122 results about "Optical flow estimation" patented technology

The invention belongs to the technical field of computer vision and deep learning, and particularly relates to an optical flow estimation method and system fusing Mama and visual basic model knowledge. The method comprises the following steps: performing down-sampling feature extraction on two adjacent frames of input images by using a convolutional neural network to obtain local texture features; performing down-sampling on the first frame image to obtain context features; meanwhile, extracting global semantic features of two adjacent frames of images by using a pre-trained visual model, and performing adaptive fusion enhancement through an adaptive semantic texture feature fusion module to obtain an image coding feature pair after semantic enhancement; constructing a related volume through pixel-by-pixel dot product operation; and finally, based on the obtained related volume and context features, iteratively optimizing the output optical flow through a loop iteration updating module. The method solves the problems that in a low-texture, repeated-texture or sheltered area, feature expression is unstable, self-adaptive modeling capacity is lacked, different scenes are difficult to generalize, and model performance and efficiency cannot be balanced.

The invention relates to the technical field of computer vision, in particular to a space-time consistency data generation method for visual target tracking. The method comprises the following steps: firstly, training a path generator on a target tracking training set, learning a motion law of a target in a time sequence by using optical flow estimation and conditional variation coding technologies, and generating a target motion track conforming to physical constraints; and then, based on the generated target trajectory, introducing a space-time consistency attention mechanism to guide a text-video generation model, and under the condition of keeping basic model parameter freezing, constraining the position, scale and continuity of a target in a generation frame through an attention network, thereby synthesizing a video frame sequence with real motion features. According to the method, target tracking video data with real motion characteristics and high time sequence consistency is generated, and the robustness of the model to complex motion, illumination change and shielding conditions can be improved in different scenes.

The invention discloses a power grid intelligent inspection method and system based on an unmanned aerial vehicle, and the method comprises the following steps: collecting image flow data of a target region, constructing a three-dimensional point cloud model, and generating an initial inspection path based on a fast marching tree method in combination with spatial position information; controlling the unmanned aerial vehicle to fly according to a path and collect image frames in real time, and executing an optical flow estimation algorithm through an edge calculation chip to obtain a pixel motion vector; associating the motion vector with a space coordinate corresponding to each inspection point in the inspection path, dividing an optical flow detection area and distributing an initial detection weight; recognizing a dynamic abnormal area according to the motion features, extracting image data and space coordinates, and driving an acousto-optic load assembly carried by the unmanned aerial vehicle to respond; and based on the space coordinates of the abnormal region, re-executing the fast marching tree method to generate a local update path, and adjusting the detection weight of the related region. According to the invention, dynamic sensing and path updating linkage in the unmanned aerial vehicle inspection process can be realized.

The invention discloses a lightweight optical flow neural network method for river velocity measurement, and the method comprises the steps: collecting a dynamic monitoring image of a water area surface flow field, calculating an approximate optical flow field through combining with a DeepFlow optical flow algorithm, and generating an optical flow label; an improved lightweight ConvFFN-Flow optical flow estimation model is built, and the model comprises a feature extraction module, a 4D related structure body module, a GRU optical flow residual error iteration module and an up-sampling module; by introducing a ConvFFN module, the model parameter scale and the calculation amount are reduced, and the optical flow prediction precision is kept; and based on pixel coordinates of front and rear frames of images of the optical flow field data, obtaining corresponding positions in a real three-dimensional space, calculating an actual displacement amount, and deducing to obtain a real flow velocity vector size and a flow direction of the water body surface. The method still has stable and high-precision optical flow estimation capability when being used for processing scenes such as fine river water surface texture and complex flow state change, and meanwhile, the calculation overhead and the storage demand are remarkably reduced.

The invention discloses a nut looseness detection method and device based on a mask guided segmentation network and a medium, and the method comprises the steps: image collection and diversified data enhancement; a step of enhancing a multi-path segmentation network by a mask guide boundary; a step of annular and segmented division based on the fine segmentation mask; a step of constructing an optical flow estimation model; a step of obtaining a displacement time sequence and a time sequence loosening characteristic; and judging the loosening state of the nut based on the self-adaptive threshold value. Compared with the prior art, the nut looseness detection method based on the mask guided segmentation network has the advantages that the definition of the segmentation boundary and the sensing capacity of small-amplitude rotation / radial displacement are remarkably improved under the background of strong reflection and noise, short-time random disturbance and continuous looseness can be distinguished, physical marks are not needed, and the detection accuracy is high. The method is suitable for in-service monitoring of wind power equipment and visual state detection of other fastener structures.

The invention discloses an optical flow estimation method, device, equipment and medium, and the method comprises the steps: extracting the multi-scale features of an image through a pyramid network structure, and obtaining the initial optical flow of each level through feature warping processing and cost body calculation; calculating an uncertainty map of each level of optical flow based on the initial optical flow; calculating the fusion weight of the current layer by adopting the uncertainty map of the current layer and the fused uncertainty map of the previous layer, and carrying out weighted fusion to obtain the fusion optical flow of the current layer; calculating the motion gradient consistency loss based on the optical flow gradient difference of the adjacent pyramid layers; constructing a total loss function based on the endpoint error loss, the uncertainty loss and the motion gradient consistency loss of the initial optical flow; and training the optical flow estimation network by using the total loss function to obtain an optimized optical flow estimation model, and performing optical flow prediction on the to-be-estimated image. According to the invention, the overall precision, robustness and interpretability of optical flow estimation in a complex scene are improved.

The application discloses a mixed expert network-based optical flow estimation method and system, belonging to the field of computer vision and artificial intelligence. First, the input two frames of images are preprocessed, low-resolution features are extracted through a mixed expert feature extractor (MoEE) containing a sparse activation mechanism to reduce redundant calculation; then, dot product operation is performed on the low-resolution feature maps to construct a 4D correlation volume to capture pixel motion matching relationship; subsequently, a mixed expert updater (MoEU) is used to iteratively update the hidden state and regress the residual flow increment through a dynamic expert selection mechanism to optimize the optical flow accuracy; finally, a multi-scale upsampling module is used to reconstruct the high-resolution optical flow field to output the high-precision optical flow result. The algorithm realizes dynamic resource allocation through the MoE architecture, significantly reduces the calculation cost while ensuring the accuracy of optical flow estimation, and can adapt to resource-constrained scenarios such as automatic driving and unmanned aerial vehicles, and can balance efficient inference and flexible deployment.

The invention provides a curvature-enhanced large-displacement image variational optical flow method for motion estimation, relates to the field of image processing, and aims to describe the local structure complexity of an image by introducing an image contour curvature. On the basis of the curvature, limited self-adaptive weighted adjustment is carried out on the brightness invariant constraint and the gradient invariant constraint on the data item level of the opto-rheological model, so that the interference of unreliable matching in a complex structure region on optical flow estimation is inhibited; the robustness of optical flow estimation in illumination variation, complex texture and large displacement scenes is improved; and the numerical stability and convergence of the model in the multi-scale calculation process are ensured.

The application provides a curvature-enhanced large displacement image variational optical flow method for motion estimation, relates to the field of image processing, and aims to depict the complexity of local structures of an image by introducing the curvature of the image contour line, and to restrict the adaptive weighting adjustment of the brightness invariable constraint and the gradient invariable constraint on the basis of the curvature in the data item level of the variational model of the optical flow, so as to inhibit the interference of unreliable matching pairs on the optical flow estimation in the complex structure area, improve the robustness of the optical flow estimation under the scenes of illumination change, complex texture and large displacement, and guarantee the numerical stability and convergence of the model in the multi-scale calculation process.

The invention discloses a dynamic environment visual map construction method and system based on OpenSeeD (Open SeeD). Relates to the technical field of computer vision and robot autonomous navigation. Comprising the following steps: acquiring a continuous RGB image sequence of a dynamic object; performing instance segmentation and optical flow estimation on the continuous RGB image sequence to obtain an initial mask and optical flow information; fusing and optimizing the initial mask and the optical flow information, and performing dynamic object recognition on an optimized fusion result to obtain a dynamic mask; and removing the dynamic feature points of the dynamic mask, retaining the static feature points of the static region, and inputting the static feature points into the ORB-SLAN3 library to complete pose estimation and map construction. According to the method, the problem of pose estimation failure or precision reduction caused by feature pollution of the traditional visual SLAM in a dynamic environment can be remarkably improved, and the method has important theoretical value and engineering application prospect in the aspects of reliable autonomous navigation and persistent map construction in a complex dynamic scene.

The invention discloses a frame insertion method and system based on a 3D movie and computer equipment, and the method comprises the steps: obtaining a left 2D video signal and a right 2D video signal in the 3D movie, extracting an input frame signal in each 2D video signal, carrying out the optical flow estimation according to the input frame signals, and obtaining a left optical flow estimation result and a right optical flow estimation result; performing network fitting correction on the optical flow estimation results of the left path and the right path to obtain an optical flow signal with consistent space; according to the left and right paths of corrected optical flow signals, carrying out pixel capture to obtain frame insertion candidate frames; performing joint optimization on the frame insertion candidate frames according to 3D information to obtain an optimized frame insertion result; and fusing the optimized frame insertion result to generate an intermediate frame for frame insertion so as to obtain a high-frame-rate video sequence. According to the invention, the consistency, coherence and robustness of the 3D film frame insertion process can be effectively improved.

The invention provides a millimeter-wave radar personnel perception method based on a time-frequency domain and a deep CNN, and relates to the technical field of radar signal processing, and the method comprises the steps: carrying out the preprocessing, spectrogram conversion and enhancement of a millimeter-wave radar echo signal; spatial features are extracted by using depth separable convolution of a residual structure, time-frequency features are extracted by combining dual-tree complex wavelet transform and attention-enhanced cavity convolution, motion features are extracted through optical flow estimation and three-dimensional convolution, and the three features are adaptively fused; the method comprises the following steps: constructing a dynamic spatio-temporal reasoning network, obtaining key spatio-temporal features by using a recurrent neural tensor network, a non-local neural network and deformable convolution, extracting multi-scale features and performing adversarial feature alignment, and completing personnel target classification through a dynamic routing mechanism based on a capsule network. According to the method, clutter interference can be effectively suppressed, space, time frequency and motion information is fully utilized, accurate perception of personnel targets is realized, and the classification accuracy is improved.

The application provides a millimeter wave radar personnel sensing method based on a time-frequency domain and a deep CNN, relates to the technical field of radar signal processing, and comprises preprocessing, spectrum conversion and enhancement of a millimeter wave radar echo signal; spatial features are extracted by using a deep separable convolution with a residual structure, time-frequency features are extracted by combining a dual-tree complex wavelet transform and an attention-enhanced hollow convolution, and motion features are extracted by optical flow estimation and three-dimensional convolution, and the three are adaptively fused; a dynamic space-time reasoning network is constructed, key space-time features are obtained by using a recurrent neural tensor network, a non-local neural network and a deformable convolution, multi-scale features are extracted and aligned in an adversarial manner, and a dynamic routing mechanism based on a capsule network is used to complete personnel target classification. The application can effectively suppress clutter interference, fully utilize spatial, time-frequency and motion information, realize accurate sensing of personnel targets, and improve classification accuracy.

The invention discloses a multi-mode nuclear magnetic image brain tumor segmentation system based on optical flow method pixel correlation, relates to the field of image processing, and solves the problems that complex physiological correlation among multiple modes cannot be captured and specific characteristics of a tumor area are difficult to reproduce in an existing missing mode completion method. The segmentation system comprises a data preprocessing module, a cross-modal optical flow estimation module, a pixel association weight calculation module, a missing modal complementation module and a fusion segmentation module. The system is simple in structure and reasonable in design, breaks through the limitation that a traditional optical flow model depends on a gray level consistency hypothesis, and adapts to the characteristic that the gray level difference of the multi-mode MRI is remarkable. By capturing the consistency of the gray gradient direction of the same anatomical structure in different modals, cross-modal pixel correlation mapping is accurately established, optical flow estimation deviation caused by gray mismatching is avoided, a reliable correlation basis is provided for subsequent deletion completion, and the accuracy of cross-modal information transmission is guaranteed.

The invention discloses an optical flow estimation method and system based on adaptive flow propagation, and the method comprises the steps: carrying out the feature extraction of adjacent frames of images, and constructing a four-dimensional correlation body between a source frame and a target frame; obtaining an initial optical flow based on the correlation body and a cyclic updating module; a self-adaptive stream propagation module is introduced on the basis, neighborhood expansion is carried out on the current optical stream, a structure guiding weight is generated according to source frame structure features, and weighted aggregation is carried out on neighborhood optical streams to obtain self-adaptive optimized branch optical streams; through structure guiding filtering, local self-attention calculation is carried out on optical flow features, a structure perception mask is injected into key features, structure enhancement filtering is carried out on an optical flow boundary and a shielding area, and a final optical flow result is obtained. According to the invention, the problems of over-smooth optical flow result and mispropagation of the optical flow in the shielding area caused by fixed weight in the existing optical flow estimation algorithm based on flow propagation can be relieved.

The invention relates to the technical field of micro-expression recognition, in particular to a micro-expression recognition method based on motion amplification and hierarchical Transform. According to the technical scheme, the method comprises the following steps: performing motion amplification processing on a start frame and a peak value frame in a micro-expression video sequence to obtain an amplified peak value frame; calculating an optical flow graph between the start frame and the amplified peak frame based on an RAFT optical flow estimation method; and inputting the optical flow diagram into a hierarchical Transform network, and extracting motion relation characteristics among all the parts of the face. According to the method, a weak motion signal of a micro expression is enhanced through a motion amplification technology, key face areas such as eye peripheries and mouth corners are accurately focused by using a local attention mechanism in a hierarchical Transform, multi-scale feature fusion is realized in combination with block aggregation, and finally higher recognition accuracy and stronger generalization ability are verified in a plurality of data sets.

The application relates to the field of image processing and discloses a hybrid neural network-based light flow prediction method for a biomimetic visual sensor, which solves the problems of existing light flow prediction methods, such as the existence of smearing or target loss, the influence on light flow estimation accuracy, the increase in calculation amount, the loss of the low data amount advantage of an event camera and the like, and the method comprises the following steps: firstly, acquiring a public data set on a network and adaptively slicing spatio-temporal data flow; then, constructing a hybrid neural network for light flow prediction of the biomimetic visual sensor, so that the network can perform light flow prediction; then, designing a loss function for overall network training, performing supervised training on the network, and obtaining a supervised light flow estimation model with faster speed and higher accuracy; and finally, using the trained model to perform light flow prediction on the spatio-temporal data flow of the biomimetic visual sensor. The light flow prediction method improves the light flow prediction accuracy and the accuracy of predicted light flow.

The invention relates to the technical field of visual SLAM positioning and mapping, and discloses a visual map data processing method and device, computer equipment and a storage medium. Firstly, panoramic semantic segmentation is carried out on an original image, a panoramic segmented image is generated, the panoramic segmented image comprises semantic information of each pixel, and the semantic information comprises prior dynamic semantic information. And performing optical flow estimation processing on the original image to obtain an optical flow image. Then, based on the panoramic segmented image, the optical flow image and the original image, feature points of a dynamic area are removed in real time, and a static feature image with higher precision is obtained; and finally, under the condition of performing motion tracking based on the static feature image and determining that the static feature image is a key frame, generating a dense point cloud map endowed with semantic information based on the original image, the static feature image and the panoramic segmentation image, and removing a point cloud with prior dynamic semantic information from the dense point cloud map. And a more accurate static environment map is constructed to update the point cloud map.

The application discloses a target tracking and early warning method, system and device, relates to the technical field of computer vision and video analysis, and comprises the following steps: acquiring a target image sequence; based on the target image sequence, performing dense optical flow estimation through a multi-scale feature extraction and iterative optimization mechanism to generate a high-precision dense optical flow field; based on the high-precision dense optical flow field, combining a preset target perception model, a motion prediction mechanism and an adaptive confidence mechanism, and through a preset multi-modal correlation tracking strategy, target trajectory data is obtained; and through a multi-dimensional behavior analysis algorithm, abnormal behavior analysis is performed on the target trajectory data to obtain early warning data. The scheme significantly improves the tracking accuracy and robustness of the target by fusing the motion representation guided by the high-precision optical flow and the multi-cue adaptive tracking mechanism. Meanwhile, through the construction of a multi-dimensional behavior analysis model and a hierarchical early warning decision engine, efficient real-time processing and intelligent accurate early warning are realized.

The invention discloses a panoramic imaging method and device, equipment and a medium, and relates to the technical field of automatic driving, and the method comprises the steps: carrying out the projection transformation of an original image collected by a vehicle camera, and generating an aerial view; determining the size of an adaptive target window according to the distance from each pixel to a central pixel in the aerial view, and performing optical flow estimation on each pixel in two adjacent frames of aerial views in the size of the target window based on vehicle motion information measured by an inertial measurement unit to obtain a dense optical flow field; extracting inner points from the dense optical flow field by using a preset improved RANSAC algorithm, and estimating global motion parameters based on an extracted inner point set; fusing the global motion parameters and motion parameters measured by the inertial measurement unit and the positioning system by using a Kalman filtering method to obtain fused motion parameters; and projecting the historical panoramic image to an aerial view coordinate system by using a homography matrix calculated based on fusion motion parameters, and splicing the historical panoramic image with the current aerial view to generate a new panoramic image.

The application provides a soft tissue real-time tracking and navigation method based on inter-frame deformation field guidance and a related device. The method comprises the following steps: S1. acquiring an intraoperative endoscope image sequence, and extracting a current frame and a historical frame; S2. performing dense optical flow estimation on the current frame and the historical frame to obtain an inter-frame deformation field; S3. performing spatial transformation on a historical mask based on the inter-frame deformation field to obtain a deformation compensation segmentation prior; S4. performing prior guided fusion on the segmentation prior and a current frame feature to obtain a refined anatomical segmentation mask; S5. performing key point matching based on the refined mask and a pre-stored atlas to obtain a spatial registration parameter; and S6. rendering a safety boundary based on the registration parameter and the deformation field, and outputting a safety navigation view. The application also provides a related device corresponding to the method, and the related device comprises a device, an electronic device, a computer readable storage medium and a computer program product.

This invention proposes a smart irrigation alarm and monitoring system, method, device, and storage medium for farmland. It utilizes idle mobile phones as data acquisition devices to collect, report, and preprocess images of farmland irrigation. A deep learning model is used to detect and identify objects in the images, including crops, land, and water flow. Simultaneously, optical flow estimation motion detection technology is used to analyze the water flow state, determine whether irrigation is complete, and report the analysis results to an IoT platform to control the start and stop of networked water meters in farmland wells. Furthermore, this invention can push irrigation completion notifications to user terminals and deduct fees from farmers' water meter accounts. Through these information technology methods, this invention can achieve alarm and monitoring of farmland irrigation at a relatively low hardware cost, significantly improving farmers' productivity.

This invention discloses a motion blur compensation image enhancement method for high-speed cameras, relating to the field of computer vision technology. The method includes acquiring raw sequence data recorded by a high-speed imaging device, analyzing overlapping regions of adjacent frames in the sequence using a temporal convolutional network to determine the temporal continuity features corresponding to highly overlapping frames, obtaining a separated frame sequence, and using an optical flow estimation algorithm to track the object's movement trajectory based on the separated frame sequence. If the optical flow vector exceeds a preset threshold, it is judged as an interference factor in a dynamic scene, resulting in a motion-compensated frame sequence. This motion blur compensation image enhancement method for high-speed cameras not only improves the accuracy of image enhancement but also enables real-time and accurate extraction of key information frames in complex dynamic environments, optimizing the defect detection process. It has broad application prospects, particularly in high-speed motion analysis and industrial inspection, significantly improving detection efficiency and accuracy.

The invention discloses an ultrasonic myocardial movement tracking method based on a data-enhanced end-to-end multi-frame optical flow estimation network. The method comprises seven steps to obtain a forward optical flow of a to-be-estimated multi-echocardiogram. According to the ultrasonic myocardial movement tracking method based on the data-enhanced end-to-end multi-frame optical flow estimation network, segmentation dependence is abandoned, and a data enhancement strategy aiming at ultrasonic image characteristics is introduced, so that the network is automatically focused on a myocardial region in a training process, interference of non-myocardial tissues is effectively inhibited, and end-to-end tracking without segmentation is realized. Meanwhile, on the basis of a network structure of multi-frame optical flow estimation, a motion feature transmission mechanism is introduced, multi-frame motion information can be fused, and cross-period iteration correction can be performed by utilizing the periodicity characteristic of heart motion, so that the limitation of short-time inter-frame tracking is broken through, and the accuracy of heart motion is improved in a complex acoustic environment. For example, stable and accurate myocardial motion estimation can still be obtained under the conditions of low signal-to-noise ratio and fuzzy boundary.

This invention discloses an event-visual-inertial semantic simultaneous localization and map building method, which is applicable to navigation and localization of mobile robots, unmanned vehicles and drones. This method acquires event data streams from event cameras and inertial data from IMUs, and optionally standard camera images. Based on the event data, it constructs an event activity surface, performs coarse-to-fine event corner detection on it, constructs an event representation for event optical flow estimation, and estimates the event optical flow. It then uses the event optical flow to perform temporal tracking of event corners to obtain multi-time-stack related observations. Based on the event corners and event representations, it extracts descriptive information, performs loop closure detection and relocalization, and introduces loop closure constraints as additional residual terms into a sliding window graph optimization. Within the sliding window, it combines residuals from events, images, IMUs, edge detection, and loop closure relocalization, and sets adaptive weights for multi-source fusion to obtain a six-degree-of-freedom pose sequence. It performs target detection on the event data stream, outputting detection boxes, categories, and confidence scores. Based on the event optical flow, it implements missed detection compensation and motion consistency checks to identify dynamic target regions, eliminates dynamic corners and their observation constraints, and generates an object-level semantic map based on pose estimation. This improves the robustness of localization mapping and environmental understanding in complex lighting, high-speed motion, and dynamic interference scenarios.

The invention provides an optical flow estimation method and device, a storage medium and electronic equipment, and relates to the technical field of motion tracking. The method comprises the following steps: acquiring a first original feature map and a second original feature map, wherein the first original feature map and the second original feature map are obtained by respectively performing feature extraction on two adjacent frames of original images; estimating an initial optical flow based on the first original feature map and the second original feature map; selecting a plurality of reference coordinate anchor points based on the initial optical flow, and generating candidate optical flows corresponding to the reference coordinate anchor points; and performing fusion processing on the candidate optical streams to obtain a target optical stream. According to the invention, the precision performance of optical flow estimation under different displacement amplitudes can be improved, and the adaptability to complex motion scenes is enhanced.

The method comprises the following steps: 1) collecting continuous frame images; 2) building an optical flow estimation network, the input of two continuous frame images is respectively extracted by a depth perception feature encoder to obtain a fusion feature map, then zero optical flow is taken as an initial optical flow estimation, and multiple iteration optimizations are carried out: in each iteration, a self-adaptive feature alignment module is entered for feature alignment, a global-local collaborative refinement module is entered for global-local feature collaborative refinement, and an optical flow prediction module is entered for optical flow updating, the optical flow of one-half resolution of the input image is up-sampled to the original image resolution, and a final optical flow map is output; 3) training the optical flow estimation network; 4) inputting the continuous frame image pairs in a test set into the optical flow estimation network for optical flow estimation. The application alleviates the feature matching difficulty in strong occlusion and large range weak texture area, and improves the precision and robustness of optical flow estimation.

The present application relates to the technical field of computer vision, and particularly relates to a spatiotemporal consistency data generation method for visual target tracking. Firstly, the present application trains a path generator on a target tracking training set, learns the motion law of the target in a time sequence by using optical flow estimation and conditional variational encoding technology, and generates a target motion trajectory conforming to physical constraints. Then, based on the generated target trajectory, a spatiotemporal consistency attention mechanism is introduced to guide a text-video generation model, and under the condition of keeping the parameters of the basic model frozen, the position, scale and continuity of the target in the generated frame are constrained by an attention network, so as to synthesize a video frame sequence with real motion characteristics. The present application generates target tracking video data with real motion characteristics and high temporal consistency, and can improve the robustness of the model to complex motion, light changes and occlusion conditions in different scenes.

The embodiment of the invention provides a driving scene image generation method and device based on Mama, electronic equipment and a storage medium, and relates to the technical field of image processing. The method comprises the following steps: performing coding quantization processing on an image sequence in an acquired real driving scene to obtain a discrete potential representation sequence, performing optical flow estimation and feature tracking on the image sequence to generate a sparse motion trail diagram of the image sequence, selecting multiple frames of images from the image sequence as key frames, and obtaining a sparse motion trail diagram of the image sequence; extracting guide pixels from each key frame based on the sparse motion trail diagram, inputting the sparse motion trail diagram and the guide pixels into a double-branch conditional encoder to obtain a fused conditional vector, inputting the potential representation sequence and the conditional vector into a Mamba model to generate a predicted potential representation sequence with a target length, and decoding the predicted potential representation sequence to obtain a predicted potential representation sequence with a target length. And obtaining a predicted image sequence. Therefore, the sense of reality and the dynamic continuity of the predicted image sequence can be obviously improved.

This invention relates to the field of video image processing technology and discloses a method for denoising high-speed motion continuous frames. The invention receives a raw video sequence in Bayer format and establishes a multi-scale image pyramid with a variance-stabilized domain. At each level of the pyramid, based on local brightness statistics, local texture contrast information, and estimated motion amplitude, a motion texture discriminability index representing texture change relative to noise intensity is calculated. This index is used to dynamically adjust the data fidelity term and smoothing constraint term in the variational optical flow estimation process to obtain a motion-compensated image. Subsequently, this index is mapped to an alignment deviation estimate and a similarity measurement range, multi-frame spatiotemporal fusion weights are calculated, and a primary fusion image is generated. Finally, an adaptive filtering threshold is calculated based on this index, high-frequency texture details are extracted from the differential data and superimposed back onto the primary fusion image, and a denoised video sequence is output after inverse transformation processing.