525 results about "Normalization (image processing)" patented technology

The invention relates to the technical field of image defect detection, in particular to an intelligent drug particle detection system and method based on image processing. According to the method, a high-resolution industrial camera and a polarized light source are used for dynamically shooting assembly line medicine particles so as to obtain multi-angle high-quality original images; noise suppression and microstructure enhancement are realized through multi-scale residual information enhancement and structure maintenance adaptive filtering, and the detail resolution is improved by using local contrast normalization; a multi-scale structure gradient of direction perception and region shape prior are fused, and a hierarchical contour evolution and dynamic threshold strategy is adopted to accurately segment a particle main contour and a microcrack; micro defects are identified based on a multi-scale texture sensing algorithm guided by a boundary relative potential graph, and multi-class defect classification is realized through a spatial and semantic dependency relationship between graph neural network modeling regions; and finally, a traceable quality detection report is generated. The intelligent level, the accuracy level and the automation level of medicine particle detection are improved.

The invention belongs to the technical field of medical image processing, and particularly relates to a semi-supervised cardiac image segmentation method based on a Mamb-Transform double structure and contrast learning. A segmentation model used in the method comprises an MTSeg encoder and a VNet decoder, an input image enters the MTSeg encoder after being subjected to linear layer and position embedding operation, the MTSeg encoder comprises a plurality of MTSeg encoding modules, the VNet decoder comprises a plurality of decoding modules, and output features of the MTSeg encoding modules are in jump connection with the corresponding decoding modules after being subjected to projection operation; performing normalization operation on the output feature vector of the VNet decoder to obtain a segmentation result; the MTSeg coding module comprises an improved Mama branch and a Transform branch, and output feature vectors of the two branches are fused through a cross attention layer to obtain an output feature vector of the MTSeg coding module; the improved Mama branch comprises an MS module and a DRFB module which are connected in series, the MS module is used for enhancing the feature extraction and calculation efficiency of the model, and the DRFB module is used for enhancing the expression ability and training stability of the model. The method achieves the effective integration of a long-range dependency relationship, global information, multi-scale information and local information, improves the feature extraction capability of the model, and obtains a finer and more accurate segmentation result.

The invention discloses a dark light enhancement method under the view angle of an unmanned aerial vehicle, and relates to the technical field of image processing and enhancement, and the method comprises the steps: collecting a continuous frame dark light image sequence of a target when the unmanned aerial vehicle flies, carrying out the preprocessing operation including denoising and normalization, and forming a dark light image set; and estimating the motion between adjacent frames by using an image recognition algorithm. According to the invention, through dynamic range compression and detail enhancement processing, the image definition and visibility in a dark light environment are improved, the brightness difference of the image is balanced by adopting a dynamic range compression algorithm, local overexposure or underexposure is avoided, the image can keep a good visual effect under different illumination conditions, and the image quality is improved. And the detail enhancement processing highlights texture and edge information in the image through multi-scale gradient fusion and adaptive sharpening, and improves the detail definition in dark light, so that the unmanned aerial vehicle can recognize a target more clearly when executing a task at night or in a low-light environment, and the task execution efficiency and accuracy are improved.

The invention relates to the technical field of remote sensing image processing, and discloses a remote sensing image adaptive identification method and system for territorial space planning, and the method comprises the steps: obtaining a multi-source remote sensing image data set of a research region, feature extraction, cloud detection, quality evaluation and adaptive preprocessing are carried out; carrying out prototype network coding, calculating a category prototype and probability, and supporting fine tuning of a set; carrying out multi-scale cavity convolution and category scale attention fusion; evaluating the adaptability score of the comprehensive fusion feature map set, and carrying out weighted fusion, classification and normalization; change detection is carried out, stable and change regions are segmented, and time sequence context features are extracted and constrained optimization is carried out; entropy is fused, a boundary is decided, uncertainty is estimated, and weighted fusion is carried out according to a change area; conditional random field optimization, confidence level grading and connected domain identification are carried out; the automation level, the adaptive capacity and the recognition reliability of remote sensing monitoring of territorial space planning are improved.

The invention relates to the technical field of medical image processing, and particularly discloses a medical image segmentation method with adaptive receptive field and feature correction, which comprises the following steps: (1) acquiring an original medical image and a segmentation label thereof, and constructing a training and testing data set; (2) carrying out size normalization and enhancement processing on the image; (3) establishing an improved U-shaped encoder-decoder segmentation network, introducing an adaptive branch mixed shape convolution module in a shallow layer, and improving edge and texture feature modeling capability by adopting a multi-branch banded convolution and channel attention mechanism; (4) a residual directional feature interaction module is introduced into a deep layer, a spatial dependency relationship is modeled through an information interaction structure in the horizontal and vertical directions, and the direction sensing ability of the heterostructure is enhanced; and (5) completing network training and reasoning, and outputting a segmentation result. The method gives consideration to the calculation efficiency and the segmentation precision, and is suitable for the automatic segmentation task of various types of medical images with complex structures.

The invention discloses a big data processing system based on AI cloud computing training, and the system comprises a data collection module which collects multi-source image data through a cloud interface, and generates a standardized data set after distributed cleaning and feature extraction; the data prediction module constructs a resource prediction model, and outputs a resource availability prediction vector in combination with a federated average algorithm and a space-time diagram neural network; the dynamic scheduling module adopts a double-queue architecture and an NSGA-II algorithm to optimize resource allocation and reduce task delay and energy consumption; the fusion decision module integrates the data and the scheduling result, generates a decision through an AI model, dynamically adjusts task migration, and improves the cloud image processing efficiency; therefore, the number of the image data fragments can be automatically adjusted according to the real-time load and the storage capacity of the cloud computing nodes, load imbalance among the nodes is avoided, the efficiency of image data cleaning, normalization and feature extraction is greatly improved, the data processing time delay is reduced, and the overall throughput is improved.

The invention relates to the field of computer technology and medical image processing, in particular to a medical image segmentation method fusing random region cutting enhancement and a pseudo label semi-supervised mechanism. In order to solve the problems of scarcity of annotation data, weak model generalization ability, inaccurate segmentation boundary and the like in a current medical image segmentation task, the invention provides a medical image segmentation method fusing random region cutting enhancement and a pseudo label semi-supervised mechanism, which is called an RCDE segmentation model. The model adopts a shared encoder and double decoder structure, combines a structure-level disturbance generation strategy, enhances the perception ability of the model for image structure change by mixing and recombining labeled images and unlabeled images, generates pseudo labels by utilizing a teacher network, and introduces a dynamic confidence coefficient screening mechanism, so that low-quality pseudo labels are effectively eliminated, and the robustness of the model is improved. The training stability and the pseudo-supervision effect are improved, preprocessing such as size normalization and image enhancement is carried out on the image, and the consistency and robustness of model input are improved.

The invention discloses an industrial robot image processing method based on image fusion, and relates to the technical field of intelligent aquaculture, and the method comprises the steps: collecting an original image in real time through deploying an image collection device integrating visible light, polarization and multispectral imaging, and extracting suspended matter density, water body light transmittance and illumination intensity change information; generating a first image set; suppressing suspension interference through image filtering and enhancement processing to obtain a first corrected image; extracting an aquatic product individual region, performing multi-source image fusion, identifying color deviation, texture interruption and reflection feature anomaly regions, and constructing a lesion candidate set; gray scale reconstruction, edge gradient and brightness normalization correction of a multispectral channel are executed based on illumination and reflection changes, and a high-quality fusion image is generated; and calculating a health anomaly probability coefficient of the target individual by using the depth recognition model, comparing the health anomaly probability coefficient with a threshold value, and recording a recognition result and collecting information if the threshold value is exceeded. The method can significantly improve the accuracy of aquatic individual lesion recognition.

The invention discloses a super-resolution image reconstruction method based on a multi-scale large-kernel convolution double-residual neural network, which is suitable for the field of image processing, and comprises the following steps: cutting a data set, inputting a cut original low-resolution image into a preprocessing module, carrying out image normalization and data enhancement operation, and carrying out image reconstruction; generating a preprocessed low-resolution image; the preprocessed low-resolution images form a distorted image block data set, and a training set, a verification set and a test set are formed; according to an existing distorted image block data set, a super-resolution image reconstruction method based on a multi-scale large-kernel convolution double-residual neural network is constructed; and inputting the data set into the constructed multi-scale large-kernel convolution double-residual neural network to extract semantic features, and amplifying a feature map by using an up-sampling module of the model to generate a super-resolution image. According to the method, a multi-scale large-kernel convolution and double-residual structure is introduced, a visual attention mechanism is used in the neural network, the extracted features better conform to human visual perception features, and super-resolution image reconstruction is more accurate.

The invention relates to the field of computer vision and image processing, and discloses a remote controller liquid crystal screen image detection method based on computer vision, which comprises the following steps: S1, acquiring an original image containing a remote controller, and carrying out gray level conversion and filtering denoising processing on the original image to obtain a preprocessed image; s2, calculating the local information entropy and the change rate of the image through a sliding window under multiple scales based on the preprocessed image, generating a multi-scale entropy gradient heat map, and screening out a candidate region with high entropy difference as a liquid crystal display region according to a set threshold value; and S3, carrying out vectorization partitioning on the candidate region, and constructing an image matrix. According to the invention, by introducing an image preprocessing mode of low-rank sparse decomposition and normalization processing, the expression ability of structural information in the character image is effectively enhanced, and the feature extraction accuracy under the conditions of complex background, low character contrast and the like is improved.

The invention relates to the technical field of image processing, in particular to a printing image-text defect detection method based on machine vision, and the method comprises the steps: obtaining an RGB image of a to-be-detected printed matter, and carrying out the Lab space conversion, and obtaining a target image; performing normalization processing on the L channel value of each pixel point in the target image to obtain a normalized brightness value so as to divide the pixel points in the target image into a preset number of illumination level intervals; obtaining a color fluctuation statistical parameter of any one illumination level interval, and obtaining an optimal color difference threshold value corresponding to any one illumination level interval; according to the optimal color difference threshold value corresponding to the illumination level interval to which each pixel point in the target image belongs, performing Lab color space template comparison on the target image to obtain an image-text color difference defect result of the to-be-detected printed matter, thereby improving the accuracy of image-text printing defect detection based on the color difference threshold values.

The invention belongs to the technical field of image processing, and discloses a multi-feature enhancement fusion Mama image denoising method, which comprises the following steps: constructing a Mama denoising model for image denoising processing, the Mama denoising model comprising a shallow feature extraction module, a depth feature extraction module and an image reconstruction module; the shallow layer feature extraction module is used for extracting shallow layer features of the input image and performing normalization processing on the shallow layer features; the depth feature extraction module comprises a plurality of DFEGs and a sixth convolutional layer, and feature extraction is performed on the normalized shallow layer features step by step through the plurality of DFEGs; according to the invention, through a plurality of DFEGs in the depth feature extraction module, gradual feature extraction from a shallow layer to a deep layer is realized, and through cooperation of the Mama long-distance dependence capture unit, the local residual error module LRB and the channel attention mechanism CAB, the problem that global structure and local detail recovery are difficult to consider at the same time in a complex noise environment in the prior art is effectively solved.

The invention relates to the technical field of image processing, in particular to an image search precision optimization method and system based on the combination of rough arrangement and fine arrangement, and the method comprises the steps: a rough arrangement model stage: calculating the rough arrangement similarity of a query image and all images in a database, and sorting and screening out a preset number of candidate images according to the rough arrangement similarity; a model fine arrangement stage: performing multi-scale reconstruction on the query image to generate reconstructed images with different resolutions; extracting feature vectors of the reconstructed images through a fine arrangement model, performing expansion operation on the feature vectors, and calculating fine arrangement similarity between the query image and the candidate image; and a similarity weighting combination stage: performing normalization processing on the rough ranking similarity and the fine ranking similarity, dynamically adjusting the weight coefficients of the rough ranking similarity and the fine ranking similarity according to the difference of the rough ranking similarity, generating a final similarity through weighted average, and finally sequencing and outputting a retrieval result. And by dynamically combining the distance of rough arrangement and the distance of fine arrangement, the image retrieval precision is further improved.

The invention provides a cultural relic crack monitoring system and method based on image processing, and relates to the technical field of image processing. The system is composed of a multispectral image acquisition module, an environment calibration module, an image processing module, a contour recognition module, a crack recognition module, a time sequence evolution analysis module and an output module. The calibration module performs reflectivity normalization according to the material spectrum and real-time illumination and inhibits specular reflection; the image processing module uses multi-scale Retinex to enhance the crack contrast; the contour recognition module is used for extracting candidate contours by using improved Canny; the crack identification module introduces a lightweight CNN to carry out texture analysis and generates crack and texture data; the time sequence evolution analysis module completes multi-moment crack registration and morphological parameter difference; and the output module summarizes the data to generate a chart and a monitoring report. The false detection rate of a traditional method is reduced.

The invention discloses a dual-branch attention and boundary enhanced remote sensing image change detection method and system based on a Mama framework, and belongs to the technical field of computer vision and remote sensing image processing. The method comprises the following steps: carrying out cutting, data set division and normalization preprocessing on a dual-tense remote sensing image; multi-scale features are extracted through a Mamba encoder, and global spatial dependence is modeled by using a state space model; constructing a boundary enhancement branch, and combining channel compression and up-sampling to generate a boundary probability graph; in the decoding stage, a space-time state space (STSS) module is introduced to fuse space-time features, and a dynamic channel attention mechanism is embedded to strengthen a change sensitive area; using deformable convolution to optimize edge offset, and performing weighted fusion on main branch and boundary branch output to generate a final change graph; and training the model through a composite loss function and optimizing parameters. According to the invention, the detection precision of a change area in a complex scene is significantly improved, and the problems of boundary blur, small target leak detection and background interference are solved.

The invention discloses a video anti-shake method and system based on multi-scale fusion and adaptive smoothing, relates to the technical field of video image processing, and aims to effectively solve the image quality problem caused by shake in a video shooting process. Gradient histograms and wavelet energy distribution characteristics of video frames are extracted through graying and normalization processing, and the gradient histograms and the wavelet energy distribution characteristics are input into a jitter type recognition network to recognize translation, rotation and Z-axis jitter probabilities. And further extracting motion, frequency domain and edge features, and generating multi-modal coupling features through combination of a dynamic feature interaction network and a dot product attention mechanism. And constructing a motion trajectory by using the features, optimizing the trajectory by using a texture perception double-layer smoothing strategy, introducing an adaptive penalty term into a dynamic planning cost function, and outputting a smooth motion compensation parameter. And finally, processing the boundary region through motion compensation and image extrapolation to generate an anti-shake video frame. Through multi-scale feature fusion and a self-adaptive smoothing strategy, the video anti-shake effect is effectively improved, and the method is suitable for complex scenes.

The invention relates to the technical field of image processing, and discloses a defect detection method and device for a wafer chip. The method comprises the following steps: preprocessing an original wafer image, wherein the preprocessing comprises at least one of the following items: contrast enhancement, noise suppression and smoothing, sharpening enhancement and normalization processing; dividing the preprocessed wafer image into a plurality of chip areas, and extracting a feature vector of each chip area by using a convolutional neural network to form an initial node feature matrix; based on the spatial position relationship between the chips, constructing a spatial adjacency graph of the wafer; and inputting the initial node feature matrix and the spatial adjacency graph into a defect detection model, and outputting a wafer-level defect detection result. According to the method, comprehensive modeling of defects in spatial distribution and associated feature levels can be realized, and the accuracy and robustness of detection are improved.

The invention relates to the technical field of image recognition, in particular to a blind sidewalk recognition method based on image processing. The method comprises the following steps: collecting a ground scene image, and carrying out noise suppression and illumination normalization processing to obtain a ground scene image to be processed; inputting the to-be-processed ground scene image into a pre-constructed convolutional neural network model to identify feature textures of the blind sidewalk bricks; dividing a blind sidewalk area in the ground scene image to be processed by using the blind sidewalk brick feature texture, and determining a direction gradient feature of the blind sidewalk area; executing blind sidewalk direction consistency constraint based on the direction gradient features to collect blind sidewalk structured path segments; recognizing a blind sidewalk fracture area based on directional gradient features; and performing cross-frame target tracking according to the structured path segment of the blind sidewalk, and outputting a continuous blind sidewalk trajectory. The automatic recognition rate of the blind sidewalk area is improved based on the image recognition technology, and the continuous recognition capacity of the blind sidewalk path in the complex illumination and shielding environment is enhanced.

The invention discloses an image processing method and system for beam crack depth based on MLP optimization, and particularly relates to the field of image data processing based on a neural network, and the method comprises the steps: carrying out the noise suppression, contrast remapping and size normalization processing of original image data, and converting the original image data into an image set with unified spatial distribution; and a structure labeling image set with a crack area label is generated through structure calibration and space mask construction. By constructing a spatial expression vector driven by image state evolution and introducing a variation MLP model with a channel decoupling and path regulation and control structure, crack region depth prediction processing dominated by image structure change is realized, so that the adaptability and prediction stability of the model to an image complex structure are improved, and the prediction accuracy is improved. The problems that an existing method is inconsistent in image semantics, lacks spatial modeling, is insufficient in depth prediction precision and the like are solved.

The invention belongs to the field of instrument identification, and particularly relates to a pressure gauge identification method and equipment in an industrial valve pressure testing process. The method comprises the following steps: extracting an instrument image from a collected video stream, and carrying out distortion correction and illumination normalization processing; feature extraction is carried out on the processed image, instrument types are identified based on a training model of multiple types of instruments, and an instrument image with angle deformation is corrected through a spatial transformation network; performing spatial sorting on the identified instrument images according to the coordinates of the central point of the detection frame, and generating a data array containing instrument information; and finally, performing enhancement processing on a single instrument image in the data array, positioning a dial plate area, performing perspective correction, determining geometric parameters of the dial plate, calculating an original reading based on the parameters, and outputting a final reading in combination with inclination compensation. Through the image processing and recognition steps, accurate recognition and reading of the pressure gauge in the industrial valve pressure testing process are achieved, and the recognition efficiency and accuracy are improved.

The invention discloses an efficient image processing and feature extraction method, which belongs to the technical field of image processing, and can effectively deal with the problems of complex illumination, noise interference, insufficient resolution and the like by introducing the steps of adaptive illumination normalization, multi-channel joint denoising, edge-oriented super-resolution reconstruction and the like. And the image quality is obviously improved. The edge-texture separation and reconstruction strategy not only solves the problem of fuzzy details in a traditional super-resolution method, but also realizes improvement on a PSNR index, and provides a higher-quality data basis for subsequent feature extraction. In addition, fractal dimension driven ROI extraction and bionic PCNN segmentation can accurately position a high-information-content region and realize noise robust segmentation, thereby further improving the accuracy and efficiency of feature extraction.

The invention relates to the technical field of remote sensing image processing, and particularly discloses a photovoltaic power station remote sensing identification method and system based on time constraint, and the method comprises the steps: reading image data, and obtaining projection parameters and geographical transformation parameters; processing the image data by a minimum-maximum normalization method; selecting a reference year, acquiring reference year photovoltaic power station vector data processed by a normalization method, converting image data and vector data processed by the reference year normalization method into the same coordinate system according to projection parameters and / or geographical transformation parameters, rasterizing the vector data, and generating a binary mask consistent with the resolution of the image data; and by taking the binary mask as a label, designing a deep learning network structure to train image data, and outputting a predicted binary mask. According to the method, the uncertainty of single-phase image classification is remarkably reduced, and the stability and reliability of a multi-phase classification result are ensured.

The invention relates to the technical field of image processing, and particularly discloses a mobile terminal real-time image style migration method based on a lightweight convolutional network, and the method comprises the steps: obtaining an original image inputted by a user, and converting the original image into standardized image tensor data through a pixel normalization and dynamic partitioning strategy; performing feature coding on the image tensor data by using a lightweight convolutional neural network, and extracting a multi-scale feature map layer by layer through a deep separable convolutional layer and a bottleneck layer structure; based on dual-channel feature coding, dynamically performing weighted fusion of style features and content features through multi-scale convolution kernel parallel deployment and learnable parameters, and generating a fused feature map; according to the method, the pixel normalization and dynamic partitioning strategy is introduced, the image partitioning size is adaptively adjusted, the calculation efficiency and the feature retention degree are balanced, the calculation burden is reduced while the high image quality is ensured, and the model calculation amount is reduced by adopting the lightweight convolutional neural network and the depth separable convolution in combination with the bottleneck layer structure.

The invention provides a thoracic surgery pathological section image super-resolution reconstruction method, and relates to the technical field of image processing, and the method specifically comprises the steps: collecting a high-resolution thoracic surgery pathological section image, and generating a low-resolution image; constructing a local directional dot matrix tensor and a response adjustment factor based on the low-resolution image, calculating an angle response aggregation kernel tensor, and performing segmented mapping in combination with a nonlinear response reconstruction function to complete directional fusion normalization so as to obtain a directional aggregation enhanced image; a robust local reference value and local texture energy are calculated in a neighborhood range, a structural strength weight is generated in combination with a variance balance parameter and a contrast balance parameter, a difference amplification item is constructed, image fusion with direction aggregation is enhanced through a residual mode, and a structure guide enhanced image is obtained; carrying out weighted mean and variance calculation to generate a difference regulation factor, and combining convolution smoothing and nonlinear amplitude limiting processing to obtain a structure contrast mapping image; and finally, inputting the image into an image reconstruction module to realize resolution improvement.

The invention discloses a multi-target license plate recognition method based on a visual attention mechanism, and belongs to the technical field of image processing and mode recognition, and the method comprises the steps: obtaining an input image, carrying out the multi-scale feature extraction, and generating a multi-scale feature map; performing spatial saliency calculation and normalization processing on the multi-scale feature map to generate an attention map; carrying out region division on the input image, and adopting differentiated image preprocessing strategies for different regions to generate a preprocessed image; based on the preprocessed image and the attention map, multi-target detection is carried out through a target detection network, candidate area screening is carried out, and a candidate license plate area is generated; and performing binarization processing, character segmentation and feature recognition on the candidate license plate region, and performing verification in combination with context information to generate a license plate recognition result. The attention map is generated by adopting a visual attention mechanism, differentiated image preprocessing and multi-target detection are guided according to the attention map, and multi-target license plate recognition can be completed in a complex scene.

The invention discloses a breast cancer neoadjuvant chemotherapy curative effect prediction method and system, and relates to the medical image processing and analysis technology, and the method comprises the steps: carrying out the Macenko dyeing normalization of a sample HE pathological image of a patient; extracting a tissue region from the normalized sample HE pathological image by using an HEST method; inputting the extracted tissue region into a Prov-GigaPath network, extracting a feature vector of a specified dimension under set parameters, and performing dimension reduction on the extracted feature vector; inputting the feature vectors subjected to dimension reduction into an ABMIL model, and training the feature vectors; and predicting the HE pathological image of the patient based on the trained ABMIL model, and visualizing the prediction result by using the attention score of the trained ABMIL model. According to the method provided by the invention, multi-level complex features such as cellular morphology and tissue structure in the HE image can be automatically mined, and the accuracy and reliability of prediction are greatly improved.

The invention discloses a landslide image instance segmentation method based on a dual adaptation mechanism, relates to the technical field of image processing, and ensures that a model can have a better effect and numerical stability on data from different sources through a complete process from multi-source data collection to data normalization processing. The designed model is based on a multi-scale state alignment mechanism, in the forward process of the model, exponential moving average fusion is carried out on feature information in the same level, transmission fusion is carried out on feature information of different levels, feature robustness is enhanced, and error accumulation caused by local deviation is reduced. A meta-context incremental learning mechanism is designed, and input data are dynamically converted into a series of key value vector sequences. In the reasoning process of the model, data distribution different from a training domain is dynamically recognized, a gradient descent process is implicitly executed according to the characteristics of data, and model parameters are finely adjusted, so that the characterization capability is greatly improved, and efficient and robust instance segmentation is realized.

The invention discloses a sports equipment management personnel identification method, system and equipment, and the method comprises the steps: starting and initializing a camera through detecting a sports equipment use trigger signal, and stopping automatic focusing; and shooting multiple frames of images at different focal lengths according to a preset interval, determining a clear area by using a gradient magnitude algorithm, and performing image registration and fusion to form a composite image. And if the whole area is not covered, shooting and splicing are carried out again. And through composite image splicing, distortion cutting, noise reduction and normalization processing, a final image is output for determining the identity of a person. According to the invention, through multi-frame shooting and accurate image processing, a high-quality image is obtained, and information limitation and quality defects of a single image are effectively avoided. Various algorithms are fused to ensure that the image is clear, complete and standardized, the accuracy and reliability of personnel identity recognition are remarkably improved, an accurate image basis is provided for sports equipment management, the normalization and safety of equipment use are guaranteed, meanwhile, the image collecting and processing efficiency is improved, and the management cost is reduced.

The invention relates to the field of digital image processing, in particular to a super-resolution imaging method based on focal plane splicing and adaptive fusion. According to the method, sub-pixel offset among nine CCDs is preset through hardware, and nine frames of low-resolution image sequences with accurate displacement are obtained in push-broom. A central image is taken as a reference frame, high-precision mapping is realized based on hardware offset, motion estimation errors are avoided, effective pixels are screened by calculating robustness weight, an anisotropic Gaussian kernel function with a self-adaptive local structure is constructed so as to maintain image edge and detail features, and each frame is accumulated to a high-resolution grid in a weighting mode, so that a high-resolution image is obtained. And a sample compensation mechanism based on cumulative robustness is introduced, a fusion strategy is adaptively adjusted in an information insufficient area, and finally a high-resolution image is generated through normalization. The method significantly improves the imaging quality, suppresses artifacts and noise, and is suitable for the field of satellite remote sensing.

The invention relates to the field of artificial intelligence and medical image processing, in particular to a mammary gland pathology visual model establishing method based on multi-model fusion and combined distillation, which comprises the following steps: performing tissue segmentation and dyeing normalization on a full-slice image; inputting the image blocks into a pre-training teacher model of a plurality of freezing parameters in parallel to extract high-dimensional features, and generating unified enhanced features through a learnable feature fusion network; constructing a student model, and performing end-to-end training by using a joint loss function including feature simulation, logic output distillation and multi-task supervision; and connecting a plurality of task specific prediction heads to the student model, and realizing full-slice-level multi-task diagnosis and treatment prediction through an aggregation strategy. According to the technical scheme, efficient knowledge migration and multi-task cooperation can be achieved, and the accuracy, generalization ability and reasoning efficiency of mammary gland pathology analysis are remarkably improved.