1080 results about "Image synthesis" patented technology

The invention discloses a metal surface defect image controllable synthesis method based on a pre-training diffusion model, belongs to the technical field of metal surface defect data enhancement, and is a multi-granularity prompt generation scheme aiming at the problems of insufficient defect data and high labeling cost in the prior art. The method comprises the following steps: performing multi-precision labeling on a metal surface defect image, and constructing a multi-granularity training set; the method comprises the following steps: designing a precision coding mask pyramid based on a pre-trained Stable Diffusion model, generating a hierarchical control signal through multi-stage downsampling, fusing category and position information by combining CLIP semantic coding, and performing hierarchical injection to denoise UNet so as to control defect generation; a VAE module and a CLIP module are frozen by adopting a transfer learning strategy, and only a denoising network and a mask coding layer are finely tuned; and a user generates various defect images through the coarse-grained mask and category prompt. The method breaks through the dependence of a traditional generation model on fine labeling, realizes flexible control of defect positions, shapes and categories, and improves the generalization ability of a detection system.

A method and apparatus for generating photorealistic synthetic images by receiving multiple forms and multiple instances of user input corresponding to a user's visual idea, executing an iterative image search to identify pre-existing images semantically aligned with the user's visual idea, and using an image synthesis deep learning model to generate at least one synthetic image based on the multiple forms and instances of user inputs.

The invention provides an industrial product defect automatic classification method and system, and the method comprises the steps: collecting original industrial product defect image data, and constructing a labeled image sample set and an unlabeled image sample set; constructing a training image sample set based on the labeled image sample set and the unlabeled image sample set in combination with a plurality of image synthesis strategies; based on the training image sample set, introducing a transfer learning strategy and fusing an attention mechanism, and constructing and optimizing an industrial product defect classification model; performing semi-supervised joint training and online learning based on the training image sample set and the real-time small-batch image sample set; and constructing an industrial product defect identification log based on the real-time image flow sample set, the classification model parameters and the corresponding classification prediction function. On the basis of multi-strategy image enhancement and semi-supervised training, transfer learning and a channel attention mechanism are fused, expansion of industrial product defect image samples and fine defect identification are achieved, and the method is suitable for an intelligent defect detection system in various industrial manufacturing fields.

To implement image synthesis without loss of surface information of SE images and shadow information of BSE images, the present disclosure proposes image processing techniques include applying data of a first quality image (low quality image) to a trained model, estimating a structural feature and a material feature of a second quality image (high quality image) corresponding to the first quality image, calculating at least one shadow datum based on the structural feature and a synthesis parameter and calculating at least one gradation datum based on the material feature and a synthesis parameter, generating a synthesized image from the at least one shadow datum and the at least one gradation datum, and outputting the synthesized image as a prediction result of the second quality image (see FIG. 8).

The invention relates to the field of intelligent detection, and provides a machined surface quality detection system and method based on microscopic image.The method comprises the steps that firstly, a multi-focal-plane image sequence of a target product sample is obtained, and a full-clear image of the target product sample is generated through super-depth-of-field image synthesis; then preprocessing and calibrating the full-clear image to obtain a standardized image of the sample, then analyzing and measuring the surface characteristics of the sample based on the standardized image to generate structured data of the product processing surface quality, and finally determining the quality of the product according to a preset quality standard. And processing the structured data to form a product processing surface quality evaluation report. In this way, the comprehensiveness and accuracy of quality detection of the machined surface can be effectively improved.

The invention provides a space-spectrum fusion high-standard farmland utilization mode remote sensing monitoring method and system, and relates to the field of remote sensing monitoring, and the method comprises the steps: carrying out the multi-temporal image synthesis and preprocessing of a remote sensing image set, and obtaining a key phenological period synthesis image set, the NDVI value of the research area in each key phenological period is calculated through the key phenological period synthetic image set; phenological characteristic analysis is performed on different farmland utilization modes based on each NDVI value, farmland utilization priori knowledge is combined to construct a cultivated land utilization map, and spectrum information of the cultivated land utilization map is acquired; performing multi-scale segmentation on the remote sensing image set to obtain field images with spatial information; and assigning spectrum information of the farmland utilization map to the field images with spatial information, and judging the farmland utilization mode of the research area. According to the invention, the spectrum information of the cultivated land utilization map and the field images with spatial information are fused, the farmland utilization mode of each field is visually displayed, and remote sensing monitoring of the farmland utilization mode at the land parcel level is realized.

The invention provides a display method and electronic equipment, and relates to the field of terminals. The invention provides a display method which is applied to electronic equipment, the electronic equipment is provided with an intelligent table setting application, and when the intelligent table setting application is started, an image synthesizer is indicated to start a display strategy of a filtering layer; when the image synthesizer receives the first image, if it is detected that the first image belongs to the image sent by the intelligent table, the first image is allowed to be displayed; otherwise, the first image is not allowed to be displayed. By adopting the method provided by the invention, when the electronic equipment starts intelligent table setting, the problem of splash screen of the electronic equipment is avoided, and the display quality of a display screen is improved.

The invention relates to the technical field of three-dimensional reconstruction, in particular to a lightweight three-dimensional reconstruction method and system based on a three-dimensional Gaussian model, and the method comprises the steps: constructing the three-dimensional Gaussian of a target scene based on a multi-view original image of the target scene through employing a 3DGS algorithm; identifying the minimum resolution of each Gaussian primitive in the three-dimensional Gaussian, setting a spherical region according to the minimum resolution, taking the intersection result of the spherical region and the Gaussian primitive as a region evaluation score, and trimming the redundancy region geometry in the three-dimensional Gaussian in combination with the opacity; and taking the maximum zoom scale of the Gaussian primitive as a spatial density proxy variable, and dynamically distributing spherical harmonic coefficient orders according to a local spatial density proxy variable so as to meet the rendering requirements of different three-dimensional Gaussian geometric regions. On the premise that the image synthesis quality is not lost, the model scale and the transmission overhead can be remarkably compressed, the resource constraint condition of the edge computing equipment can be effectively adapted, and the higher training speed and the better rendering effect are achieved.

The invention discloses a structure semantic prompt constraint-based text image generation method, which comprises the following steps of: taking a generative adversarial network as a generative model of a main body, and respectively learning and updating the generative model and a discrimination model by utilizing an existing sufficient training data set so as to finish model updating of the model and a training model; and on the basis of the generation model which completes model updating, a new image is generated by inputting the description text, so that the purpose of text image generation is achieved. According to the method, the consistency of the text and the image is remarkably enhanced while the image synthesis quality is improved. In order to reduce the dependence of the model on the training batch size and the training round, improve the training efficiency and optimize the resource demand, the difficult-to-load sample mining is introduced into the image generation task for the first time, and the difficult-to-load sample mining matching perception loss is proposed, so that the model can concentrate on the most challenging sample, and the image generation efficiency is improved. In practical application, good performance can be obtained even in an environment where computing resources are limited.

The invention belongs to the technical field of image processing, and discloses a maskless image synthesis method and system based on a diffusion model, and the method comprises the steps: inputting a background image and a noise image containing a target object into a pre-trained potential diffusion model, and enabling the potential diffusion model to execute a reverse denoising process under the guidance of a condition vector, the target object is synthesized in a self-adaptive mode, and a synthesized image is obtained; wherein the condition vector acquisition mode comprises the following steps: encoding a reference image containing a target object into an object embedding vector, and generating a group of learnable background cue words according to the object embedding vector; splicing with the background prompt word; and performing spatial mapping on the spliced feature representation to obtain a condition vector. The method can overcome the defects that in an existing image synthesis technology, the process is tedious, the efficiency is low, a large amount of manual intervention is needed, the synthesis result lacks the sense of reality and harmony, and especially the foreground object cannot be adaptively adjusted according to the background environment.

The invention relates to the technical field of image focus detection, in particular to a fiber endoscope image focus detection method and system, and the method comprises the following steps: providing a data quality guide interface; capturing narrow-band light images of the endoscope away from a first preset position and a second preset position of the intestinal wall in the axial movement process, calculating brightness gain values of a blue light channel and a green light channel, calculating an asymmetric scattering correction coefficient, and updating indication of color information calibration integrity; tracking the pixel moving speed of the interferent in the view in the propulsion operation process, calculating the relative depth of the interferent, and updating the indication of the validity of the depth data of the interference layer; identifying the image area which is not shielded by the interferent, and updating the indication of the information coverage degree of the background area; triggering an image synthesis operation; performing color compensation on the image information of the image area which is stored in the background canvas and is not shielded by the interferent; reconstructing a clear image; and performing focus detection on the clear image. The method improves the accuracy of focus detection.

The invention discloses a CBCT (Cone Beam Computed Tomography) high-quality CT (Computed Tomography) image synthesis method based on structure prior guidance, which comprises the following construction steps of patient data collection and arrangement, space guidance deformable attention convolution, structure prior guidance type conditional diffusion image synthesis and reasoning process and quality evaluation. The method specifically comprises the steps of technical design of a space guiding module, technical design of a deformable sampling kernel generator, technical design of an attention weight modulator, contour extraction operation, diffusion modeling operation, conditional denoising network design and target training. The method provided by the invention has the advantages of structural perception, high spatial selectivity, high modulability and the like; and the reduction precision of the model on the key anatomical region is enhanced by taking a contour map as a structure priori condition. The whole diffusion generation network realizes unification of local detail reservation and global structure stability while maintaining noise robustness, and gradually generates a high-quality pseudo CT image which is close to a real CT modal and has excellent structure consistency.

The invention relates to the technical field of tunnel lighting, in particular to a tunnel portal lighting control method and system, and the method comprises the steps: S1, obtaining a first image in a short exposure time mode; obtaining a second image in a long exposure mode; s2, synthesizing the second image and the first image into a high-dynamic-range image, and dividing the high-dynamic-range image into a low-brightness area, a transition area and a high-brightness area; s3, generating a brightness transition curve of the transition area; s4, determining a brightness adjusting curve of the transition area according to the maximum brightness of the tunnel portal and the light and shade adaptive pupil area change rate; s5, determining the target brightness of each part of the transition area according to the brightness transition curve and the brightness adjustment curve; and S6, adjusting the brightness of the illumination lamp in the transition area to enable the brightness of each part of the transition area to reach the target brightness. The target brightness is determined according to the brightness transition curve in combination with the brightness adjustment curve, the influence of the brightness adaptive pupil area change rate on the sight line is fully considered, and the change of the brightness of the tunnel portal is more adaptive to the function of human eyes.

The invention discloses a 5G signal tower three-dimensional rendering and high-altitude panoramic image synthesis method based on GSII, and the method comprises the steps: firstly carrying out the frame extraction processing of data shot when an unmanned plane flies around a signal tower, and obtaining an original image set; obtaining a signal tower sparse point cloud and a camera pose through a motion structure recovery SfM algorithm, and carrying out 3DGS initialization operation; the method is characterized in that in the process of iterative optimization of a three-dimensional Gaussian point cloud, a GSII network is constructed, structural similarity comparison is carried out on an original image of the same visual angle and a deblurred rendering image, a loss function is sensed by using a high receptive field and back propagation is carried out, meanwhile, parameter updating is carried out on a 3D Gaussian ball attribute and FFC residual error repair network, and the quality of a new visual angle rendering image is improved; and finally, designing a central pixel time sequence splicing module, and converting any number of single-view-angle rendering images into a panoramic image. According to the method, the phenomena of artifacts and blurring which are easy to occur under a high-altitude view angle are solved, and the signal tower high-altitude panorama is finally obtained by synthesizing a plurality of rendering images of a single view angle.

The invention relates to an intelligent operation and maintenance method for power transmission network equipment based on digital twin dynamic coupling, and belongs to the technical field of intelligent detection of power systems. The method comprises the following steps: dynamically splitting a power grid panoramic frame into an equipment physical entity layer and a static background environment layer, and generating a panoramic display picture through cloud and edge parallel rendering and a terminal HMD picture synthesis technology; establishing real-time bidirectional mapping between the physical entity and the virtual twinborn body of the power grid equipment, and synchronizing sensor data to the digital twinborn model; adopting an improved permutation entropy algorithm to extract equipment operation characteristic quantity, and dynamically distributing rendering pipeline resources in combination with a Bayesian network diagnosis model and an analytic hierarchy process evaluation system; and constructing an equipment state quality index, and quantitatively evaluating the health degree of the current equipment and predicting the future degradation trend by dynamically comparing the characteristic parameters with an expected value and combining health factor correction and a time sequence nonlinear prediction model. According to the invention, full-life-cycle efficient management and control of the operation state of the power grid equipment can be realized.

The invention discloses an infrared ship scene generation method and system based on collaborative optimization of a physical prior model and a data-driven algorithm, and a storage medium, and belongs to the technical field of infrared image generation and artificial intelligence fusion. Comprising the following steps: preprocessing an acquired real infrared ship image and an original visible light ship image; different types of ship databases are established through semantic segmentation, and a visible light ship foreground image is obtained; synthesizing a new visible light ship image, and converting the new visible light ship image into a pseudo-infrared image; segmenting the new visible light ship images, classifying the new visible light ship images according to materials, covering the segmented image masks into the corresponding preprocessed real infrared ship images, extracting real infrared gray values of ship parts, establishing a linear regression model, and outputting gray values of ships of different materials; and optimizing the pseudo infrared image, correcting the gray scale of the ship area by using the gray value, and finally generating an infrared ship scene. According to the method, the contradiction between physical consistency and generation efficiency is solved, and the problems of high cost and data shortage of infrared ship image acquisition are solved. Enhancement of infrared ship data can be used for sea ship detection and identification scenes.

The invention relates to the technical field of computer vision and medical image processing, in particular to an intra-intestinal monocular pose estimation method based on diffusion model deformation field prediction, which comprises the following steps of: inputting two frames of enteroscopy images of an intestinal capsule robot with adjacent time sequences, respectively extracting image features of front and back frames of enteroscopy images through a parameter-shared double-path depth map generation network, and generating corresponding depth maps; solving six-degree-of-freedom pose parameters of the intestinal capsule robot by utilizing the pose estimation network; generating a three-dimensional non-rigid deformation field through the non-rigid deformation field prediction network based on the physical regularization diffusion model; reconstructing a composite image aligned with the real previous frame of enteroscope image through an image synthesis module; completing self-supervised training; the problems that in an existing intestinal capsule robot positioning technology, a non-rigid deformation field lacks biophysical constraints, three-dimensional reconstruction geometric distortion is caused by monocular information, and collaborative optimization is lacked in deformation field and pose parameter decoupling are solved.

Apparatuses, systems, and techniques to infer synthesized images including digital representations of objects blended realistically with appropriate background images. In at least one embodiment, background image data and a description of semantic features for a type of object are fused together to infer such a synthesized image using one or more neural networks.

The invention discloses a target detection method and device, equipment, a storage medium and a product, and relates to the technical field of image recognition. And synthesizing the first front view image and the second front view image into a central eye image. Data labeling is performed on the multi-camera data and the central eye image, a labeling data set special for small target detection is constructed, and labeling efficiency and quality are remarkably improved. Performing aerial view coding on multi-scale features contained in the multi-camera data based on a cross-space cross attention mechanism of aerial view query position and image semantic joint modeling to obtain aerial view features; and on the basis of the annotation data set, aligning the aerial view features to the central eye image to obtain aerial view enhancement features, and realizing small target semantic enhancement in the aerial view features. And carrying out target identification on the aerial view enhanced features so as to determine a detection target contained in the multi-camera data. According to the target detection scheme provided by the invention, the detection accuracy and robustness of the small-size target are remarkably improved.

The invention belongs to the technical field of image processing, and discloses a high-end equipment product small target flaw image detection method and system. The method comprises the following steps: S1, establishing and training an Attention GAN model; s2, establishing and training a Pix2Pix model; s3, flaw detection: specifically, randomly selecting an image y with flaws to generate an image F (y), and comparing the image y with flaws with the repaired flawless image F (y) pixel by pixel to obtain a binary image of flaw positions and shapes; according to the method, the complexity of flaw image synthesis can be reduced, the problem that flaws are not seen in detection is solved, compared with a traditional naive CycleGAN, better performance can be obtained, and the effect can be comparable with that of a supervised UNET; in addition, the attention mechanism is introduced to weight the flaw area in the image, so that flaws which are not seen can be more accurately generated and detected.

The invention discloses a display screen defect image synthesis method of a hybrid network model, which is applied to the technical field of machine vision and aims to solve the problem that the prior art depends on a large number of defect samples to detect display screen defects and is difficult to adapt to scenes of industrial defect products. The method comprises the following steps: firstly, preprocessing display screen defect image data; carrying out positive sample training learning by adopting an unsupervised network model; constructing an image-level defect synthesis network model, and synthesizing multi-scale and multi-form display screen defect samples to assist in training discriminator parameters of the unsupervised network model; a feature-level defect synthesis network model is constructed, initial defect data is generated by adopting a Gaussian noise model, an initial feature sample is generated by utilizing a self-adaptive gradient rise and refined truncation projection method, and the initial feature sample and a normal feature sample are superposed to synthesize a feature-level defect sample to assist in training discriminator parameters of the unsupervised network model; and finally, the trained unsupervised network model is adopted to carry out detection and segmentation positioning on a to-be-detected image sample.

The invention discloses a fast full-color imaging method based on a Fourier laminated microscope, which solves the problems of long time consumption and poor imaging quality of a full-color imaging mode of the existing FPM microscopic imaging technology, and specifically comprises the following steps: step 1, obtaining a low-resolution color image of a target through a Fourier laminated microscope system, selecting an optimal acquisition channel from three channels of the system; 2, collecting a target through the optimal collection channel, reconstructing a high-resolution grayscale image based on the collected FPM original data, and calculating a system coherent transfer function and a system incoherent transfer function; 3, calculating a deconvolution IHS color space variable corresponding to the low-resolution color image through a system incoherent transfer function; and step 4, calculating the high-resolution grayscale images of the other two channels through the high-resolution grayscale image of the optimal acquisition channel and the deconvolution IHS color space variable, and synthesizing the three high-resolution grayscale images into a high-resolution color image to complete rapid full-color imaging.

The invention relates to the technical field of image simulation, and discloses a multispectral image intelligent simulation system based on a three-dimensional model, and the system comprises a three-dimensional modeling module, a multispectral imaging module, a light source positioning unit, a light ray tracing unit, a light source intensity calculation unit, a rendering sub-module, and an image synthesis sub-module. On the basis that light source intensity calculation is achieved through a simplified model and based on simple parameters, compared with the prior art, through light source positioning, light path tracking and light source intensity calculation, multi-aspect illumination influences are considered, the influences of surface characteristics and different wavelengths on illumination are comprehensively captured, and the light source intensity calculation accuracy is improved. The illumination effect is improved, the accuracy of light source intensity calculation is improved, data support is provided for subsequent multispectral images and image simulation, and the precision of the multispectral images and image simulation is greatly improved.

The invention discloses a concrete defect image generation method based on mask guidance. The method is suitable for data enhancement and structured image synthesis tasks of defect images such as concrete cracks under the small sample condition. According to the method, a space mask mechanism is introduced, a defect area is accurately positioned and extracted from a reference image, and the injection accuracy and control granularity of defect features are effectively improved. The method comprises three main stages: a multi-scale theme-background feature extraction stage, an adaptive time feature generation stage and a multi-scale cross attention image denoising stage. In the feature extraction stage, spatial attention enhancement and fine-grained semantic alignment are carried out on a reference image and a subject text, so that unified representation of multi-modal features is realized; in the time feature generation stage, feature adaptive time is introduced, and time step related weights are generated to dynamically regulate and control the fusion proportion of theme and background features; in the image denoising stage, guidance denoising and image generation are completed in combination with UNet and cross attention.

The invention belongs to the field of computer vision, and relates to an efficient image synthesis technology of a generative adversarial network (GAN). Aiming at the problems of insufficient high-frequency details, limited geometric denaturation and the like and low calculation efficiency in high-resolution image synthesis of the existing generation method, the invention provides a frequency domain sensing generation method based on wavelet driving WE-GPMConv-StyleGAN. According to the model, discrete wavelet transform is embedded in a StyleGAN3 architecture, a traditional convolution operation is decomposed into a two-stage process of wavelet domain feature extraction and cross-band fusion, and calculation complexity is reduced through grouping separable convolution; designing a low-frequency enhancement module to expand the global structure modeling capability, and expanding a receptive field by using secondary wavelet decomposition; a high-frequency enhancement module is introduced to enhance texture details through a residual structure and cross-channel interaction; in combination with a staged training strategy, low-frequency global features are optimized preferentially in the early stage, and local details are refined by combining a high-frequency module in the later stage. According to the invention, through a wavelet domain multiband decoupling and fusion mechanism, the high-frequency generation quality and geometric denaturation are significantly improved, and a frequency domain driving solution with both efficiency and precision is provided for high-resolution image generation.

In some implementations, the techniques described herein relate to a method including: (i) training, by a processor, a machine learning model to create composite images from background scenes and foreground objects, (ii) identifying, by the processor, a digital image file that comprises a background scene and an additional digital image file that comprises a foreground object, (iii) compositing, by the machine learning model executed by the processor, the digital image file that comprises the background scene and the additional digital image file that comprises the foreground object to produce a composite digital image file that comprises the foreground object and the background scene by performing at least one of a channel concatenation step and a reverse diffusion sampling step, and (iv) causing display, by the processor, of the composite image file that comprises the foreground object and the background scene.

The invention relates to a diffusion prior synthesis and optimization method for cross-modal medical image synthesis, which is used for solving the problems of strong dependence on source modal data, high acquisition cost and the like in the existing method. According to the method, source modal data is not needed, training is carried out only based on single target modal data, and a diffusion prior synthesis (DPS) module and a diffusion prior optimization (DPO) module are included. The DPS encodes the image to a potential space guided by general diffusion prior through a probability flow ordinary differential equation, and decodes the image into an initial image through a target diffusion model; the DPO optimizes an initial image through a linear inverse problem, recovers high-frequency details and corrects discrete errors, thereby improving image fidelity and stability.

To provide "a synthetic image generation system and a posterior image display system" which correct images photographed with a plurality of cameras whose photographing regions partially overlap, with suppressed image failure due to the correction., and synthesize the images.SOLUTION: The synthetic image generation system applies prescribed image conversion which brings the ratio of the pixel values of a first image and a second image closer to 1, to the images. The system extracts from the image-converted first image and the image-converted second image an overlapping portion of photographic regions, as comparison images. The system calculates as a correction gain the ratio of the average of the gradation values of the comparison image extracted from the image-converted first image to the average of the gradation values of the comparison image extracted from the image-converted second image, and corrects the gradation value of the image-converted second image with the correction gain, and applies inverse conversion of the prescribed image conversion to the corrected second image to synthesize with the first image.SELECTED DRAWING: Figure 3

A method includes accessing first image data captured by a camera of an environmental scene in which a subject arm performs a task. A first image is obtained from the first image data. The first image includes a subject arm object representing the subject arm and a background representing the environmental scene. A set of robot arm parameters for the first image is determined at least in part based on the subject arm object. An image of a robot arm object is rendered based on the set of robot arm parameters and a robot arm model. The image of the robot arm object is composited with the first image to obtain a synthetic robot image, which may be used for generating training data for AI model training.

An embodiment provides a method of producing a three-dimensional (3D) model of an object based on a single, frontal input two-dimensional (2D) image. In one example a method includes obtaining an actual, frontal 2D image of an object and generating a pair of synthetic 2D multiview images of the object based on the actual, frontal 2D image of the object. A 3D model of the object is produced based on at least the pair of synthetic 2D multiview images. The 3D model conserves an asymmetry of the object. An output using the 3D model of the object is produced that conserves the asymmetry.