482 results about "Imaging condition" patented technology

The invention provides a mobile phone glass quality detection method and system based on image detection, and relates to the technical field of image defect detection, and the method comprises the steps: configuring an imaging condition for collecting a mobile phone glass image, and collecting an original image of mobile phone glass according to the configured imaging condition; constructing a transparency enhancement algorithm according to a transparent background existing in the collected original image, and processing the original image according to the transparency enhancement algorithm to obtain an enhanced image; a crack boundary self-enhancement step is introduced, micro cracks and fuzzy scratches existing in an enhanced image are positioned, the enhanced image is marked according to the positioning result of the micro cracks and the fuzzy scratches, and a marked image is obtained. The phenomenon of false detection or missing detection easily occurring in mobile phone glass detection under a transparent background in related technologies is solved through a transparency enhancement algorithm; and the problem that boundary fuzzy defects such as microscopic cracks and fuzzy scratches with extremely low contrast are difficult to identify is solved through the crack boundary self-enhancement step.

The invention relates to a high-resolution remote sensing image change detection method and system based on decoupling representation learning, and solves the defects of false change and non-ideal detection precision caused by inconsistent imaging conditions of different time phases compared with the prior art. The method comprises the following steps: obtaining a training sample; constructing content branches; constructing style branches; training a multi-branch parallel framework; obtaining a high-resolution change result; and separating image styles and obtaining a recovery result. According to the method, a reliable high-resolution remote sensing image change detection result can be obtained under the condition of multi-temporal complex imaging condition interference.

The invention discloses a mold defect visual inspection method, particularly relates to the technical field of industrial machine visual inspection, and is used for solving the technical problem of image spatial variation blurring caused by mechanical vibration under a mobile shooting condition. The method comprises the following steps: acquiring a to-be-detected image on the surface of a mold, analyzing the gradient magnitude of each region, determining the fuzzy characteristics of different regions in the to-be-detected image according to the difference of the gradient magnitudes, evaluating the expected confidence of each region for executing the deblurring operation based on the fuzzy characteristics, and executing the deblurring operation on the regions to obtain a preliminary restored image; an artifact index is calculated in the uniform background area of the preliminary restored image, the distribution concentration degree of image components in each local feature area in the frequency domain is analyzed in the preliminary restored image, and the distribution concentration degree is compared with a preset defect judgment threshold value adjusted according to the artifact index; judging whether the corresponding local feature region is a defect region or not; accurate recognition of mold surface defects under complex imaging conditions is realized.

The invention discloses a process casting image surface defect automatic identification method based on machine vision, which comprises the following steps: acquiring an original image of a process casting, recording an imaging parameter set, and preprocessing to obtain a standardized image; performing homomorphic filtering to obtain an enhanced image and an illumination component, and inputting the enhanced image into an improved OfficientAD algorithm to generate an abnormal score map and a feature residual map; updating homomorphic filtering parameters according to the abnormal score map, the feature residual map and the illumination component, and performing homomorphic filtering for three times to obtain an enhanced image; inputting the enhanced image into an improved OfficientAD algorithm to obtain an abnormal score graph; and performing threshold segmentation and connected domain extraction on the abnormal score graph to obtain a defect candidate region, extracting texture statistics and residual statistics, and matching a defect type feature library to output an identification result. According to the method, stable and consistent automatic identification is realized under complex texture and imaging conditions, and the manual visual inspection cost is reduced.

A computer-implemented method includes converting by a pose tokenizer, based on a learned codebook, pose parameters of a body into a sequence of discrete pose tokens; randomly masking a portion of the sequence of discrete pose tokens; predicting the randomly masked sequence of discrete pose tokens based on multi-scale features extracted from a monocular image by an image conditioned masked transformer; optimizing the sequence of discrete pose tokens by aligning a re-projected three-dimensional (3D) pose with an estimated two-dimensional (2D) pose; directly regressing, from the multi-scale features, a shape parameter of the body and a weak perspective camera parameter; and generating a 3D mesh reconstruction of the body based on the shape parameter and the weak perspective camera parameter.

The invention discloses a method and device for detecting color reducibility of a mobile phone camera and a medium, and relates to the technical field of image quality detection.The method comprises the steps that after geometric alignment is conducted on a color card image set, color block areas are divided, and a cross-light-source color block observation matrix is obtained through color block feature extraction; converting the cross-light-source color block observation matrix into a target color space coordinate, and performing illumination correction to obtain a cross-light-source color coordinate matrix; inputting the cross-light-source color coordinate matrix into a deep learning model, outputting a color block level high-dimensional color embedding matrix, and obtaining a high-dimensional color embedding vector set corresponding to the tested mobile phone camera in combination with an attention weight; calculating a color reducibility error index based on the high-dimensional color embedding vector set in combination with the standard reference library; and performing qualification judgment according to the color reducibility error index to generate a qualification detection conclusion. According to the invention, a stable and reliable technical basis is provided for camera quality detection under a multi-light-source imaging condition.

The invention discloses a radar point cloud-based train body modeling method and device and a related medium, and the method comprises the steps: obtaining an original point cloud frame sequence, and carrying out the parameter synchronous mapping of the original point cloud frame sequence, and obtaining a standardized point cloud sequence; performing data preprocessing on the standardized point cloud sequence to obtain a preprocessed point cloud sequence; performing time sequence registration and continuous modeling by using the preprocessed point cloud sequence to generate continuous vehicle body characterization; performing semantic segmentation and region extraction on the continuous vehicle body representation to obtain a door region set; extracting morphological characteristics of the vehicle body based on the door area set to obtain a structured parameter set with a door area state label; and performing formatting packaging on the structured parameter set, and outputting vehicle body modeling parameters. According to the method, three-dimensional modeling is carried out on the train, and the feature state is extracted, so that train state identification can be stably completed under different illumination and shielding conditions without depending on external identification and imaging conditions, and the positioning accuracy is improved.

A method for training a generative image model includes receiving training data having training images and image captions and annotating the training data resulting in annotated training data having the training images, the image captions, and moderation labels, such that each training image is associated in the annotated training data with a corresponding image caption and at least one corresponding moderation label. The method further includes performing a training process to train a generative image model using the training images, the image captions, and the moderation labels, resulting in a trained model that generates images conditioned to both the image captions and the moderation labels.

The invention discloses a welding seam defect X-ray image intelligent diagnosis and credible traceability method based on multi-modal heterogeneous information collaboration, which is characterized in that a defect analysis network fusing multi-domain feature modeling and graph structure expression is constructed on the basis of bimodal data formed by a welding seam X-ray image and an industry detection standard text. In the image mode, dividing the weld seam image into a plurality of local area units through superpixel segmentation, taking the areas as image nodes, respectively extracting spatial domain, frequency domain, wavelet domain and edge domain features, and constructing a weighted graph structure by combining the spatial adjacency relation and the feature similarity relation between the areas; realizing overall modeling and correlation analysis of weld defect structure information by using a graph convolutional network; in a text mode, feature coding is carried out on an industry detection standard text, and the feature coding is used as an important prior constraint for defect judgment. Collaborative modeling of an image detection result and standard semantic information is achieved through a gating fusion mechanism, a mapping relation between a detection conclusion and a standard term is established, and interpretable expression and result credible traceability of the weld defect diagnosis process are achieved. And a welding seam X-ray film automatic digital acquisition and observation device is adopted in a matched manner, so that stable transmission, positioning observation and high-resolution digital imaging of the industrial ray film are realized, and reliable and consistent image data input is provided for the intelligent diagnosis method. The method is suitable for intelligent defect detection under complex welding seam structures and multi-working-condition imaging conditions, and has high engineering application value and popularization prospect.

The invention discloses an ISAR image ship coarseness identification method and system based on scattering feature estimation, and relates to the technical field of target identification, and the method comprises the steps: constructing and training a target identification model; the target recognition model comprises a small grazing angle image feature enhancement module based on scattering characteristic estimation, a feature extraction fusion module based on motion attitude decoupling and a full connection module; the small grazing angle image feature enhancement module comprises a scattering center parameter estimation module, a complex waveform feature extraction module and a physical data feature fusion module; the feature extraction fusion module comprises a spatial frequency domain attitude mode representation module, an attitude adaptive deformation convolution module and a frequency domain space double-attention fusion module; obtaining an ISAR image of a to-be-identified marine ship; and inputting a to-be-identified marine ship ISAR image into the trained target identification model for ship type identification. According to the invention, the identification precision of the ship type under the conditions of small grazing angle and complex motion imaging is improved.

A method for image generation includes receiving a training data that includes training images, image captions each corresponding to one of the training images, and contextual features each associated with one or more of the training images. The method further includes performing a training process to condition a generative image model using the training images, the image captions, and the contextual features, resulting in a conditioned model that generates images conditioned to both the image captions and the contextual features.

The invention provides a medical magnetic resonance image reconstruction method and system, and belongs to the technical field of image reconstruction, and the method comprises the steps: obtaining sampling coding and under-sampling k space data; processing the under-sampled k space data through inverse Fourier transform, and converting the k space data into an image domain to obtain under-sampled image data with blurring or artifacts; carrying out preliminary restoration on the undersampled image data based on sampling coding, further applying data consistency operation on the image to obtain a rough image, and taking the rough image as an image condition vector after underspace coding; acquiring a text serving as a cue word, inputting the text into a text encoder, and encoding the text into a high-dimensional semantic embedding vector by the text encoder; and inputting the high-dimensional semantic embedding vector and the image condition vector into a reconstruction model based on the correction flow to predict a full-sampling MRI image, and generating a final reconstructed MRI image.

According to one embodiment, an X-ray imaging system includes X-ray irradiator irradiates X-rays to an object, X-ray detector detects the X-rays transmitted through the object, and processing circuitry. The processing circuitry is configured to acquire a camera image including an examination portion of an object. The processing circuitry is further configured to input a prompt including at least the camera image into a generative model to acquire a virtual medical image from the generative model. The processing circuitry is further configured to set a main imaging condition, which is an imaging condition used for a main imaging, based on the virtual medical image, and to output the main imaging condition to a medical imaging apparatus. According to other embodiments, the processing circuitry of an information processing apparatus controls the output of the imaging condition to a medical imaging apparatus.

The invention relates to the technical field of engine cylinder body and cylinder cover surface defect detection, and discloses a visual detection method for engine cylinder body and cylinder cover surface defects, which comprises the following steps of: acquiring engine cylinder body and cylinder cover surface images under various different imaging conditions, preprocessing, extracting a feature vector of each pixel block, distributing weights by using a weight function, and calculating the surface defects of the engine cylinder body and cylinder cover. The method comprises the following steps: extracting a suspected defect area, introducing a loss function for verification, constructing a causal graph, calculating the probability of each defect type through causal inference, providing a decision basis for final output decision, and finally outputting a defect result. By integrating multi-angle illumination, multi-spectral imaging and laser scanning systems, rich information of the surface under different physical characteristics can be obtained, and it is ensured that various types of defects can present high contrast characteristics under at least one imaging mode.

The invention discloses a license plate character segmentation and comparison identification method, and relates to the technical field of license plate character segmentation and comparison identification, and the method comprises the steps: carrying out the perspective transformation based on the four-corner coordinates of a license plate image, and obtaining a license plate flattening image; reading gray scale percentile positions on the flattened image according to rows, calculating a bright side half width and a dark side half width, determining an unbiased boundary and a character segmentation seam in combination with gradient direction marks, and cutting character image blocks; constructing a single-side index trailing convolution kernel by half widths of two sides, carrying out line-by-line one-dimensional convolution on the standard template to generate a template appearance, selecting candidate characters by adopting normalized cross-correlation, and obtaining a confidence coefficient in combination with a pixel residual error ratio; according to the invention, stable segmentation and high-precision identification can be realized under complex illumination and imaging conditions.

The invention relates to the technical field of workpiece alignment, in particular to a machined workpiece automatic alignment method and system based on image measurement, and the method comprises the following steps: collecting a workpiece image, analyzing a local curvature change rate, adjusting sampling density, reconstructing surface features, generating micro-topography data, calculating a stability score, and outputting an attitude correction parameter. And tracking a time sequence deviation derivation instruction response deviation track, fitting a deviation curve to invert delay mapping to generate an error correction table, and mapping a control signal to adjust a step length to generate an adaptive driving instruction. According to the method, the morphological stability score is calculated through the ratio of the boundary curvature continuity to the area compactness of the equal-optical-density region, accurate judgment on the boundary contour under different image conditions is achieved, the response offset function is constructed with the help of the least square method, the execution error correction table is mapped, the response precision and execution consistency of the alignment instruction are improved, and the accuracy of the alignment instruction is improved. And the alignment efficiency and the posture stability of the workpiece under the complex morphology condition are improved.

The invention discloses a method and a system for simplifying large model reasoning service deployment in a heterogeneous cluster environment, mainly relates to the technical field of service deployment, and is used for solving the problems of repeatability cost of heterogeneous adaptation, dependence on manual allocation of containers to specific nodes during containerization deployment and version library compatibility in an existing scheme. And crash is caused, and dynamic resource allocation based on load characteristics is lacked. Comprising the steps of performing preset mirror image condition matching on a node hardware feature tag and a model service demand tag of an implementation node and a pre-built reasoning service container mirror image in a mirror image warehouse; when the pre-constructed reasoning service container mirror image with the matching degree of 100% does not exist, dynamically constructing a customized reasoning service mirror image, and storing the customized mirror image in a mirror image warehouse as the pre-constructed reasoning service container mirror image for implementing node matching; and pulling a pre-constructed reasoning service container mirror image for implementing node matching from the mirror image warehouse, mounting a standardized model file mirror image, and starting a reasoning service container.

A radiation image processing apparatus includes a hardware processor that obtains radiation images of a subject captured under imaging conditions different from one another, and an image processor that corrects a pixel value of at least one radiation image of the radiation images, and combines the radiation images to obtain one composite image. Each of the radiation images has an image overlap area with one or more other radiation images of the radiation images. The image processor corrects the pixel value of the at least one radiation image such that a maximum value of the pixel value of the at least one radiation image is equal to or smaller than a maximum value of an image gradation.

The invention discloses a CAD (Computer Aided Design) program generation method based on a multi-view freehand sketch and a cross-modal Transform, which comprises the following steps: acquiring freehand sketch images of three views, and preprocessing each freehand sketch image; the preprocessed freehand sketch images are used as input of a trained CAD program model; according to the method, by designing a CAD program model comprising a visual encoder network, a fusion module, a cross-modal alignment module and a CAD sequence generator, the structural integrity and geometric consistency of freehand sketch image modeling are effectively improved, the generated CAD program sequence conforms to parametric modeling logic, and the reliability and accuracy of a modeling result are improved; according to the method, hierarchical position coding vectors and hierarchical constraint masks are added in a CAD program model training stage, the stability of a generation result in the aspects of structural rationality, geometric consistency and performability is improved, and the method is suitable for freehand sketch input under different drawing habits and imaging conditions and has good robustness and universality.

A processing system includes: a processing apparatus for processing an object by irradiating the object with an energy beam to form a melt pool on the object and supplying a build material to the melt pool; an imaging apparatus for imaging the melt pool to generate a melt pool image; and a control apparatus for generating melt pool image information based on the melt pool image and for controlling the processing apparatus based on the melt pool image information so that a size of the melt pool is a target size, the melt pool image information is generated based on a plurality of melt pool images, the control apparatus changes, based on a processing condition of the object, an imaging condition for the imaging apparatus imaging the melt pool.

The invention relates to the technical field of liquid level reading identification, and discloses a liquid level meter reading identification method based on computer vision. The method comprises the following steps: acquiring and preprocessing an image, acquiring a digital image through image acquisition equipment, and executing normalization, noise removal and contrast enhancement; feature extraction and weight generation: extracting features such as a scale line position and a pointer angle from the preprocessed image, and generating a reading preference weight vector; performing parameter mapping and optimization, mapping the weight vector into an identification control parameter, referring to a predefined parameter anchor point in an offline optimization database, and performing interpolation calculation to adapt to a current image condition; and reading calculation and output: executing a liquid level value calculation algorithm by using the optimized parameters and outputting a result. The method adapts to different image environments through multi-stage cooperation, the adaptability and stability of liquid level reading recognition are improved, and the method is suitable for automatic reading scenes of various liquid level instruments.

The invention discloses a crystal grain detection and analysis method and related equipment, effective crystal grains are obtained through corrosion expansion pretreatment denoising, OfficientNet and bidirectional feature pyramid network detection, contour obtaining through region segmentation and three-step filtering, qualification is judged in combination with a threshold value, a report is generated, and the existing technical problems are solved in a targeted mode. On the aspect of reducing calculation complexity, preprocessing reduces model data volume and interference, the model abandons complex modules, parameters are few, calculation cost is low, hardware occupation can be reduced, and detection and training time can be shortened; on the aspect of improving generalization ability, features are preprocessed and purified, interference of imaging conditions and material types is reduced, the number of layers of the model is simplified, multi-scale fusion ability is achieved, and detection in different fields can be adapted without optimization of a specific data set; on the aspect of reducing data dependence, invalid and pseudo crystal grains are accurately removed through three-step filtering, defects of a small-scale or low-quality data set are cooperatively made up, dependence on a high-quality large-scale data set is greatly reduced, and robustness is improved.

The invention discloses a viscoelastic wave reverse time migration imaging method based on finite difference algorithm solution, and the method comprises the steps: S1, combining with the complex modulus of a constant Q model, and deriving a viscoelastic wave attenuation equation suitable for finite difference algorithm solution through introducing weighting function approximation; s2, establishing a viscoelastic wave compensation equation by introducing a compensation item for compensating energy attenuation when seismic waves are propagated in a viscoelastic medium, and combining the wave field compensation item with low-pass filtering for calculation to realize stable compensation of a wave field; s3, performing longitudinal and transverse wave field separation on the seismic source wave field and the detector wave field; s4, applying a seismic source normalization cross-correlation imaging condition to the separated wave field to obtain an attenuation compensation reverse time migration imaging result; and S5, setting a seismic wave excitation source and a signal receiving device for wave field data acquisition. The method provided by the invention can effectively improve the precision and resolution of seismic imaging in the viscous medium.

The invention discloses a dried orange peel year identification method based on multi-scale deep feature fusion and meta-learning, and relates to the technical field of medicinal material year identification and quality monitoring. Aiming at the problems of changeable imaging conditions, inconsistent image quality and the like of different devices, a deep learning model combining multi-channel feature interaction and meta-learning is provided, and a pericarpium citri reticulatae year classification framework suitable for multi-source image data is constructed. The method comprises the following steps: (1) sample preparation and multi-device image acquisition; (2) partitioning and preprocessing the complete image, the peel surface layer and the inner capsule image; (3) deep network training based on three-channel multi-scale feature extraction and fusion; and (4) realizing rapid adaptation and generalization of the model to different equipment domains by introducing a meta-learning mechanism. According to the method, a pericarpium citri reticulatae year recognition model suitable for various image sources is created, the generalization ability of the model is improved, and certain scientific basis and reference significance are provided for pericarpium citri reticulatae year recognition marketization.

Methods, systems, and apparatuses, including computer programs encoded on computer storage media, for training an image generation neural network and, once the image generation neural network is trained, generating new output images using the image generation neural network. In particular, the described techniques include obtaining a training data set that includes training examples that each include a training conditioning image and training target image that has been identified to being semantically similar to the training conditioning image. Then training, on the training data set, an image generation neural network that is configured to generate an output image conditioned on a conditioning image. By using the described techniques to train an image generation neural network the system achieves high quality image generation that can be used to generate new output images semantically similar to a conditioning image without the need to fine-tune the image generation neural network to a specific subset of semantic attributes.

The invention provides a diffusion converter method for realizing material migration only by using an image condition, which comprises the following steps of: acquiring a target image and a material image, generating an illumination image and a depth image, and converting the illumination image and the depth image into a unified image mark through a shared potential space encoder; and performing unified processing through a multi-modal attention mechanism of a diffusion converter, realizing cross-modal interaction in combination with cross deviation modulation, enhancing depth control by using a low-rank adaptive module, performing background maintenance, mixing and fusing a foreground and a background, and finally generating an output image inheriting a target structure and a material texture. According to the MaTe framework of the method, input images are integrated at the token level, potential space is shared, multi-modal attention unified processing is carried out, dependence on an adapter, ControlNet and the like is eliminated, text prompt and model fine adjustment are not needed, zero-shot high-quality material generation is achieved, framework complexity is reduced, efficiency and quality are improved, and the method is suitable for the fields of digital creation, industrial design and the like.

An information processing apparatus comprising: a memory storing instructions, and at least one processor configured to execute the instructions to: perform time synchronization processing for a plurality of video signals obtained by capturing a subject by a plurality of cameras in a space and a plurality of acoustic signals collected by a plurality of microphones in the space; calculate position information of the subject using the plurality of video signals and imaging conditions of the plurality of cameras; separate a sound source signal of the subject from the plurality of acoustic signals having undergone the synchronous processing; and generate an acoustic signal for listening, at an arbitrary listening position in the space, the sound source signal generated from the subject, by assigning the sound source signal of the subject to a position corresponding to the position information of the subject.

The signal processing device 10 includes a three-dimensional information with reliability reconstruction unit 11 which generates three-dimensional information with reliability including three-dimensional information constructed with estimated values of intensity and a phase at a three-dimensional position in a steady state reconstructed using an observed SAR image, and information indicating reliability of the three-dimensional information, and a simulated SAR image generation unit 12 which generates a simulated SAR image which is a complex image representing the steady state suitable for an imaging condition of a SAR image to be analyzed, using the three-dimensional information and the imaging condition of the SAR image to be analyzed, and calculating reliability information representing the reliability of the simulated SAR image.

A method or system for enhancing vehicle identification accuracy. The system identifies a gap in a training dataset for training a license plate identification model. The gap represents underrepresented visual characteristics in misidentified license plates. A guidance prompt is generated based on the visual characteristics of the misidentified license plate. Condition embeddings are then generated from the guidance prompt, which are used to condition a diffusion model to create synthetic license plate images. The diffusion model is trained to receive a real license plate image, encode it into a vector, apply forward diffusion to add noise, and then apply reverse diffusion to remove the noise, resulting in a denoised vector that represents a synthetic license plate image conditioned by the condition embeddings. The denoised vector is then decoded to produce the synthetic license plate image. The synthetic images are then used to retrain the license plate identification model.

A processor acquires two radiation images with contrast based on a first characteristic corresponding to a first imaging condition. It derives a body thickness distribution of a subject from at least one of the images and removes scattered ray components from both images using the first characteristic. A soft part image and a bone part image representing respective regions of the subject are derived from the two images. These images, based on the first characteristic, are converted into soft and bone part images with contrast based on a second characteristic corresponding to a second imaging condition, using the body thickness distribution and both characteristics. Finally, the processor adds the converted soft and bone part images to derive a processed radiation image with contrast based on the second characteristic.