111 results about "Image Artifact" patented technology

Systems and techniques are provided for automatically analyzing and processing domain-specific image artifacts and document images. A process can include obtaining a plurality of document images comprising visual representations of structured text. An OCR-free machine learning model can be trained to automatically extract text data values from different types or classes of document image, based on using a corresponding region of interest (ROI) template corresponding to the structure of the document image type for at least initial rounds of annotations and training. The extracted information included in an inference prediction of the trained OCR-free machine learning model can be reviewed and validated or corrected correspondingly before being written to a database for use by one or more downstream analytical tasks.

The invention discloses an X-ray image artifact definition automatic correction and enhancement method based on deep learning, and belongs to the technical field of image artifact correction, and the method specifically comprises the steps: obtaining an X-ray image, employing frequency domain decomposition and edge detection to generate an artifact image and a sharpness image, and extracting an imaging parameter set and an anatomical region label; establishing a parameter-driven artifact migration network, inputting an imaging parameter set and an artifact graph, learning an artifact vector field and phase prior, and forming reversible representation of an artifact source; constructing a double-branch decoder, dissecting branches to generate a structural skeleton diagram, and performing cross attention fusion to obtain a candidate correction field; according to the candidate correction field, geometric remapping and sub-band compensation are carried out on the image to obtain an intermediate correction result image, and an artifact vector field is sent to a feedback loop; and the consistency of the sharpness graph and the structural skeleton graph is taken as a loss item, artifact residual constraint is combined, a correction field and decoder parameters are optimized, and non-anatomical textures are suppressed.

A method includes obtaining, using at least one under display camera, one or more first image frames associated with a first diffraction pattern and one or more second image frames associated with a second diffraction pattern. The first diffraction pattern and the second diffraction pattern are related through a transformation. The method also includes generating a first deblurred image using the one or more first image frames and a second deblurred image using the one or more second image frames. The method further includes combining the first and second deblurred images while exploiting complementary types of image artifacts created by the first and second diffraction patterns to generate an image of a scene.

The application discloses a TDI-CCD image deartifact method and device, equipment, medium and product, and relates to the technical field of image processing. The method comprises the following steps: constructing a deep deblurring network; the deep deblurring network comprises a Wiener filter module and a generative adversarial network, and the Wiener filter module is arranged at the input end of a generator of the generative adversarial network; the Wiener filter module is used for filtering a frequency domain image; the deep deblurring network is trained, and the trained deep deblurring network is used as a deartifact model; during the training of the deep deblurring network, a point spread function and a regularization parameter of the Wiener filter module are used as learnable network parameters; and an image with row smear obtained from a TDI-CCD camera is processed by using the deartifact model to obtain a deartifact image. The application can effectively remove image artifacts and improve the definition of a TDI-CCD image.

The application relates to a mode filter, a topology optimization method and an ultrasonic guided wave damage imaging artifact suppression method, relates to the field of ultrasonic detection and imaging, and the topology optimization method comprises the following steps: acquiring initial design parameters, generating an initial population, each chromosome individual in the initial population is coded by a binary logic matrix representing material distribution and meeting design constraints; a population is optimized by using a genetic algorithm, an adaptability function in the genetic algorithm is constructed based on a mode purity index and an energy transmission index, the mode purity index is the ratio of in-plane displacement integral in a transmission area in finite element simulation with the mode filter to total displacement integral, and the energy transmission index is the ratio of relative energy at a transmission end in the finite element simulation with the mode filter to relative energy of an S0 mode; and final optimized material distribution of the mode filter is output. Compared with the prior art, the application has the advantages of effectively suppressing background artifacts, giving consideration to mode regulation performance and substrate structure integrity and the like.

Techniques for generating for training a machine learning model to detect image artifacts include training, based on a first plurality of video frames having synthetic artifacts, a machine learning model to generate a trained machine learning model, generating, based on a second plurality of video frames, a plurality of first artifact detections using the trained machine learning model, selecting, from the second plurality of video frames based on the plurality of first artifact detections, to generate refinement data, and re-training, based on the first plurality of video frames and the refinement data, the trained machine learning model to detect image artifacts in video frames.

The utility model provides an X-ray detector and an imaging system. According to the X-ray detector, the positioning columns are matched with the positioning holes of the positioning sleeves fixed to the circuit board assembly, so that the flicker module and the photoelectric conversion module of the circuit board assembly are positioned and aligned more accurately. Moreover, the structure of the existing circuit board assembly does not need to be greatly changed, and the functions can be realized only by using the additionally arranged positioning sleeve with smaller processing error, so that the cost for developing a new circuit board assembly is saved. In the imaging system, more accurate positioning and alignment are realized by the flicker module and the photoelectric conversion module of the circuit board assembly, so that the signal-to-noise ratio of the X-ray detector for imaging can be improved, the risk of image artifacts is reduced, and better imaging quality is obtained.

The invention relates to the technical field of reconstructed image anomaly detection, in particular to an image data anomaly point detection and correction method and system based on magnetic resonance fingerprints, and the method comprises the steps: generating a simulation data dictionary containing random anomaly points based on a magnetic resonance scanning plane; obtaining a to-be-detected reconstructed image, extracting a reconstructed image signal evolution curve, and matching the reconstructed image signal evolution curve with the simulation data dictionary to obtain an abnormal point position in the to-be-detected reconstructed image; and correcting the image data according to the positions of the abnormal points in the to-be-detected reconstructed image to obtain a corrected image. According to the method, the scanned and acquired data is matched with the pre-generated data dictionary containing the random abnormal points, so that whether the abnormal points are contained or not can be detected, the abnormal points are corrected, related image artifacts are eliminated, reconstructed image anomaly detection can be realized without an AI model, manpower and material resources are saved, the stability is better, and the detection efficiency is improved. Deployment and implementation can be carried out in magnetic resonance imaging equipment.

The application provides an automatic artifact detection and repair method and system based on an endoscope image, constructs an artifact removal model, acquires an original endoscope image, pre-processes the original endoscope image, extracts features of the original endoscope image through a feature pyramid encoder, combines an attention gate module to perform feature screening and feature fusion, acquires an endoscope feature map, inputs the endoscope feature map into a multi-branch decoder, outputs a multi-branch image, reconstructs the multi-branch image through a differentiable rendering layer, acquires a reconstructed endoscope image, acquires a constraint loss according to a multi-level perception constraint mechanism, adjusts parameters of the artifact removal model according to the constraint loss, performs detail enhancement on an intrinsic image to acquire an enhanced image, scores and screens the enhanced image, and acquires a repaired image. Through the technical solution with the multi-artifact decoupling repair function, the accuracy of endoscope image artifact repair is improved.

Deep learning algorithms are applied to electrocardiogram (ECG) traces to more accurately detect, locate, and / or predict true R-waves within the ECG traces. Accurate R-waves may be used to reduce image artifacts due to ECG gating or arrhythmia in real-time imaging. Corrected R-waves may be used for retrospective image reconstruction of cine cardiac MRI by re-binning spatial frequency maps to produce corrected cine MRI with reduced artifacts and uncovered cardiac anatomy.

The application discloses a work robot screen cabinet target detection method and system, comprising: preprocessing a screen cabinet image; constructing a bidirectional network, which is trained based on a loss function comprising a screen cabinet target shape loss and a screen cabinet target position loss; using a first branch to sequentially extract feature maps from the preprocessed screen cabinet image, using a second branch to perform three-layer convolution operation on the preprocessed screen cabinet image with a stride of two to obtain 1 / 8 feature maps; performing first feature fusion on the 1 / 8 feature maps of the first branch and the 1 / 8 feature maps of the second branch, performing second feature fusion on the obtained fusion 1 / 8 feature maps and 1 / 32 feature maps to obtain a to-be-recognized feature map, and obtaining a detection result of the screen cabinet target according to the to-be-recognized feature map. The method avoids the interference of image artifacts or noise, reduces target detection error, has strong anti-interference ability and high recognition accuracy.

The utility model belongs to the field of magnetic resonance imaging, concretely relates to a kind of test cell for artifact quantification measurement in magnetic resonance environment, including test cell body, adjustable positioning mechanism and fixed mechanism;Adjustable positioning mechanism is multi-axis adjusting mechanism, adjustable positioning mechanism sliding assembly in the lateral wall of test cell body;The end of adjustable positioning mechanism into test cell body inside installs sample one, and the spatial position of sample one in test cell body is adjusted by adjustable positioning mechanism;Fixed mechanism is fixedly assembled in the inside bottom surface of test cell body, and sample two is installed on fixed mechanism.Compared with prior art, the utility model solves the quantitative measuring device of the image artifact produced under a group of standard scanning conditions in the test device of artifact evaluation in prior art, and the detection of artifact imaging parameter of different materials in MR is realized in the scheme, to ensure that effective artifact evaluation data is output.

The invention discloses a multi-modal medical data conjoint analysis system based on artificial intelligence, and particularly relates to the field of data analysis. Comprising a data acquisition module, a data quality evaluation module, a data dynamic association module, a heterogeneity feature fusion module, a model real-time optimization module, a data feedback iteration module and a data abnormal feature detection and analysis module. The data acquisition module is used for acquiring multi-modal medical data and executing modal specificity preprocessing on the multi-modal medical data; the data exception feature detection and analysis module is used for constructing an exception association model based on the optimized model and generating an exception association analysis report; according to the method, the problems that in the prior art, the data quality sensitivity is high, and the stability of an analysis result is poor due to the fact that the method is easily influenced by image artifacts, signal noise and text errors are solved by combining modal specificity preprocessing and quality self-adaptive correction with the quality index and the targeted correction formula.

The application provides a medical image three-dimensional reconstruction method and system, and relates to the technical field of medical image processing. The method comprises the following steps: acquiring a multi-layer medical image slice sequence to be processed, performing multi-scale feature analysis on the multi-layer medical image slice sequence, and constructing an initial voxel space distribution matrix; based on the direction consistency and spatial adjacency topology of adjacent voxel feature vectors in the initial voxel space distribution matrix, calculating the spatial potential field strength parameters between each voxel node to obtain an initial spatial constraint topology network; receiving the initial spatial constraint topology network, performing spatial affinity alignment of cross-slice voxel nodes based on the spatial potential field strength parameters, constructing a voxel-level cross-slice feature alignment and spatial interpolation field, performing feature fusion and spatial position interpolation operation on the initial spatial constraint topology network, and obtaining a continuous voxel three-dimensional feature map. The application improves the three-dimensional reconstruction efficiency of medical images, reduces image artifacts, and enhances the visualization effect.

The invention provides an image artifact removal model training method, an image artifact removal method, an image artifact removal device, equipment and a computer readable storage medium. The method comprises the steps that training samples are acquired, and the training samples comprise image samples without artifacts and image samples with artifacts; generating a first artifact-removed sample image with the artifact image sample through an image artifact removal model; performing coding processing on the first artifact-removed sample image through an image artifact removal model to obtain a sample prior information feature representing an artifact-free image sample; calling an image artifact removal model to generate a second artifact-removed sample image based on the sample prior information feature and the first artifact-removed sample image; determining a loss value based on the artifact-free image sample, the second artifact-removed sample image and the artifact-carried image sample; and updating parameters of the image artifact removal model based on the loss value to obtain a trained image artifact removal model. According to the invention, the image artifact removal effect can be improved.

The invention discloses a tail end needle holding clamp for an ablation needle and a surgical robot for performing ablation, and belongs to the technical field of medical instruments. The clamp comprises a base, a sliding platform, a fine adjustment assembly, an anti-spinning locking sleeve, an ablation needle and a tail end reference part. The base is provided with a front end guide seat, and the sliding platform is driven by a lead screw to achieve puncture feeding. The fine tuning assembly comprises an adjusting seat, an adjusting block and a fine tuning screw rod, and is matched with an axial graduated scale and an axial pointer; an anti-counterfeiting reference marking ball is arranged on the needle tip side of the ablation needle, and a circumferential pointer is arranged on a reference piece at the tail end and matched with an annular dial of the anti-spinning locking sleeve to accurately indicate the circumferential rotation angle. The surgical robot comprises a moving trolley, a multi-degree-of-freedom mechanical arm, the clamp and a main control module. Image artifact interference can be reduced, accurate quantitative regulation and control of the ablation needle in the axial direction and the circumferential direction are achieved, and the accuracy of puncture positioning and ablation range control is improved.

The invention discloses a CT sparse angle reconstruction method based on deep learning, and relates to the technical field of industrial image processing and computed tomography, and the method comprises the following steps: S1, data classification and preprocessing: obtaining original CT sinogram data, segmenting the original CT sinogram data through a 128 * 128 pixel sliding window with the overlapping rate of 15%, and obtaining a CT sparse angle; carrying out classification through a three-level classification index system to obtain 12 types of small sinogram blocks with labels; s2, sine domain deep learning data enhancement training: performing interval sampling deletion operation on the classified 12 types of sinogram small blocks, constructing a sample triple, establishing a sine domain generation network of a double-discrimination network structure, and taking the sample triple as training data to obtain a sine domain completion model after training is completed; s3, training and repairing the projection domain repair graph based on the U-Net network; s4, carrying out artifact removal processing on the reconstructed image; the method effectively optimizes the image quality.

The invention provides an active shimming method for TMS-fMRI dynamic compatibility regulation and control, and relates to the field of dynamic electromagnetic compatibility regulation and control in a synchronous transcranial magnetic stimulation and magnetic resonance imaging system. The active shimming device comprises a coil shell, a multi-channel shimming coil is arranged on the side wall of the coil shell, the multi-channel shimming coil comprises a plurality of coil channels, and a uniform magnetic field generated by the multi-channel shimming coil is used for counteracting an interference magnetic field generated by a TMS coil. The active shimming method is adopted, the magnetic field uniformity can be remarkably improved, the problems of image artifacts, geometric distortion, signal non-uniformity, signal loss and the like are effectively restrained, static magnetic field disturbance introduced by a TMS coil is greatly weakened, real-time dynamic compensation based on algorithm control is supported, and the method is suitable for large-scale popularization and application. The method can adapt to the space change of the static magnetic field caused by the TMS coil at different target positions, and solves the problems of poor space flexibility and the like.

The invention relates to the technical field of image artifact recognition, in particular to an image recognition method for cardiovascular intervention postoperative complication risk prediction. The strip-shaped artifact structures are independently distributed in a regular manner and generally show the distinguishing characteristics of high consistency at the angle, and the strip-shaped artifact structures are screened out according to the linear texture distribution consistency condition; calculating the possibility of the artifact structure by combining the complex branch structure of the blood vessel, the sequential change of the diameter of the blood vessel and the characteristic that the contrast agent fluctuates along with blood flow, so as to screen out a tubular artifact structure with relatively low conformity with the blood vessel characteristics; the finally obtained banded artifact structure and tubular artifact structure representing the artifact area are more accurate, and compared with a target detection method, the image recognition processing method based on machine learning is higher in efficiency.

A method for training an image processing system with a machine learning model to perform a virtual multi-angle reconstruction of image stacks recorded with a light-sheet microscope. The method includes: recording at least one light-sheet fine stack comprising multiple image stacks at different illumination angles; determining target outputs from the fine stack, the model, and a classical multi-angle reconstruction; determining learning inputs from a light-sheet coarse stack comprising one or more image stacks at different illumination angles using the model and the classical reconstruction; creating an annotated dataset of the target outputs and learning inputs; and optimizing the model for the virtual multi-angle reconstruction based on the dataset. The light-sheet coarse stack has fewer images than the fine stack, and a virtual reconstruction produced by the trained model exhibits fewer image artifacts than a reconstructed image stack computed from the same coarse stack by the classical multi-angle reconstruction.

The invention relates to a multi-modal medical image artifact intelligent detection method and system based on deep learning, and the method comprises the steps: receiving a multi-modal original medical image, and carrying out the preprocessing of the original medical image, so as to generate a standardized target medical image; performing global detection based on the target medical image to obtain artifact confidence; when the artifact confidence exceeds a set threshold value, identifying a corresponding artifact type according to a preset multi-classification network, performing pixel-level segmentation positioning on the target medical image based on the artifact type to extract a main artifact area and position, and calculating a severity score according to the main artifact area; according to the method, the interpretable thermodynamic diagram is generated based on the target medical image, and the key features in the interpretable thermodynamic diagram are extracted to generate the structured detection report, so that the problems of dependence on manpower and high modal specificity in the prior art are solved.

An electronic device may include an electronic display to display frames of image data and processing circuitry to determine a display strain transition event of the electronic display and change a refresh rate of the electronic display based on the determination of the display strain transition event. The electronic device may temporarily increase the refresh rate based on determining that the display strain transition event includes folding or unfolding the electronic display. The electronic device may sequence down the refresh rate to a lower refresh rate after a period of time. As such, perceivable image artifacts in displayed image content caused by folding or unfolding the electronic display may be reduced or eliminated.

The invention provides a deep learning semiconductor defect microscopic super-resolution method based on physical guidance, and the method comprises the steps: obtaining an accurate point spread function through a microscopic imaging system, building an accurate mathematical degradation model through the point spread function, and calculating the difference between an image and a condition image based on the degradation model. And the obtained gradient information is used for restraining diffusion model image prediction. And finally, gradient information is used as a guide item to be introduced into backward diffusion in the denoising diffusion implicit model, and an offset item is introduced into mean value prediction in the diffusion process of each step so as to prevent the image from generating artifacts and non-existing information.

The invention discloses a single-light-source energy spectrum imaging device, and relates to the field of medical science and technology, the single-light-source energy spectrum imaging device comprises a ray light source, a detector assembly and a shooting grid assembly, the detector assembly and the shooting grid assembly are both arranged in the emitting direction of the ray light source, and the detector assembly is used for synchronously collecting at least two types of ray image signals in single exposure of the ray light source; the at least two types of ray image signals correspond to different energy ranges; the emission grid assembly is located between the ray light source and the detector assembly and comprises an enhanced anti-scattering grid, a micropore array is arranged on the enhanced anti-scattering grid and used for allowing rays to pass through, the enhanced anti-scattering grid comprises a barrier layer and an absorption layer, and the absorption layer is arranged on the surface of the barrier layer in a composite mode. The absorption layer is used for absorbing the characteristic rays generated by the barrier layer. The technical problems that in existing single-light-source single-exposure dual-energy imaging, scattered rays are prone to polluting detector signals, energy spectrum data distortion is caused, and image artifacts are serious are solved.

This invention relates to the field of visual inspection technology for electroacoustic devices, and discloses a method and system for quality inspection of horn drum paper based on visual inspection. The method separates a base layer, a detail layer, and a noise layer through multispectral image fusion and multi-scale decomposition. A three-dimensional mesh model of the horn drum paper is reconstructed based on the base layer, and the detail layer is mapped to the surface texture of the model, realizing the correlation analysis of defects in three-dimensional morphology and texture features, thus improving the ability to identify minute three-dimensional defects. By analyzing the local energy distribution of candidate defect regions in the noise layer and combining it with the local curvature of the mesh for authenticity judgment, it can accurately distinguish between real material damage and imaging artifact interference, reducing the false detection rate. Finally, repair instructions are generated based on the determined defect set. This method overcomes the limitations of two-dimensional image detection, has higher accuracy and reliability, and is suitable for automatic quality inspection on production lines.

The present application relates to image correction technology, disclose a kind of CBCT image bone hardening artifact correction method, comprising: the high-energy projection image of target object and low-energy projection is collected and is calculated after image reconstruction to obtain preliminary artifact-free image, based on threshold segmentation from preliminary artifact-free image extraction bone image and soft tissue image and calculate skeleton weight, artifact image of target object is used as training input data, bone image and soft tissue image are used as reference data, and dual-channel artifact correction model is trained, and preprocessed bone image and preprocessed soft tissue image are output according to the bone hardening artifact image to be processed according to dual-channel artifact correction model, according to the preprocessed bone image and preprocessed soft tissue image are weighted fusion according to skeleton weight, and artifact-free reconstruction image is obtained.The present application also proposes a kind of CBCT image bone hardening artifact correction device, electronic equipment and storage medium.The present application can improve the effect of image artifact correction.

The invention discloses a medical image focus automatic identification system based on deep learning, and relates to the technical field of medical image intelligent analysis. The system supports the access of CT, MRI and X-ray equipment, adapts to various image formats and carries out calibration to keep the physical significance consistent; processing image artifacts by adopting a denoising-normalization-cutting three-stage process, and unifying the size of the region of interest; a multi-branch CNN and Transform hybrid model is adopted, and based on large-scale annotation data set training, local and global features are extracted; outputting the category, size, position and malignant risk level of the focus; a structured report, DICOM (Digital Imaging and Communications in Medicine) labeling and visual display are supported; regularly updating the model by adopting a hierarchical storage architecture; monitoring a system operation index, and triggering an alarm when the index is abnormal. The image processing consistency and the focus identification accuracy are improved, the risk judgment is optimized in combination with clinical information, the data privacy and the system quality are guaranteed, and the diagnosis efficiency and reliability are high.

The invention relates to a mode filter, a topological optimization method and an ultrasonic guided wave damage imaging artifact suppression method, and relates to the field of ultrasonic detection and imagines.The topological optimization method comprises the steps that initial design parameters are obtained, and an initial population is generated; each chromosome individual in the initial population is formed by coding a binary logic matrix which represents material distribution and meets design constraints; a genetic algorithm is adopted for population optimization, a fitness function in the genetic algorithm is constructed based on a mode purity index and an energy transmission index, and the mode purity index is a ratio of a transmission region in-plane displacement integral to a total displacement integral under finite element simulation added with a mode filter; the energy transmission index is a ratio of relative energy of a transmission end under finite element simulation added with a mode filter to relative energy of an S0 mode; and outputting the final optimized material distribution of the mode filter. Compared with the prior art, the method has the advantages that background artifacts are effectively inhibited, and the mode regulation and control performance and the matrix structure integrity are both considered.

The invention discloses a deep seabed weak illumination image enhancement method based on non-uniform illumination correction. The deep seabed weak illumination image enhancement method comprises the following steps: designing a non-linear guide filtering algorithm to realize balance of a bright region and a dark region of an underwater image; an adaptive MSRCR algorithm is adopted to eliminate artifacts of the underwater image; and carrying out multi-scale image fusion on the basis of the Retinex model to obtain an enhanced clear image. The method has good adaptability in the underwater environment with non-uniform illumination and low illumination, the influence of non-uniform illumination on the image can be corrected, the brightness and color of the underwater image can be effectively recovered, artifacts of the image can be adaptively eliminated, and technical support is provided for high-quality visual perception of a deep sea exploration sampling vehicle.