19 results about "Fuzzy graph" patented technology

The application discloses a kind of unmanned aerial vehicle image enhancement method and system.Therein, the method includes: obtaining first fuzzy image and its corresponding semantic prompt text;The structural feature extraction is carried out to first fuzzy image, and the semantic feature extraction is carried out to semantic prompt text, to obtain prior information;Prior information is injected into denoising model using cross attention mechanism;First fuzzy image is input into the denoising model injected with prior information to carry out multiple rounds of training and obtain target denoising model;Second fuzzy image is obtained, which is subjected to frequency domain transformation and input into frequency domain residual error model for multiple rounds of training to obtain target frequency domain residual error model;The fuzzy image to be detected is input into target denoising model for denoising to obtain spatial clear image;The fuzzy image to be detected is input into target frequency domain residual error model for prediction to obtain frequency domain clear image;Spatial clear image and frequency domain clear image are fused to obtain high-resolution image.The method can obtain higher quality clear image.

The application belongs to the technical field of simultaneous localization and mapping, and particularly relates to a SLAM algorithm based on feature reinforcement and motion judgment in a dynamic scene, which applies a feature reinforcement instance segmentation network FENET and comprises the following steps: step S1, collecting image information and realizing feature recovery of a dynamic fuzzy object through a fuzzy feature recovery module; step S2, guiding a model to focus on key features of an object based on a reinforced feature recognition mechanism, and recognizing potential dynamic objects; and step S3, jointly estimating the pose of a camera itself and judging the motion of an object to remove a dynamic object. The application can reconstruct and recover lost feature information from a fuzzy image, greatly improves the recognition accuracy of a system for a dynamic object, greatly improves the recognition accuracy of a dynamic object, and avoids misjudgment of static features.

The invention discloses a small sample hyperspectral image classification method and device based on fuzzy contrast graph convolution. The method comprises the following steps: determining membership information, fuzzy node initial feature vectors and similarity of each initial graph node corresponding to each fuzzy cluster based on a hyperspectral image; generating a fusion weight, a fusion feature and a target feature according to convolution features output by convolution of different layers of fuzzy images; performing linear transformation operation and probability distribution operation on the target features in sequence to obtain the probability that the corresponding initialized graph node belongs to each category; and performing argmax operation on the probability to obtain a category index corresponding to the hyperspectral image. The uncertainty of pixels is effectively processed through fuzzy learning, interference of neighborhood pixels is weakened, the uncertainty between the pixels is better processed, and meanwhile richer and more expressive feature representation is obtained. The method has higher classification effect and robustness, and plays an important role in multiple fields of mineral exploration, environment monitoring, forest management, precision agriculture and the like.

The invention relates to the technical field of image display, in particular to a method and device for generating a fuzzy layer of an auxiliary screen, a display screen and a vehicle. The screen is divided into a main screen and an auxiliary screen, the auxiliary screen is simulated through a virtual screen, and the auxiliary screen covers display content, located in an auxiliary screen display area, of the main screen; under the condition that a first image layer displayed on the auxiliary screen comprises a fuzzy image layer, the method comprises the steps that a second image layer which is located below the fuzzy image layer and has an intersection with the fuzzy image layer is determined, and the second image layer is an image layer on the main screen; under the condition that the second image layer meets a first condition, performing fuzzy processing based on rendering information of an application corresponding to the second image layer to generate a fuzzy image layer; and under the condition that the second image layer meets a second condition, obtaining a snapshot of the main screen, and performing fuzzy processing on the snapshot to generate the fuzzy image layer.

The application relates to the technical field of vehicle speed identification methods, and discloses a fuzzy image frame analysis vehicle speed identification method based on field reproduction, wherein, according to the differences of a camera, a field environment and road conditions, a field reproduction positioning measurement method in the fuzzy image vehicle speed identification method is divided into a direct measurement method and an indirect measurement method. Through the use of the method, the limitations and deficiencies of target vehicle calibration distance selection in the fuzzy video image for vehicle driving speed identification are overcome, the problems of identification failure and inaccuracy are solved, most of the video image identification materials which cannot be identified by using the traditional method regain the use value, the problems caused by the defects of the identification materials are solved, the requirements for the identification materials are reduced, the problems of fruitless vehicle speed identification and inaccurate vehicle speed identification results are solved, the scientificity and accuracy of the identification opinions are improved, and strong evidence support is provided for the traffic police department for accident handling.

This invention proposes a dynamic scene deblurring method and system based on 4D Gaussian representation and pseudo-ground value supervision. The method includes: obtaining a sharp image and a rendered depth map from a 4D Gaussian representation; predicting the motion velocity of dynamic pixels using a velocity MLP network based on the rendered depth map; constructing a fuzzy weight network and predicting the contribution weight of each sampling point on the sampling trajectory; fusing the pseudo-sharp image with the original blurred input using a dynamic region mask to obtain a hybrid pseudo-ground value image; synthesizing a physically blurred image through weighted integration; and constructing a dual-domain reconstruction loss and geometry-motion regularization constraints by combining hybrid pseudo-ground value supervision and gradient decoupling strategies to jointly optimize the scene representation, motion parameters, and fuzzy model, thereby obtaining a sharp dynamic scene representation. This invention achieves explicit modeling of the physical blurring process through 4D Gaussian representation and a fuzzy weight network, improving the realism and view consistency of motion blur synthesis.

The invention provides an automatic commodity scene supplementing method and device based on a BAGEL model, equipment and a medium, and the method comprises the steps: carrying out the processing of a set commodity image through employing a BiRefNet model, and obtaining a commodity main body image; constructing a preset cue word template, inputting the commodity main body graph into the multi-modal model, and generating a corresponding scene description cue word according to the preset cue word template; inputting a commodity main body graph and the scene description prompt word into a BAGEL model, wherein the BAGEL model generates a required scene commodity graph; determining Gaussian filtering kernel parameters according to the size of the scene commodity image, converting the scene commodity image into a scene commodity grey-scale image, and performing Gaussian blur on the scene commodity grey-scale image through the Gaussian filtering kernel parameters to obtain a scene commodity fuzzy image; and mixing the commodity graph and the scene commodity fuzzy graph according to a soft light mode to obtain a scene commodity final graph, so that the proportion of the commodity in the scene is balanced, and the composition coordination is improved.

The application relates to a spaceborne SAR signal post-facto fine analysis system and belongs to the field of radar signal reconnaissance; can complete time-frequency analysis and PDW parameter measurement functions; can perform pulse rapid retrieval based on a time scale, realizes fine analysis on a single pulse signal, and includes a fuzzy diagram, a spectral peak degree coefficient and the like; has a radiation source signal sorting function, can manually add a radiation source database, can automatically store a radiation source result obtained through detection and collection into a warehouse, and can sort known radiation sources according to the warehouse; GPU is adopted to accelerate processing of an analysis algorithm, time-frequency analysis speed can reach 1.6GS / s, and PDW measurement speed can reach 80MS / s; the whole processing system is completed in a 2U server disk array all-in-one machine, the system has high integration degree and full functions.

The application provides a scanning file fuzzy retrieval self-adaptive splicing method, and relates to the technical field of document digitization and computer vision. The method comprises the following steps: obtaining a plurality of fuzzy image blocks obtained by non-fixed block scanning of a long document by a mobile terminal, wherein the long document comprises a layout structure composed of text lines and lines; performing local fuzzy retrieval on each fuzzy image block to extract local features of each image block, and determining an initial pose relationship and an overlapping area between the image blocks according to the local feature matching; determining a local rigid texture primitive in the overlapping area of each image block; performing preliminary splicing on the plurality of fuzzy image blocks according to the initial pose relationship and the overlapping area to obtain a preliminary splicing image, and determining a global topological constraint manifold feature point in the preliminary splicing image. The application can realize accurate splicing of fuzzy image blocks and layout structure restoration.

The application provides a kind of image visual identification processing method for foundation pit monitoring, belong to data processing technical field, it includes: step 1: to the foundation pit carries out multimodal image and motion information acquisition;Step 2: background server carries out multimodal feature fusion and fuzzy image screening to foundation pit image;Step 3: background server carries out rigid-nonrigid structure classification and feature enhancement based on multimodal feature map;Step 4: background server constructs dynamic fuzzy kernel and self-adapting deblurring for rigid structure area and nonrigid structure area;Step 5: based on complete deblurring image, carry out fine crack identification of foundation pit.Through four core steps of multimodal feature fusion extraction, dynamic fuzzy kernel construction, nonrigid structure enhancement and fine crack identification, the efficient deblurring and crack accurate identification of foundation pit image are realized.

The invention discloses a low light blurring image processing algorithm based on an improved LEDNet model, and belongs to the technical field of image processing, and the algorithm comprises the following steps: S1, constructing a low light blurring-clear image data set fitting a police service scene; s2, constructing an improved low-light blurred image processing network based on the LEDNet; s3, a multi-task loss function is designed, the LEDNet network is improved, and the multi-task loss function comprises pixel-level loss, perception loss and improved structural similarity loss; and S4, performing model training and image processing, and training the improved LEDNet network based on the data set. According to the low-light blurred image processing algorithm based on the improved LEDNet model, the processing effect of the low-light blurred image is improved, a new scheme is provided for landing application of the deep learning technology in public security actual combats, and meanwhile reference is provided for design of a police service adaptive multi-task image processing model.

The application provides a kind of detection method, system, storage medium and computer of vehicle plastic parts, the method comprises: real-time acquisition vehicle plastic parts is detected image, and corresponding separation threshold is constructed, and according to separation threshold, feature separation is carried out to the image to be detected to obtain feature image;Increase regression prediction in convolution neural network model to obtain deep learning model;Get the labeled image of detection completed, and the labeled image is fuzzy processed, the obtained fuzzy image and labeled image are calculated, the obtained fuzzy degree is input into deep learning model for optimization, to obtain deep learning optimization model;Feature image is input into deep learning optimization model for image processing, and the detection result of image to be detected is generated according to image processing result.The application optimizes deep learning model using fuzzy degree, and uses the constructed deep learning model to carry out image processing to feature image, to generate corresponding detection result.

The invention discloses an acceleration method and device for fast plane diffraction lens imaging based on an FPGA platform, and the method specifically comprises the steps: carrying out the pruning processing of a neural network model, evaluating the loss of the neural network model through a loss function, carrying out the fine adjustment of parameters, carrying out the quantification processing of the neural network model, and carrying out the calculation of the loss of the neural network model. Performing fine tuning and repairing on the quantized neural network model; compiling the neural network model to ensure that the model can run smoothly on the FPGA, and providing a system environment for deployment of the model through hardware design; and finally, deploying the neural network model on an FPGA, processing a blurred image formed by the plane diffraction lens by using the deployed neural network model, and outputting a low-delay high-frame-rate reconstructed image. According to the method, on the premise of considering the imaging quality, the neural network model reasoning acceleration is realized, the imaging delay is reduced, and the method is suitable for miniaturized and low-delay imaging equipment.

The application discloses a mine coal mining face low-quality fuzzy image super-resolution reconstruction method, comprising the following steps: S1, inputting a low-resolution mine image and extracting multi-receptive field feature information of the image; S2, for the multi-receptive field feature, using a multi-dimensional collaborative covariant feature weighted adaptive residual group to carry out high-frequency feature fusion processing, realizing lightweight weighted adaptive collaborative feature fusion; S3, for the fused features, carrying out multi-receptive field feature fusion and up-sampling processing, realizing high-resolution mine image reconstruction, and finally outputting mine image super-resolution reconstruction results; and the reconstruction system is also disclosed, comprising a mine image acquisition module, a pretreatment module, a feature extraction and coding module, a multi-dimensional collaborative covariant modeling module, a super-resolution reconstruction module and a reconstructed image evaluation and feedback module. The application can realize the improvement of image definition, meanwhile, the lightweight of the model structure is maintained, and the dual requirements of calculation resources and processing effect under the mine environment are met.

The invention discloses a traffic cone real-time target detection method, system and device based on a YOLOv8 recognition model, and relates to the technical field of image processing. A blurred image of the target area is acquired, an energy distribution model is calculated according to the blurred image, and adaptive parameters are determined according to the blurred image; calculating according to the energy distribution model and the adaptive parameters to obtain an energy distribution operator, and fusing the energy distribution operator with the blurred image to obtain a clear image; and identifying the clear image through the target model to obtain an identification result, and performing sound and light alarm according to the identification result. And an adaptive deblurring operator is quickly constructed, and the image definition is restored by using one-time convolution, so that the processing speed and the real-time performance of a subsequent target model are improved.

The invention belongs to the technical field of image processing, and particularly relates to an image deblurring method based on event fusion and multi-head attention. The method comprises the following steps: S1, constructing a training set based on a blurred image data set and event voxel grids in one-to-one correspondence with blurred images contained in the blurred image data set; s2, constructing an event fusion multi-head attention network, and training the event fusion multi-head attention network by using the training set to obtain a trained event fusion multi-head attention model; and S3, inputting the to-be-processed blurred image and the event voxel grid corresponding to the to-be-processed blurred image into the event fusion multi-head attention model for processing to obtain a deblurred image. According to the method, the understanding capability of the model on a global structure and context information can be enhanced, and the method is particularly beneficial to recovering large-range fuzzy object contours and textures.

The invention relates to an image processing method and device, equipment and a storage medium, and relates to the technical field of image processing. The method comprises the following steps: performing color adjustment on a to-be-processed image to generate a first image; performing downsampling processing on the first image to obtain a second image; cutting the second image to form a third image; performing fuzzy processing on the third image to obtain a first fuzzy image; and generating a shadow image corresponding to the to-be-processed image based on the first blurred image. According to the invention, the shadow is generated based on the first image corresponding to the shape of the to-be-processed image, so that the generated shadow effect can well adapt to graphs of different shapes, and is more flexible and better fits the graph shape of the image. And the image is subjected to downsampling processing before the fuzzy processing, so that the data processing amount during the fuzzy processing can be greatly reduced, and the processing speed is improved.

The application discloses a kind of intelligent electronic police dynamic scene imaging clear method, first, obtain electronic police dynamic scene video frame and clean, get clear image.Secondly, using the average method of frame insertion obtains fuzzy image and gradient diagram, establishes data set.Then, EADN network is constructed, and edge information and multi-scale information are used to realize fuzzy image clearification, and loss function is involved to train network.Finally, the network is tested to verify the network effect, and the output is the clear image of deblurring.The application enhances the ability of network learning nonlinear feature, generates more accurate attention map, reduces data redundancy, accelerates network convergence, so that the restored image is clearer.