54 results about "Resolution recovery" patented technology

The invention discloses a low-illumination image enhancement method based on a YCbCr space and a Fourier frequency domain, and the method comprises the steps: extracting a brightness component of a low-illumination image in the YCbCr space, and constructing a training set together with a normal-illumination image and the low-illumination image; constructing a low-illumination image enhancement network model, and inputting the training set into the model for training; constructing a loss function between the amplitude enhancement network branch and the normal illumination image as well as between the phase enhancement network branch and the normal illumination image, and adjusting model parameters by minimizing the loss function to obtain an optimization model; and inputting a to-be-enhanced low-illumination image into the optimization model for image enhancement. According to the method, the YCbCr brightness component and the amplitude information of the Fourier frequency domain are combined, and the Transform is utilized to enhance the phase information, so that the visual quality of the low-illumination image is remarkably improved, the visibility and definition of the image are improved, and the method has wide application prospects in tasks of image enhancement, denoising, super-resolution recovery and the like.

The invention discloses an urban green plant identification method and device, a medium and equipment, and relates to the technical field of image identification, and the method comprises the steps: obtaining a to-be-identified remote sensing image; deep features in the remote sensing image are extracted, multi-scale spectral features in the deep features are obtained, the deep features and the multi-scale spectral features are fused, and a multi-spectral feature map is generated; wherein the deep features comprise color features and texture features; extracting strip features and multi-scale shape features in the remote sensing image, and fusing the strip features and the multi-scale shape features to generate a multi-shape feature map; performing weighted fusion on weights corresponding to the multi-spectral feature map and the multi-shape feature map to determine a fused feature map; and recovering the resolution of the fused feature map to the size of the original image, and classifying the feature map with the recovered resolution to obtain a final urban green land classification result.

The invention discloses a power line image super-resolution recovery system and method for power transmission line patrol, and relates to the technical field of unmanned aerial vehicle image processing. The system provided by the invention comprises three core modules, namely a physical perception generator, a differentiable physical discriminator and a spatial transformation network (STN), wherein the generator is combined with a U-Net and a residual structure, and the output is ensured to accord with the geometry and material characteristics of a power line through physical constraint loss; the discriminator adopts a double-branch structure, and synchronously evaluates image authenticity and physical rationality; and the STN module corrects perspective distortion by using the pose data of the unmanned aerial vehicle to realize geometric alignment of the power line. The system integrates advanced components such as Swin Transformer, ViT and differentiable RANSAC, supports end-to-end processing from original Bayer data to a high-definition image, and outputs a defect detection confidence map with enhanced physical constraints. Compared with a traditional data driving method, the method strictly guarantees the physical credibility of a reconstruction result while improving the resolution, and is suitable for a power line inspection task in a complex environment.

The invention relates to a high-fidelity generative video stream transmission system based on a visual base model, which belongs to the field of image communication, and is characterized in that a visual enhancement-oriented generative codec is designed, and high-fidelity video reconstruction under a high compression ratio is realized through an asymmetric space-time compression strategy and time sequence consistency enhancement; a resolution scaling module is provided, the calculation complexity is remarkably reduced through a video super-resolution recovery module of adaptive resolution control and joint optimization, and real-time high-definition video processing is achieved; and constructing a network adaptive video stream transmission controller, an intelligent token discarding mechanism based on semantic importance and a mixed packet loss processing strategy to realize code rate scalable control and robust transmission under network fluctuation.

The present disclosure relates to methods, apparatus, systems, and non-transitory computer-readable storage media for video coding using super-resolution restoration with residual frame coding. According to some examples, a computer-implemented method includes receiving a coded frame of a video; performing a video coding on the coded frame of the video to generate a resultant for the coded frame at a second lower resolution than a first resolution; upsampling the resultant in at least a vertical direction to a higher resolution than the second lower resolution to generate an upsampled resultant; generating a decoded frame based on at least the upsampled resultant; and transmitting the decoded frame to a frame buffer or to a display device.

An information processing system includes a processor. The trained model is trained to resolution recover a low resolution training image generated by low resolution processing performed on a high resolution training image to a high resolution training image that represents a high resolution image captured with a predetermined object through the first imaging system. The low resolution processing represents processing that generates a low resolution image as if captured with the predetermined object through the second imaging system and processing that simulates the second imaging method, and includes processing that simulates a resolution characteristic of an optical system of the second imaging system. The processor uses the trained model to resolution recover the processing target image captured through a second imaging system to an image having a resolution at which the first imaging system performs imaging.

This application discloses a landslide segmentation method, apparatus, equipment, and medium based on hybrid Mamba and frequency domain calibration, relating to the field of landslide disaster detection technology. The method involves fusing optical satellite imagery with slope and lithological topographic constraints, preprocessing the data, and then inputting it into a model containing a hybrid Mamba encoder, a multi-scale adaptive gating module, and a progressive frequency domain calibration decoder for processing. Through multi-stage hybrid coding, contextual aggregation modulation, and progressive resolution recovery, accurate landslide segmentation is achieved, effectively improving global feature modeling capabilities, enhancing the identification accuracy of multi-scale and boundary-ambiguous landslides, balancing lightweight design and robustness, and better adapting to the actual monitoring needs of highway landslides.

The present invention provides an ATFormer architecture for medical image segmentation, comprising a patch partitioning layer, an encoder, a bottleneck layer, a decoder, and a linear projection layer; the patch partitioning layer divides an input medical image into non-overlapping patch images; the encoder generates feature maps at different resolutions from the non-overlapping patch images, wherein the resolution of the feature maps is reduced and the feature dimension is increased; the bottleneck layer performs deep feature learning on the feature maps; the decoder gradually upsamples the feature maps at different resolutions from the encoder under the joint action of a progressive guided fusion (PGF) module, wherein the resolution of the upsampled feature maps is increased and the feature dimension is reduced, and the resolution of the feature maps is restored to the original resolution; and the linear projection layer outputs the feature maps output by the decoder as pixel-by-pixel segmentation prediction results. The present invention provides a new ATFormer architecture that can be used for medical image segmentation and provides accurate segmentation results.

This invention relates to a dual-branch remote sensing image semantic segmentation method and system based on Mamba and adaptive convolution, belonging to the fields of remote sensing image semantic segmentation and artificial intelligence technology. This method aims to solve the problems of existing technologies, such as difficulty in coordinating global dependencies and local texture information, low segmentation accuracy at multiple scales, poor multi-band adaptability, and low training efficiency. The invention includes using a multispectral adaptive processor for spectral grouping and attention enhancement, inputting the data into a dual-path backbone network after downsampling through a convolutional backbone. The global path uses a Mamba module to achieve linear complexity global modeling, while the local path uses adaptive convolution to dynamically adjust weights to capture details. A multi-head cross-attention fusion module then enables bidirectional interaction of multi-scale features, and a feature pyramid network performs cross-scale optimization and resolution restoration. This invention significantly improves segmentation accuracy, reduces computational burden, and adapts to the real-time application requirements of high-resolution remote sensing images.

The application discloses an underwater lining plate crack detection method and device based on image recognition, and relates to the technical field of underwater lining plate cracks. Standardized underwater lining plate image data is obtained; an initial spatial feature map is output; enhanced sequence feature representation is obtained; a global context enhanced feature map is generated; a first resolution recovery stage, a second resolution recovery stage and a third resolution recovery stage are sequentially executed; a first-level high-resolution feature map is generated; a second-level high-resolution feature map is generated; a third-level high-resolution feature map after adaptive rearrangement is generated; a crack segmentation prediction map is generated; a connected region label result is output; and underwater lining plate crack detection report data is output. The application can more completely express the long-distance continuous structure characteristics of underwater lining plate cracks, improve the identification capability of weak cracks, intermittent cracks and low-contrast cracks, and reduce the occurrence probability of crack breaking, missing detection and local false segmentation.

A deep learning-based automatic segmentation method for diabetic retinopathy fundus images includes the following steps: S1 resizing the diabetic retinopathy fundus image to 512*512, S2 starting image preprocessing, S3 passing the preprocessed image into a network model for node feature initialization, S4 feature transfer and fusion, S5 each node passing features to the upper layer node, segmenting the vascular structure in the image through deconvolution / interpolation / depooling methods, and when the last node of the first layer completes feature fusion, the resolution is restored to the same size as the original input image. S6 After the vascular structure segmentation is completed, a grayscale image of the vascular structure segmentation is output through a single layer of convolution. The method provides automatic segmentation of fundus vascular structures for medical assistance systems, solving the problems of slow image reading, unclear vascular ends, and incomplete segmentation caused by the easy loss of new small blood vessels. The present invention has the advantages of automatic extraction of fundus blood vessels, low computational resource consumption, and high segmentation accuracy.

The application discloses a kind of based on multi-scale feature refinement network's image super-resolution construction method, comprising the following steps: step one: using DIV2K dataset constructs high-resolution and low-resolution image pair, and the generalization ability of network model to diversified image features is improved by data enhancement technology;Step two: construct multi-scale feature refinement network: the network includes detail extraction engine, global multi-scale context module and feature refinement module;Step three: the parameter of multi-scale feature refinement network model is optimized by minimizing L1 loss function to train;Step four: using the network model trained to low-resolution image carries out super-resolution reconstruction, generates high-resolution image output;The model is evaluated on test dataset.The application introduces detail extraction engine (DEE), global multi-scale context module (GMCM) and feature refinement module (FRM), enhances cross-region feature interaction and the limitation in the aspect of detail in single image super-resolution recovery.

The application discloses a gas diffusion layer fiber binder segmentation method based on global and local feature fusion, and aims to accurately segment three types of regions of carbon fibers and phenolic resin binders in a gas diffusion layer in a SEM image. The method first collects SEM images of the gas diffusion layer with different phenolic resin loadings through an SEM (scanning electron microscope), and obtains an image data set after pretreatment, manual labeling and data enhancement and the like; then, an improved neural network model containing a GLIB global-local aggregation module, an LBWA boundary weight learning feature interaction module and an LKAT large kernel convolution attention module is constructed, global features are extracted by using a Transformer encoder, key region feature expression is enhanced in combination with multiple modules, multi-scale feature fusion and resolution recovery are completed through a decoder, and finally, a segmentation result is output.

The embodiment of the invention provides a wafer image matching method and device and a computer storage medium, and the method comprises the steps: firstly carrying out the down-sampling of an obtained to-be-detected and template image, and rapidly calculating an initial matching position in a low-resolution space, so as to greatly reduce a search range; performing up-sampling on the low-resolution to-be-detected image by using a pre-trained neural network, and reconstructing a detail-enhanced high-resolution image; and finally, performing fine matching in an initial position neighborhood based on the high-quality image, and determining an optimal sub-pixel position. According to the method, the search space is efficiently compressed through low-resolution rough matching, the overall calculation complexity is remarkably reduced, and the speed is increased; meanwhile, by means of super-resolution recovery driven by the neural network, the problem that details of a traditional interpolation method are fuzzy is effectively solved, high-fidelity image data are provided for fine matching, and therefore the sub-pixel-level matching precision is guaranteed.

The application discloses a video super-resolution method based on space-time convolution attention, and is specifically implemented according to the following steps: step 1, converting an input high-resolution video sequence into a low-resolution video sequence, so as to obtain a low-resolution video set; step 2, constructing a deep neural network structure for video super-resolution; step 3, training the network based on paired video data; step 4, video super-resolution reconstruction, so as to realize super-resolution recovery of the input video sequence. The application solves the problem in the prior art that the network is regarded as a black box, performance is improved by increasing the network depth and parameter scale, the rapid growth of the calculation complexity is ignored, and the efficiency and the reconstruction quality are difficult to be considered.

The invention provides an image data processing method and device and electronic equipment, and belongs to the technical field of image data processing, and the method comprises the steps: obtaining a resolution reduction image of an original image, and original image data of the original image in a first preset region; performing sequence-variable combination processing on the resolution-reduced image to generate an intermediate image of which the resolution is the same as that of the original image; wherein the combination processing comprises fuzzy processing and resolution recovery processing; replacing the image data corresponding to the first preset area in the intermediate image with the original image data to generate a target image; according to the technical scheme, the data transmission amount of the display device when the display device displays the high-definition image can be reduced, key information of the original image is reserved in a partition processing mode, and the experience feeling of a user is improved.

A source video is encoded in a streaming protocol comprising a normal stream and a companion stream. The normal stream contains I-frames and predicted frames at high resolution. The companion stream contains frames at a lower resolution. When needed, an I-frame from the companion stream may be decoded and upscaled from the second resolution to the first resolution and injected into a decoded picture buffer. At an interval corresponding to an amount of time that is shorter than an interval between I-frames in the normal stream, a first frame is downscaled to the second resolution and then upscaled again to the first resolution. The next frame is then encoded with reference to the upscaled frame such that an output stream recovers to the first resolution based on the second frame prior to receipt of the next I-frame in the normal stream.

The application provides an interventional medical instrument image segmentation method and system, comprising the following steps: acquiring DSA image data from an interventional surgery case database system and dividing the DSA image data into a training set and a test set; building a Unet model, converting an input picture into a semantic feature map through an Encoder module, converting the semantic feature map into a binary classification result through a Decoder module, completing resolution recovery, and obtaining a final segmentation result map; defining a weight function during the training process; defining a new energy function in combination with cross entropy and softmax; and fusing and mapping the segmentation result of the model to a DSA original map and highlighting the DSA original map. The application greatly improves the precision and stability of surgical operation by means of computer and robot technology, and can effectively reduce the harm of radiation to interventional doctors and the probability of intraoperative accidents.

The invention discloses a sock production line defect monitoring method and system based on image recognition, and relates to the technical field of industrial defect detection.The sock production line defect monitoring method comprises the steps that an elastic deformation manifold is constructed based on entanglement feature vectors, and a Lagrange function is established on the elastic deformation manifold; solving a differential homeomorphic mapping function on the elastic deformation manifold through a Lagrange function, and performing deformation compensation on the entanglement feature vector according to the differential homeomorphic mapping function; performing multi-scale feature extraction on the entangled feature vector after deformation compensation by using a separable quantum convolution kernel to obtain a defect feature map, and performing spatial resolution recovery on the defect feature map by using a deconvolution layer to generate a defect type and a position coordinate; according to the method, by constructing the elastic deformation manifold and solving the differential homeomorphic mapping function, the spatial continuity characteristics of deformation are captured, coordinate transformation of entangled characteristic vectors is achieved, and the problem that non-rigid deformation of socks in high-speed movement lacks a compensation mechanism is solved.

A technique is disclosed for analyzing and displaying the extent to which the images and structures inferred by a physically seeded multiscale network correspond to genuine resolution improvement and the extent to which they correspond to the hallucination of realistic looking structures. A selected reconstruction generated using a trained neural network is compared against a baseline representation (e.g., a baseline reconstruction) by calculating image similarity metrics between progressing spatially filtered versions of selected reconstruction. As the amount of spatial filtering increases, at some point the image similarity metrics will reach an extrema (e.g., a lowest distance between the images). At that extrema, the parameter(s) of the spatial filter can be used to identify a resolution score (e.g., a length scale) associated with that extrema. The resolution score is indicative of an amount of resolution recovery associated with the trained neural network with respect to the baseline.

The invention provides methods and systems pertaining to a method for recovering resolution of spectral base images acquired using spectral CT imaging techniques. In particular, embodiments aim to provide a method for generating resolution recovered spectral base images of a field of view of a high resolution based on low-resolution spectral base images of the field of view of a first resolution and a high resolution CT image of the field of view of a high resolution.

The application discloses a kind of urban green plant identification method, device, medium and equipment, it is related to image recognition technical field, including obtaining remote sensing image to be identified;Deep feature in remote sensing image is extracted, obtain multiscale spectral feature in deep feature, deep feature and multiscale spectral feature are fused, generate multispectral feature map;Wherein, the deep feature includes color feature and texture feature;Strip feature and multiscale shape feature in remote sensing image are extracted, and strip feature and multiscale shape feature are fused, generate multi-shape feature map;By corresponding weight of multispectral feature map and multi-shape feature map is weighted fusion, determine the feature map after fusion;The resolution of the feature map after fusion is restored to original image size, then the feature map of recovery resolution is classified, obtains the final urban green classification result.

Techniques are disclosed for analyzing and displaying the extent to which images and structures inferred by a physical seed multi-scale network correspond to true resolution improvements and their extent to which they correspond to illusions of realistic appearance structures. A selected reconstruction generated using a trained neural network is compared to a baseline representation (e.g., a baseline reconstruction) by calculating an image similarity measure between progressive spatially filtered versions of the selected reconstruction. As the amount of spatial filtering increases, the image similarity metric will reach an extreme value (e.g., the lowest distance between the images) at a moment. At the extremum, a parameter of a spatial filter can be used to identify a resolution score (e.g., a length scale) associated with the extremum. The resolution score indicates a resolution recovery amount relative to a baseline associated with the trained neural network.

The application relates to a multi-dimensional grid data compression method, system, device and storage medium, and relates to the field of data processing. The method is based on a Transformer super-resolution network, wherein the method comprises the following steps: after original floating-point scientific data is mapped into a grid image sequence, spatial downsampling preprocessing is performed; the down-sampled image sequence is subjected to time series compression coding based on inter-frame prediction to remove time redundancy; in the decoding stage, a super-resolution reconstruction model is used to restore the resolution of the compressed image; further, the error between the reconstructed data and the original data is calculated, the pixel points with errors exceeding a preset threshold are identified, and the corresponding pixel values are replaced with the original data values; meanwhile, the spatial positions and error information of the pixel points are subjected to entropy coding and stored together with the compressed data. The technical effect of the application is that the data compression rate is significantly improved, the reconstruction error is strictly constrained, and the application is suitable for the storage and transmission of remote sensing and scientific calculation data with high precision requirements.

The application provides an image reconstruction method, system and medium of a compressive sensing lens based on a coded aperture, the method comprising: acquiring an original scene image, performing projection modulation to obtain an image received by a sensor; constructing a sensing matrix and a projection matrix, mapping a transfer function of a mask based on the sensing matrix, performing spatial coding sampling on the image received by the sensor through the projection matrix to obtain a compressed measurement result; constructing a deep learning reconstruction network, processing the compressed measurement result based on the deep learning reconstruction network to obtain a two-dimensional feature map; capturing different subspace information based on the two-dimensional feature map through a multi-head attention mechanism, restoring the image resolution based on the different subspace information, and outputting a reconstructed image; and using the mask to realize physical compression, greatly reducing the data acquisition amount, restoring information through the deep learning reconstruction network, improving the image reconstruction efficiency and sensing quality, and adapting to multiple scene environments.

To provide an imaging apparatus capable of acquiring a higher quality image by reducing the influence of an image magnification fluctuation.SOLUTION: An imaging system includes an imaging apparatus 1 and a lens device 31 and performs focus adjustment by the control of the focus adjustment operation of an imaging optical system included in the lens device 31. The imaging apparatus 1 acquires AF (autofocus) information or gyro information and calculates the shake amount in the optical axis direction of the imaging optical system (S202). The imaging apparatus 1 detects an image magnification change when performing the focus adjustment (S205) and calculates the degree of recovery of image resolution, to perform setting for post-photographing processing (S206). The imaging apparatus 1 determines whether to fix or adjust the focus on the basis of an image magnification change amount and the degree of recovery of the image resolution (S207).SELECTED DRAWING: Figure 2

The present application relates to a kind of real-time fire monitoring image enhancement systems, including noise removal sub-model and resolution recovery sub-model;The noise removal sub-model removes the noise of original fire monitoring image, and the image after processing is input resolution recovery sub-model, and resolution is promoted and image is reconstructed.The present application improves the image definition and resolution, and guarantees monitoring quality.

The invention provides methods and systems pertaining to a method for recovering resolution of spectral base images acquired using spectral CT imaging techniques. In particular, embodiments aim to provide a method for generating a resolution recovered spectral base image of a field of view of a third resolution based on a low-resolution spectral base image of the field of view of a first resolution and a high resolution CT image of the field of view of a second resolution, where the first resolution is lower than the second and third resolutions.

The invention discloses a remote sensing visual semantic segmentation method, device and system, and a storage medium. The method comprises the following steps: S1, extracting text features and visual features of a remote sensing image; s2, obtaining a target thermodynamic diagram and a random shielding text according to the text features and the visual features; s3, performing reconstruction processing on the target thermodynamic diagram through the random shielding text to obtain an updated target thermodynamic diagram; and S4, performing multi-scale fusion and resolution recovery according to the updated target thermodynamic diagram and cross-modal features. By adopting the technical scheme of the invention, the problems of insufficient cross-modal alignment capability, low complex scene multi-scale target segmentation precision and inaccurate boundary fuzzy region prediction of the current remote sensing visual language segmentation (RRSIS) are solved.

A technique is disclosed for analyzing and displaying the extent to which the images and structures inferred by a physically seeded multiscale network correspond to genuine resolution improvement and the extent to which they correspond to the hallucination of realistic looking structures. A selected reconstruction generated using a trained neural network is compared against a baseline representation (e.g., a baseline reconstruction) by calculating image similarity metrics between progressing spatially filtered versions of selected reconstruction. As the amount of spatial filtering increases, at some point the image similarity metrics will reach an extrema (e.g., a lowest distance between the images). At that extrema, the parameter(s) of the spatial filter can be used to identify a resolution score (e.g., a length scale) associated with that extrema. The resolution score is indicative of an amount of resolution recovery associated with the trained neural network with respect to the baseline.