and carrying out the next round of training until a preset training round or loss convergence is reached. The method can adapt to errors existing in an optical system.

The invention provides an intelligent vehicle-mounted electronic rearview mirror system based on optical fiber transmission, and the system comprises a camera module which is installed outside a vehicle body; the host is connected with the camera module through an optical fiber and is used for receiving and processing the video data acquired by the camera module; and the rearview mirror display screen is connected with the host and is used for displaying the video processed by the host. Data transmission is carried out through the optical fiber, high-speed and uncompressed video signal transmission is ensured, delay and data loss are reduced, and the stability and reliability of the system are improved. The camera module comprises a self-adaptive optical assembly, the self-adaptive optical assembly comprises a wavefront detector, a wavefront controller and a wavefront corrector, and the self-adaptive optical assembly can correct subtle image distortion introduced by lens deformation or aerodynamic effect during high-speed driving, so that the imaging definition under the extreme illumination condition is further improved.

The invention discloses an adaptive optical anti-shake control method for a camera module. The adaptive optical anti-shake control method comprises the following steps: S1, collecting real-time vibration data of the camera module in a working process; s2, preprocessing the real-time vibration data to obtain pose change information; s3, constructing a global search model, and defining an anti-shake compensation parameter search space; s4, performing multiple rounds of global iteration by using an improved seagull optimization algorithm, and determining a candidate parameter solution and a candidate region; s5, performing local sampling and iterative tuning by adopting a Bayesian optimization algorithm to obtain an optimal anti-shake compensation parameter; s6, the optimal anti-shake compensation parameters are issued to a camera module control unit, and an optical anti-shake assembly is driven to adjust the position or angle of a lens group; and S7, carrying out quality evaluation on the compensated image, and triggering whether to carry out global search or local optimization again or not. According to the invention, by constructing an adaptive optimization control mechanism, high-precision and low-delay real-time optical anti-shake compensation of the camera module in a dynamic environment is realized.

The invention discloses a self-adaptive optical system parameter adjusting method and device combining offline pre-training and online reinforcement learning, and belongs to the field of intelligent control combining self-adaptive optics and reinforcement learning. Multi-modal state modeling, off-line strategy pre-training and an online strategy enhancement mechanism guided by behavior advantages are introduced, and key control parameters in the AO system are dynamically optimized by using a reinforcement learning algorithm. According to the method, multi-modal heterogeneous data can be effectively integrated, the dynamic modeling and prediction capability of the system is improved, and stable, efficient and robust control strategy updating is realized.

The invention relates to a self-adaptive optical matrix calculation acceleration method based on an ARM framework, and belongs to the field of self-adaptive optical wavefront correction technology and embedded parallel calculation. The method comprises the following steps: analyzing a mathematical mapping relationship between a microlens array and wavefront phase distribution in a Hartmann sensor image, and constructing a wavefront matrix operation model based on an ARM multi-core architecture; designing a Hartmann image data rearrangement strategy, dividing a single-frame Hartmann image into multiple task calculation units, and realizing parallelization processing of wavefront slope calculation; a wavefront matrix operation kernel is optimized according to ARM NEON instruction set characteristics, and vectorization calculation of multi-channel wavefront reconstruction is realized by adopting an SIMD parallelization technology; and computing resource allocation is coordinated through a dynamic thread scheduling algorithm, and the task allocation proportion among multiple cores is adjusted according to the real-time load condition. According to the method, the wavefront calculation efficiency is remarkably improved, so that the overall bandwidth of the adaptive optical system is greatly improved.

The invention discloses a large-field-of-view high-resolution fundus imaging system and method, and belongs to the field of adaptive optical imagines.The large-field-of-view high-resolution fundus imaging system is characterized in that light of a light source is divided into multiple beams by a beam splitter, and the multiple beams of light enter human eyes through a first spectroscope and an aberration correction module to be reflected by fundus; the multiple beams of reflected light carry respective small-view-field human eye aberrations, pass through the aberration correction module and are reflected by the first spectroscope to reach the confocal detection module; the confocal detection module collects small-view-field human eye aberration of the multiple beams of reflected light to generate multiple image signals and evaluates the aberration according to the quality of the multiple image signals so as to generate feedback signals to drive the aberration correction module to carry out aberration correction on the multiple beams of reflected light at the same time until the system reaches the diffraction limit. The large-view-field high-resolution fundus imaging system can synchronously correct aberration of different view fields, the aberration correction time is shortened, and the overall imaging speed is increased.

In one example, an on-mirror adaptive optics system may include a substrate including a deformable surface, a controller and a plurality of pockets defined in a substrate. Each of the pockets may include a an electrooptical sensor and an actuator. The controller may be communicatively coupled to the electrooptical sensor and the actuator. The controller may be configured to generate control voltages based on signals received from the electrooptical sensor to deform a portion of the deformable surface proximate a corresponding pocket of the plurality of pockets.

The invention relates to an adaptive optical compensation method and system for underwater optical communication. The method comprises the following steps: acquiring real-time transmission data of a light beam in an underwater optical communication network, preprocessing the real-time transmission data, and performing feature extraction to obtain a high-frequency disturbance signal and a low-frequency disturbance signal; then, based on the high-frequency disturbance signal and the low-frequency disturbance signal, a first compensation control instruction and a second compensation control instruction are generated through an algorithm; the first compensation control instruction and the second compensation control instruction are fused and input into a trained compensation model, and an optical transmission quality baseline value in the current environment is obtained; and finally, according to the optical transmission quality baseline value and the historical environmental parameter change trend, adjusting the parameters of the compensation model by means of a gradient descent method, and generating a dynamic compensation scheme. According to the method and the system, the stability, the reliability and the transmission quality of underwater optical communication can be improved, and negative effects caused by signal attenuation and distortion are effectively reduced.

The invention discloses a wavefront correction method for a wavefront detection-free adaptive optical system based on deep learning, and the method comprises the steps: S110, constructing a neural network for training a mapping relation from the light intensity of a far-field focal plane to a group of complex conjugate wavefront data corresponding to the far-field focal plane; s120, constructing a data set, taking the far-field focal plane light intensity as the input of a neural network, and taking a pair of complex conjugate wavefront data corresponding to the far-field focal plane light intensity as the standard output of the neural network; s130, the loss of back propagation is processed by using a loss function, and the network learns one of the determined mapping relationships for the relatively small loss of back propagation; and S140, designing a wavefront correction algorithm of the wavefront-detection-free adaptive optical system for the neural network, and realizing closed-loop correction of the wavefront-free adaptive optical system. According to the technical scheme, the wavefront correction speed and precision are greatly improved, and rapid and high-precision wavefront correction of the wavefront-detection-free adaptive optical system is realized.

The invention, which relates to the technical field of machine vision, discloses a high-precision machine vision positioning system comprising the following components: an adaptive multispectral imaging module, an image data processing module, a positioning calculation module and a storage and control module. The self-adaptive multispectral imaging module is integrated with a visible light imaging subunit, an infrared imaging subunit, an ultraviolet imaging subunit, a self-adaptive optical adjustment subunit and a spectral band self-adaptive switching subunit. Optical parameters are adjusted in real time according to various factors such as ambient light conditions, target distances and target reflection characteristics, so that adaptive switching and cooperative imaging of spectral bands are realized, and the system can acquire comprehensive information of an object in multiple spectral bands; the problems of incomplete information acquisition, image distortion and the like of traditional single visible light wave band imaging in a complex environment are effectively solved.

The present invention relates to a multiphoton imaging system, a multiphoton stimulation system and a multiphoton imaging and stimulation system. The system switches the system optical path to a multiphoton imaging optical path, an adaptive optical multiphoton imaging optical path, a multiphoton optical stimulation optical path or a multiphoton holographic optical stimulation optical path by adjusting the position of an electrically controlled movable reflector. Compared with the prior art, the present invention can achieve high-precision multiphoton imaging and single-cell precision multiphoton optical stimulation.

The invention belongs to the technical field of optical measurement, and particularly relates to a digital holographic microscopy system phase distortion correction method based on adaptive optics, which has the advantage that optical wave complex amplitude information can be acquired in a wide field by means of digital holographic microscopy, and uses a distortion phase image measured by a digital holographic microscopy system as input to correct phase distortion of the digital holographic microscopy system. And iterating the image by using an optimization algorithm until the image evaluation function converges, and obtaining a correction pattern for removing phase distortion. And uploading the correction pattern to a self-adaptive correction device and modulating the wavefront of the light wave to finally realize accurate compensation of phase distortion. The method can provide an in-situ, accurate and rapid phase distortion correction mode and means for the field of optical microscopic imaging, especially the field of quantitative phase imaging.

An adaptive optical module (38A) for a microlithographic projection exposure system (10; 210) is provided, comprising an optical surface (32) for interacting with an exposure radiation (14) of the projection exposure system and several actuators (36n-1, 36n, 36n+1) for changing the shape of the optical surface. The adaptive optical module has a dielectric medium (48) that can be deformed by applying an electrical voltage (68), and each of the actuators includes a separate control electrode (62n-1, 62n, 62n+1), each of which is arranged to generate a respective electric field in a layer of the dielectric medium.The control electrodes are connected to each other by means of a weakly conductive structure (60) with an electrical conductivity of at least 0.1 mS / m and for measuring an impedance (86n-1, 86n, 86n+1) at least one of the actuators is arranged between the control electrode of the actuator being measured and the dielectric medium a measuring electrode (58n-1, 58n, 58n+1).

The present invention discloses a jitter compensation device and method for a high-precision pointing system, which belongs to the field of adaptive optics. The method constructs a proprietary optical device, uses a high-power microscope system to perform jitter analysis on the autocollimated image reflected from the product end, and then drives the rapid rotation of the two-dimensional reflector, thereby realizing jitter compensation of the entire system. The method involved in the present invention can achieve a jitter compensation accuracy better than 0.1″ and a resolution better than 0.01″, thereby providing high-precision test conditions for high-precision sensors.

The invention discloses a self-adaptive optical polishing system and method based on dynamic temperature control and multi-modal detection, and relates to the technical field of intelligent manufacturing and precision machining.The optical polishing system comprises a dynamic temperature control and multi-modal detection module, a flexible execution mechanism, a closed-loop control and optimization module and a cross-scale collaboration and generalization module; the method has the advantages that the temperature field, surface appearance and microcrack signals of a polishing area are monitored in real time through the dynamic temperature control and multi-mode detection technology, thermal deformation is rapidly compensated in combination with a flexible execution mechanism, the problem that the surface shape precision is reduced due to temperature rise in a traditional technology is effectively solved, and the polishing quality is improved. Multi-dimensional defects in the polishing process can be comprehensively captured through cooperative work of the multi-mode sensor, the defect detection sensitivity and the process stability are remarkably improved, and the machining quality of the high-precision optical element is ensured.

Phase-based optoretinography (ORG) is performed using tissue velocity obtained from a modified optical coherence tomography (OCT) system. A swept-source generated OCT A-scan is used to assemble a 2-dimensional profile of the retinal tissue (B-scan) by laterally scanning the imaging beam. These B-scans can be taken at multiple time-points and assembled into a M-scan, which displays motion of a particular cross-section of retinal tissue (after correction for bulk motion). The method captures the changes between each timestep of the M-scan to deduce tissue velocities of the COST, ROST, and IS / OS, among others. An initial contraction of outer segments is detected in outer segments, followed by an elongation shortly thereafter. The tissue velocity measurements can be used to infer retinal dysfunction without the need for costly and computationally expensive scans using adaptive optics.