457 results about "Image noise" patented technology

The invention provides an image noise mark feature selection method and system, a storage medium and a computer. The method comprises the steps of obtaining a to-be-processed image noise mark data set; embedding a sample set in the image noise mark data set into a multi-granularity fuzzy cluster to construct a dynamic fuzzy membership evaluation matrix; dynamically evolving a multi-level high-precision granular ball cluster; obtaining mark distribution with high identification degree; constructing a rough perception feature evaluation framework based on granular ball topology driving, extracting decision equivalence classes by combining rough set upper and lower approximation and extended positive domain models, and determining and measuring the contribution degree of each feature to a decision system by fusing multi-granular-ball decision boundary information based on a dependency degree quantitative model; a particle and ball structure consistency verification mechanism is introduced, and multi-level evaluation is carried out on the importance of the features through dependency and consistency. According to the method, the optimal feature subset with strong anti-noise performance and high discrimination capability is obtained, and stable and efficient input support is provided for a subsequent image noise mark learning model.

The invention provides a dolomite multi-scale digital core reconstruction method and device, electronic equipment and a medium, and the method comprises the steps: inputting a to-be-processed CT scanning two-dimensional slice image of a dolomite sample into a reconstruction model, and obtaining a multi-view SEM style image and view parameters; inputting the style image and the visual angle parameter into a NeRF model to obtain a three-dimensional high-resolution image; the reconstruction model is obtained based on the following steps: adjusting a CT scanning two-dimensional slice image and an SEM image of a dolomite sample according to a unified standard; inputting a CT scanning two-dimensional slice image and an SEM image which are in a unified standard into a FastGAN network for preprocessing of image enhancement and augmentation to obtain a standard simulation image; and training a generator in the CycleGAN network based on the standard simulation image to obtain a reconstruction model. The method can solve the problem that a traditional digital core technology is limited by image noise and equipment cost, and nanometer and micrometer multi-scale fusion is difficult to achieve.

An image processing method, performed by a computer device, comprising: extracting sketch texture features at multiple scales from a sketch image; extracting image noise features at multiple scales from preset noise; determining color guide information corresponding to the sketch image; encoding, for each scale, a noise feature based on a sketch texture feature and the color guide information to obtain multi-scale image features; and performing multi-scale decoding on these image features to obtain a colored image comprising a sketch texture corresponding to the sketch image and a color based on the color guide information. A related training method and apparatus are also provided to develop models for this image processing technique.

The invention relates to the field of computer vision, in particular to an image feature recognition method based on computer vision, which comprises the following steps of: preprocessing an input image, removing image noise and correcting image gray deviation to obtain a preprocessed image; performing multi-scale feature extraction on the preprocessed image, obtaining image local texture features and global contour features under different scales, performing dynamic weight fusion, calculating dynamic fusion weight based on a feature response value and a local complexity factor, and performing dimension reduction processing on fusion features through a principal component analysis algorithm to obtain a target feature set; and matching the reference features, calculating the Euclidean distance between the target feature set and the reference features, determining an adaptive matching threshold, judging a matching result, and completing image feature recognition. According to the method, through adaptive filtering and gray level equalization preprocessing, Gaussian pyramid multi-scale feature extraction and dynamic fusion and adaptive threshold matching, feature extraction accuracy, characterization capability and recognition robustness are improved.

The invention relates to the technical field of machine vision and product defect visual prediction, in particular to a product defect visual prediction method and device based on X rays. According to the method, edge extraction and entropy-based analysis are utilized to enhance defect positioning, meanwhile, the challenges of image noise and non-uniform intensity change are solved, and in order to improve robustness, a self-adaptive threshold strategy combined with DBSCAN clustering is adopted to distinguish defects and noise. Therefore, the problems that in the prior art, the ratio of product defects is low, the visual prediction efficiency of edge blurring is low, and the accuracy is not high are solved. Compared with the prior art, noise reduction, edge detection and crack identification are enhanced. By integrating X-ray image processing, the method improves accuracy, reduces false positive and improves detection efficiency.

The invention relates to a circuit board defect identification method and system based on multi-dimensional image data, and belongs to the technical field of data identification processing, and the method comprises the following steps: obtaining synchronous image data of a circuit board to be detected in a plurality of imaging modes; performing space-spectrum joint registration on each modal image to generate a multi-dimensional image cube with a unified coordinate system and pixel alignment; inputting the multi-dimensional image cube into a pre-trained multi-branch heterogeneous fusion neural network; generating a pixel-level defect probability graph by utilizing a defect sensing context decoder, and performing geometric constraint optimization on the probability graph by combining prior information of a circuit board design layout; outputting defect types, positions and confidence coefficients, and establishing an interpretable defect fingerprint database according to the multi-dimensional response characteristics of the defects; the method has the beneficial effects that false defect signals generated by image noise and circuit board surface texture interference can be effectively inhibited, the omission ratio and the false detection ratio are greatly reduced, and pixel-level accurate defect positioning is realized.

The invention belongs to the technical field of medical image processing, and particularly relates to a medical image processing method and system based on multi-modal fusion, and the method comprises the steps: obtaining a CT image and a corresponding digital pathological image, and carrying out the pairing of the CT image and the corresponding digital pathological image; respectively inputting the paired CT image and digital pathological image into a CT branch and a pathological branch; the CT branches extract macroscopic morphology and internal density distribution to obtain a CT feature map, and the pathological branches extract microscopic details to obtain a pathological feature map; the CT feature maps and the pathological feature maps from different layers are fused in a layer-by-layer cascade and feature splicing mode, and a comprehensive feature map is obtained. The method can still stably extract and identify key image features and modes in the face of image noise, different imaging devices or image visual differences among individual samples, and has high image analysis robustness.

The invention provides a noise modeling and denoising method for a low-light-level image intensifier. The noise modeling and denoising method comprises the following steps: step 1, establishing an imaging noise model for the low-light-level image intensifier; step 2, collecting output data of the low-light image intensifier under different illumination conditions, calibrating parameters of an imaging noise model, and superposing the calibrated imaging noise model on a noise-free image sample to construct a training data set conforming to actual imaging characteristics; step 3, designing an image denoising network based on the convolutional neural network, the input of the network being a noise-containing image, and the output being a noise-suppressed clean image; and step 4, training and optimizing the convolutional neural network based on the training data set, and verifying synthetic data and real data. According to the method, the adaptive capacity and generalization performance of the neural network to complex noise can be effectively enhanced, and finally, more accurate and stable image enhancement and denoising effects can be realized in a low-light imaging scene.

The invention provides an ultrasonic image focus automatic identification method based on deep learning, and the method comprises the steps: obtaining an ultrasonic probe pressure and an ultrasonic image of a body surface contact region of a patient, carrying out the tissue hierarchy analysis of the ultrasonic image, and recognizing the tissue density and the three-dimensional space coordinate information of the depth position of a target organ; driving sound wave signals of the ultrasonic probe through the target probe frequency, monitoring the contact pressure change of the probe and the body surface in real time, and judging whether the current contact state meets the pressure requirement of deep imaging or not according to the contact pressure change; identifying a deep focus boundary of the enhanced ultrasonic image, and performing multi-scale feature extraction and texture analysis on the ultrasonic image to obtain contour coordinates, tissue gray level distribution and internal uniformity of a focus area; meanwhile, the image noise level and the tissue contrast difference are evaluated to obtain the definition, the contrast ratio and the boundary sharpening degree of the current image.

Systems and methods for improving the image quality of images of epithelial tissue structures are disclosed. The systems include training a first cycle-GAN model and a second cycle-GAN model simultaneously, where the first cycle-GAN model is trained to remove noise from an image and the second cycle-GAN model is trained to learn the structure of the image. Additional systems and methods include deploying the trained cycle-GAN model to identify an unknown image segment and / or generate a protocol for following the identified skin care treatment recommendation for an identified image segment.

The invention relates to the field of visual inspection, and discloses an intelligent automatic visual inspection system and method for an automobile brake disc, and the method comprises the steps: carrying out the time-sharing sampling imaging of a friction working surface of the brake disc through a multi-station imaging device when the brake disc enters a detection station along with a conveying mechanism; analyzing and separating the main texture direction, periodic distribution and continuity characteristics in the image group around regular turning textures formed in the brake disc machining process; performing stratified analysis on abnormal signs by combining distribution characteristics and gray disturbance amplitude differences of residual information under different spatial scales; non-structural fluctuation caused by local reflection or imaging noise is eliminated by analyzing the difference between the texture extension trend and the gray scale stability of adjacent areas; and comprehensively judging the distribution position and the influence range of the suspicious area on the working surface in combination with the reference position of the brake disc structure. The method has the advantage of improving the surface quality detection accuracy of the brake disc.

The invention relates to the technical field of medical images, in particular to a multi-modal medical image registration and fusion analysis method, which comprises the following steps of: eliminating image noise and artifacts based on a modal adaptive filtering strategy; constructing a pyramid type feature extraction network to realize multi-scale feature extraction, calculating feature matching degrees among different modal images, and dynamically adjusting matching weights by combining feature differences among modals; a focus area attention mask is constructed, targeted enhancement of registration image features is realized, and a hierarchical fusion strategy is adopted to evaluate the quality of a fused image; and constructing a multi-task deep learning model to complete focus automatic detection, segmentation and benign and malignant preliminary judgment on the fused image. According to the multi-modal medical image registration and fusion analysis method, a three-layer feature pyramid is constructed, a multi-feature fusion matching cost function is introduced, and cross-modal feature matching is optimized through an adaptive weight iteration nearest point ICP algorithm, so that the information richness, marginal definition and focus discrimination of a fused image reach the standard.

The invention provides a slit-type spectral imaging method based on filtering phase reconstruction and a grating spectrometer. The method comprises the following steps: measuring wavefront aberration of a slit-type grating imaging spectrometer before slit filtering by using a wavefront sensor; based on a slit filtering principle, reconstructing a phase function after slit filtering by using wavefront aberration before slit filtering; and capturing a slit image, and reconstructing the slit image by using the reconstructed filtered phase function of the slit to obtain a final reconstructed image. According to the method, effective recovery of the slit imaging data of the grating spectrometer based on the slit is realized, the influence of optical aberration and image noise on the spectrum and spatial resolution of the spectrometer is effectively eliminated, and the slit imaging quality is remarkably improved and is close to the diffraction limit.

The invention relates to the technical field of artificial intelligence, and discloses a low-illumination scene-oriented unmanned aerial vehicle imaging data enhancement method, which comprises the following steps of: S1, acquiring unmanned aerial vehicle imaging data and constructing a training data set; s2, data preprocessing based on illumination partition; s3, constructing and training a data enhancement model; and S4, enhancing the imaging data of the unmanned aerial vehicle. The data preprocessing based on the illumination partition in the step S2 comprises the following steps: generating a low-illumination area mask and a medium-illumination area mask based on the brightness information of the low-illumination image; a Gaussian smooth convolution operation is applied to the low-illuminance region identified by the low-illuminance region mask to suppress noise, and an original value is reserved for the medium-illuminance region identified by the medium-illuminance region mask to generate a pre-processed image. The problems of low-illumination image noise amplification and detail loss are solved by generating low-illumination region masks and medium-illumination region masks and respectively adopting noise reduction and retention strategies, and the balance of dark region noise reduction and bright region detail protection is realized.

The invention discloses a ViT-based automatic detection and recognition system and method for Manchu ancient book characters, and belongs to the technical field of artificial intelligence and digital literature protection. The system comprises a YOLO target detection module, a positioning network module, a ViT module and a classifier module. The YOLO target detection module is used for identifying and positioning each individual character area in the whole Manchu ancient book image, the positioning network module is used for automatically adjusting the position, the rotation angle and the scaling of the character image, and the ViT module is used for extracting the deep feature of the character image and modeling the character visual representation. And the classifier module is used for mapping the feature vectors output by the ViT module into specific character tags and splicing the specific character tags into a transcription text. The invention provides an end-to-end intelligent processing system, solves the problems of character pattern diversity, image noise, multi-language mixing and the like of Manchu literatures, and realizes high-precision identification of Manchu characters and intelligent sorting and translation of Manchu and Chinese bilingual literatures.

The present disclosure relates to systems, methods, and non-transitory computer readable media for utilizing a diffusion neural network for mask aware image and typography editing. For example, in one or more embodiments the disclosed systems utilize a text-image encoder to generate a base image embedding from a base digital image. Moreover, the disclosed systems generate a mask-segmented image by combining a shape mask with the base digital image. In one or more implementations, the disclosed systems utilize noising steps of a diffusion noising model to generate a mask-segmented image noise map from the mask-segmented image. Furthermore, the disclosed systems utilize a diffusion neural network to create a stylized image corresponding to the shape mask from the base image embedding and the mask-segmented image noise map.

The invention discloses a multi-frame fusion fixed image noise dynamic compensation system and method, particularly relates to the technical field of image processing, and is used for solving the problem that fusion boundary noise has artifacts due to existing independent noise compensation. The method comprises the following steps of: establishing a mapping relation between pixel-level imaging parameter difference degree and fusion weight by acquiring a multi-frame image sequence of different imaging parameters and identifying a transition region of brightness gradient dramatic change; extracting a frequency domain energy skewness feature and a chroma phase change boundary feature of the transition region, and generating a noise visibility correction factor through nonlinear fusion; dynamically constraining the compensation intensity range to realize multi-frame noise compensation collaboration; and finally, performing cooperative compensation and fusion output within the constraint range. Fusion boundary noise discontinuity is effectively eliminated, and the visual quality of high-dynamic-range imaging is remarkably improved.

This application relates to the field of medical image processing technology, and provides a method, apparatus, device, and storage medium for reconstructing ECT images. This allows for more thorough capture of subtle changes in the concentration distribution of radioactive tracers at high temporal resolution, helping to reduce image noise levels and improve image quality. This application obtains a first ECT image sequence; the first ECT image sequence is obtained by scanning with an ECT scanning device; the first ECT image sequence is input into a pre-constructed intermediate frame prediction network to obtain an estimated second ECT image sequence; using the estimated second ECT image sequence as a priori image, a target second ECT image sequence is reconstructed; the estimated second ECT image sequence and the target second ECT image sequence belong to the second ECT image sequence; the temporal resolution of the second ECT image sequence is higher than that of the first ECT image sequence.

The present description relates to an image sequence motion correction method and apparatus based on intensity correction optical flow registration. The method comprises: determining a reference image of each deformed image in a deformed-image sequence, and performing motion correction on the deformed image by means of adjusting the weight of a mechanical regularization term between the reference image and the deformed image, so as to obtain a displacement field of the deformed image relative to the reference image; on the basis of the displacement field and an intensity correction value of a finite element node, performing intensity correction on the reference image; adjusting the weight of the mechanical regularization term, performing motion correction on the deformed image, performing intensity correction on a first corrected reference image on the basis of an optimized displacement field, repeatedly executing the process until an image residual function converges, determining that the deformed image has been corrected, and recording the optimal solution of a weight value of the mechanical regularization term of the deformed image; and on the basis of the optimal solution of the weight value, determining the actual displacement of the deformed image. In the present description, the mechanical model behind physiological motion and an intensity correction technique are integrated into an image registration algorithm to reduce image noise, thereby improving the algorithm accuracy.

The invention discloses an electrical equipment insulation defect image recognition system and method based on deep learning, and relates to the technical field of electrical equipment intelligent detection, and the method comprises the steps: receiving an original image, dividing the original image into a foreground region, a mid-scene region and a background region, extracting electrical topological structure data, and generating a discharge path risk thermodynamic diagram in combination with the mid-scene region; identifying candidate defect areas in a foreground and a middle scene by using a target detection model, extracting discharge risk distribution data of the defect areas in combination with a thermodynamic diagram, analyzing background image features to estimate an image noise probability, fusing the discharge risk and the noise probability through a preset risk confidence coefficient calculation function, and calculating a defect risk confidence coefficient score; candidate defect areas with the risk confidence scores higher than a threshold value are marked in the image; the method has the beneficial effects that the spatialization and quantification risk judgment of the defect is realized through the discharge path risk thermodynamic diagram, the accuracy and robustness of insulation defect identification can be improved, and the accurate marking of the high-risk defect is realized.

The invention relates to the technical field of scanning electron microscope image processing, and discloses a scanning electron microscope image restoration method based on a wavelet frequency domain adjustment diffusion model, and the method comprises the steps: constructing a two-stage degradation pipeline comprising random light path disturbance and fixed circuit collection limitation, and generating a specific domain training data set; extracting multi-scale high-frequency energy characteristic modulation noise distribution by using a frequency domain prior encoder, and training a wavelet frequency domain adjustment diffusion model; and inputting a target low-quality degraded image into the pre-training model, initializing a potential noisy state, then executing reverse denoising iteration, calculating wavelet domain consistency gradient correction noise prediction, and updating the potential noisy state until a restored image is generated. According to the method, data distribution mismatch is relieved through physical degradation modeling, high-frequency detail perception is enhanced by means of a frequency domain adjustment mechanism, the structural fidelity is improved through wavelet domain gradient guidance, and balance between scanning electron microscope image noise suppression and detail recovery is achieved.

The invention provides a wake detection method based on multi-layer graph interactive guidance and a multi-head voting type Transform. According to the method, depth features are extracted through FcaNet in combination with Fourier transform, frequency domain key details are explicitly modeled, and a frequency domain-time domain joint information feature map is output; a graph structure is constructed, and through multi-layer graph interaction, the discrimination capability of a complex wake form is enhanced by using an MHVT module; an MSFM module is introduced to realize space and channel dual fusion, and then a wake and an anchor frame are positioned through a regression network, so that the interference of SAR image noise and wake form variability is effectively overcome, and efficient and accurate ship wake detection is realized.

The invention provides an image conversion model establishment method and device, an infrared image labeling method and device and a medium, and relates to the field of automatic driving, and the method comprises the steps: obtaining a visible light image and an original infrared image which are aligned in time and space; inputting the visible light image into a material attribute estimation branch to output a material semantic graph, and inputting the original infrared image into an image noise adding branch to output a noise-added infrared image; the two and the visible light image are jointly input into a diffusion model to output predicted noise; the predicted noise, the noise-added infrared image and the original infrared image are input into a multi-task loss module to output weighted summary loss, and diffusion model parameters are iteratively updated according to the weighted summary loss so as to construct an image conversion model; and inputting the visible light image into the image conversion model to output a target infrared image, and mapping and labeling to obtain infrared labeling information. According to the method, the physical consistency of the infrared generated image and the visible light image is remarkably improved, the task of the infrared generated image is friendly, and a large-scale high-quality infrared data set can be quickly constructed.

The invention relates to the technical field of oral decayed tooth detection, and discloses an oral decayed tooth detection system and detection method. Through quadric surface fitting and smoothing processing, image noise is reduced and local details are reserved. By calculating an image gradient and constructing a gradient field, the gray level change direction and amplitude are effectively extracted. The divergence calculation and the weighted aggregation degree improve the accuracy of abnormal region identification, and especially in early-stage decayed tooth detection, the abnormal region can be highlighted and the normal region can be ignored. The self-adaptive threshold setting is combined with the local mean value and the standard deviation, the threshold is dynamically adjusted, and false detection and missing detection are reduced. And 8, gathering the scattered candidate points into a complete area according to an adjacency principle and connectivity analysis, and ensuring that a decayed tooth area with an irregular shape is not missed. The method ensures that the final detection area has clinical significance through the screening of features such as physical area and boundary pixel number, and enhances the operability of the detection result through red contour labeling.

The invention discloses an elastic reverse time migration imaging method based on deviatoric stress characterization, and relates to the technical field of exploration geophysics. Deviatoric stress is introduced into a conventional elastic wave equation, and a speed-deviatoric stress elastic wave equation describing stress propagation characteristics is obtained through derivation; based on Hooke's law, P wave contribution in total partial strain is removed through partial strain decomposition, decoupling of P-S wave stress components is achieved, and correctness of amplitude, phase and physical unit of a decoupling wave field is guaranteed; and then a joint imaging condition with the average stress as a seismic source wave field and the multi-stress component as a receiving wave field is constructed, PP waves are subjected to normalization superposition through the average stress and the deviatoric stress, and PS waves are subjected to shear stress imaging. According to the method, the sensitivity to heterogeneity and local stress variation of underground media can be remarkably improved, P-S wave crosstalk and imaging noise are effectively eliminated, the imaging precision and resolution of a complex geologic structure are greatly improved, and a reliable technical means is provided for high-precision seismic exploration of complex oil and gas reservoirs.

The present application belongs to the technical field of engineering structure health monitoring, and discloses a photovoltaic flexible support cable force non-contact monitoring method and system based on video vibration. First, a cable video sequence is acquired, and a dynamic sensitive area is adaptively divided along the cable image. The time sequence average value of the pixel intensity in each area is calculated, the two-dimensional image information is converted into a one-dimensional vibration time history signal, and the light drift and image noise are suppressed through difference and wavelet denoising. After filtering the signal of each dynamic sensitive area, the initial frequency set is formed by preliminarily extracting the multi-order dominant frequency estimate value. The frequency observations from all areas are fused and analyzed, and the real frequency cluster is identified. The weighted centroid frequency of the cluster is used as the multi-order vibration frequency. Finally, the multi-order frequency is substituted into the model, and the real-time cable force value is output. The whole process of the method does not need to contact the measured cable, and provides an innovative and reliable solution for long-term online safety monitoring of the flexible support cable force in complex environments.

The invention discloses an image binarization method based on a dynamic window. The method comprises the steps of 1, target image noise suppression and gray level conversion; 2, carrying out compression processing on the dynamic range of the bright part of the image through a logarithmic function; 3, calculating a global threshold value of the grayscale image by utilizing a maximum between-class variance method; 4, dividing the grayscale image into image blocks according to the image gradient, and calculating a local threshold value for each block by using a Sauvola algorithm; 5, according to the global threshold value and the local threshold value of each image block, the threshold value of each image block is obtained through calculation according to the weight proportion; and 6, segmenting each image block by using a corresponding threshold value to obtain a final segmented image. According to the invention, the size of a local window is adaptively changed when any pixel point is binarized; and the image binarization capability under the condition of non-uniform illumination in a complex illumination environment is remarkably improved.

The application discloses a UAV autonomous obstacle avoidance method and system based on vision-laser radar fusion, belongs to the technical field of UAV autonomous flight, and comprises the following steps: S1, collecting vision image data and laser point cloud data of the environment around a UAV through a vision sensor and a laser radar synchronously; adopting a hardware trigger and a timestamp double synchronization mechanism to control the time synchronization error of the vision image data and the laser point cloud data within + / -1 mu s; respectively carrying out denoising, enhancement and distortion correction processing on the vision image data; and adopting a bilateral filtering algorithm to remove image noise. The hardware trigger and the timestamp double synchronization mechanism are adopted to control the time synchronization error of the vision and the laser radar data within + / -1 mu s, the problems of fusion feature space dislocation and obstacle calibration distortion caused by a traditional synchronization mode are solved, a high-precision data foundation is laid for multi-sensor feature fusion, and the real-time sensing demand of high-speed flight of the UAV is met.

The application relates to a multi-modal feature interaction method and device based on information filtering. The method comprises the following steps: extracting image local features and text features; constructing a text semantic condition; generating adaptive weight filtering image noise based on the condition, and obtaining filtered features; updating the text features by using the filtered features; and guiding the image features again by using the updated text features. The filtering and interaction processes are jointly optimized by using a loss function containing a regularization term. The method can effectively suppress noise in multi-modal data, and improve feature consistency and task robustness through bidirectional closed-loop interaction.

The present application relates to the technical field of ultrasound image processing, and particularly relates to an image processing method, device and system based on an interventional array ultrasound transducer. The method comprises: processing a two-dimensional ultrasound image through noise evaluation to obtain a noise evaluation image; the noise evaluation image can be used to accurately partition image noise; a first filtering algorithm and a second filtering algorithm can be used to preliminarily process the partitioned noise to obtain a preliminarily denoised image; the preliminarily denoised image is then processed accurately using a U-Net model to obtain an accurately denoised image; the accurately denoised image is regionally divided to obtain an edge image and a smooth image; corresponding algorithms are executed on the edge image and the smooth image to obtain an edge denoised image and a smooth denoised image; finally, the edge denoised image and the smooth denoised image are weighted and fused to obtain a denoised ultrasound image; and the accuracy of ultrasound image denoising is improved.