315 results about "Quantization (image processing)" patented technology

The invention belongs to the technical field of image processing, and discloses a miniature ultrasonic image reconstruction method, device and equipment based on deep learning. The method comprises the following steps: constructing a lightweight U-Net model by a processed signal; a mixed loss function and a quantitative perception optimization and compression lightweight U-Net model are adopted, the processed model is deployed on an FPGA, a layered mapping strategy and a double-buffering mechanism are adopted to execute reasoning, and a reconstructed image is output; an up-sampling strategy is dynamically selected through a closed-loop feedback super-resolution module to optimize a reconstructed image, the ultrasonic imaging quality is improved through a lightweight U-Net architecture, hardware collaborative optimization and an end-to-end imaging pipeline, and a depth separable convolution, multi-scale attention gating and mixed loss function optimization algorithm is adopted; the power consumption is reduced through FPGA accelerated reasoning, dynamic voltage frequency adjustment and on-chip memory optimization, the imaging effect is improved through quantitative perception training, and the system is high in resolution, low in power consumption, high in real-time performance and suitable for portable medical equipment.

The invention belongs to the technical field of image processing, particularly relates to an image effect enhancement method and system, and aims to improve the image quality through multi-domain collaborative adaptive processing. The method comprises the following steps: firstly, determining an optimal quantization level through content-aware adaptive quantization, generating a quantization image and a region complexity graph, then decomposing the image into smooth, edge and texture sub-bands through multi-scale domain analysis, constructing a characteristic spectrum, then, applying histogram equalization, anisotropic diffusion and nonlinear sharpening to different sub-bands through adaptive characteristic enhancement, and finally, carrying out adaptive characteristic enhancement on the different sub-bands. And residual error optimization is carried out. And finally, multi-domain collaborative fusion is carried out to construct an enhancement matrix, fusion masks and feature correction are applied to obtain a final enhanced image, and the method is suitable for various images, retains details, enhances different regions in a targeted manner, and has a remarkable effect.

The invention discloses a single-bit image recovery method and device, a terminal and a medium, and relates to the field of image processing. The method comprises the following steps: performing single-bit quantization on a to-be-recovered image by adopting a non-zero quantization threshold, and determining an initial image quantization model; replacing a sign function in the initial image quantization model with a smooth hyperbolic tangent function, performing low-rank matrix decomposition on the to-be-recovered image, and determining a target image quantization model; determining an initial optimization model according to the truncated least square function and the target image quantization model; a semi-quadratic minimization method is adopted to convert the initial optimization model, and a target optimization model is determined; solving the target optimization model by adopting a near-end block coordinate descent algorithm to obtain each low-dimension sub-matrix and a noise matrix; and determining a reconstructed image by using each low-dimensional sub-matrix. The problem that in the prior art, quantization noise cannot be effectively restrained, and therefore image restoration precision is affected is effectively solved.

A neural network-based image processing method and apparatus according to an embodiment of the present invention may: acquire a feature tensor from an input image by using a first neural network including a plurality of neural network layers; acquire a symbol tensor by performing quantization on the acquired feature tensor; and generate a bitstream by performing entropy encoding on the basis of the symbol tensor.

The invention relates to the technical field of video coding and image processing, and discloses a code rate control method and system based on video image segmentation, and the method comprises the steps: carrying out the pixel-level semantic segmentation of a to-be-coded video frame sequence; extracting a foreground region of interest; calculating a corresponding segmentation uncertainty parameter; establishing semantic mutation parameters of the foreground region of interest; generating a semantic perception weight; executing region-level target code rate redistribution and quantization parameter mapping; and executing partition coding control. In the prior art, code rate control mainly depends on motion intensity or pixel complexity, and especially when a foreground target suddenly appears or disappears in a monitoring scene, a technical problem that a key target is blurred or a background code rate is wasted is easily caused. Due to the fact that the uncertainty modeling and semantic mutation sensing mechanism of the semantic segmentation result is introduced, priority coding of the foreground interest area is achieved under the frame-level code rate constraint condition, and the video coding quality and the code rate utilization efficiency are improved.

The invention provides a video coding method and device, electronic equipment and a storage medium, relates to the technical field of image processing, in particular to the field of video coding and the like, and can be applied to application scenes such as video live broadcast and the like. The specific implementation scheme is as follows: acquiring a reference frame which is adjacent to a current frame and has established initial ROI hierarchical division; multiplexing a motion vector generated by the reference frame in a motion compensation time domain filtering process, and determining an ROI position predicted value of the current frame; extracting feature points of the five sense organs of the current frame, and correcting an ROI position predicted value based on a feature point spatial topological relation in the initial ROI; respectively configuring differential quantization parameters for the corrected main face region, the corrected secondary face region and the corrected background region; dynamically allocating a three-layer area code rate according to a real-time network bandwidth; and outputting the coded frame of the current frame and the associated ROI level metadata. According to the scheme, the coding efficiency and quality can be improved.

The invention relates to the technical field of medical image processing, and discloses a fundus focus segmentation method, system and device based on an uncertainty OCT image and a medium, and the method comprises the steps: constructing an encoder-decoder network architecture; an uncertainty guiding module is embedded into the tail end of the encoder, an uncertainty graph is generated, and weighted enhancement is carried out on input features; a cross-layer fusion module is embedded in the jump connection of the encoder and the decoder, features of different levels are processed through parallel maximum value fusion and multiplication fusion branches, and fusion features are generated and then transmitted to the decoder; performing independent uncertainty quantization at the tail end of the decoder to generate an uncertainty graph of each lesion category; and combining the target-level uncertainty excitation loss function and the pixel-level uncertainty loss function, and optimizing network parameters through a total loss function. According to the method, the automatic identification and segmentation performance of a complex focus area can be remarkably improved, and accurate quantitative analysis of ophthalmic diseases is facilitated.

The invention discloses a large-scale deformation compensation method, device, equipment and medium, and relates to the field of image processing, and the method regards a salient point in an image as a leader in a group, is similar to a leading animal in a migration sheep flock, and plays a key role in guiding the whole registration process. In order to estimate the aggregation manifold, large-scale deformation registration is divided into three parts according to an animal migration mode: route decision based on aggregation quantization, route execution with aggregation maintenance and cascade migration with aggregation inheritance. The three parts are integrated into an aggregation cascade migration (CCM) framework, and large-scale deformation of the image is effectively compensated. According to the method, the registration precision and robustness during large-scale deformation compensation can be greatly improved. The method is a universal method, is not limited to specific feature descriptors or deformation models, and can be easily expanded to other image registration tasks.

The invention provides an impact deformation feature extraction method based on high-speed camera shooting and image processing, and the method comprises the steps: collecting the transient deformation data of the edge package of a remote controller housing under impact through high-speed camera shooting, recording the change speed of a strain rate, obtaining the material characteristics, the size of a package structure, and the geometric discontinuous features of an opening and closing part, generating continuous frame sequence images; performing strain field data quantification processing on the third visualization result, extracting strain field characteristic data, and judging influence weights of stress wave reflection on strain fields of different functional areas of the shell in combination with a falling height and an impact angle of the remote controller; and improving edge packaging design parameters according to the deformation buffer effect and discontinuous quantitative indexes at the opening and closing positions, verifying the improvement amplitude of the deformation buffer effect, and obtaining strain field distribution data of the optimized shell.

The invention provides a 2D image generation method and system under a limited sample, and belongs to the field of image processing. The method comprises the following steps: acquiring a limited data set of a target domain, and performing privacy desensitization processing; preprocessing the target domain limited data set subjected to privacy desensitization processing; a non-pre-trained StyleGAN generator is initialized, a native hidden space and a quantized codebook are constructed, and codes in the codebook represent different 2D image features; external knowledge is cited to initialize a codebook, and a generative network of a target domain is trained; the randomly sampled noise is input into the trained generative network, data of a target domain are obtained, generation of the 2D image is completed, and the data of the target domain comprise the 2D image. The method solves the problems that in the fields of medical treatment, judicial and the like, due to the limitation of privacy, copyright or collection cost, the available data size is extremely limited (generally less than thousands of examples), and consequently the model is prone to mode collapse, overfitting and the like.

The invention discloses a post-earthquake building segmentation system and method based on an MDDFAN-Net multi-scale dynamic fusion network, relates to a remote sensing image processing technology, and solves the technical problems that the multi-scale feature extraction capability is insufficient, the feature fusion capability is weak, the boundary quantization effect is poor, and the background has great influence on target segmentation. The system comprises an input module, a segmentation module and an output module, the input module is used for processing a to-be-segmented post-earthquake satellite remote sensing image and then inputting the processed post-earthquake satellite remote sensing image into the trained segmentation module, an MDDFAN-Net network is arranged in the segmentation module, the segmentation module carries out building damage segmentation according to the input data, and the output module outputs the segmented post-earthquake satellite remote sensing image. The output module is used for outputting the segmentation effect for checking; according to the method, global and local features are extracted at the same time through double branches, fragile edge local information of the building is extracted through curved convolution, the global features are extracted through standard convolution branches, and complex and multi-scale dynamic building edges are better captured.

The invention relates to the technical field of point location labeling, in particular to a self-adaptive load balancing point location labeling method and system based on image processing, and the method comprises the following steps: obtaining an image source division region, extracting an edge pixel number texture direction number color channel standard deviation, generating region detail redundant structure distribution information, and monitoring node state data. The method comprises the following steps: constructing a node resource load state mapping set, matching according to a regional detail redundancy ratio and a node state, establishing a task allocation relationship, extracting a boundary gray derivative, screening a boundary stable candidate point location set, calling point location coordinates, sending the point location coordinates to corresponding nodes, executing target verification, writing in an image frame, and generating a labeling result. Quantitative extraction is performed on distribution among image region edge pixels, texture directions and color channels, and task assignment is accurately paired in combination with image structure complexity and node states, so that resource mismatching and processing retardation are avoided, and task distribution accuracy is improved.

The invention discloses an intelligent image processing method and device for an AI chip and a medium, and relates to the technical field of intelligent image processing, and the method comprises the steps: collecting original image data, carrying out K-MEANS filtering and regularization processing, and generating an image data matrix; based on the image data matrix, calculating a similarity score matrix through a query-key-value matrix, extracting an image key region, and generating a key region mask matrix and an attention weight matrix; calculating a quantization precision level parameter according to the attention weight matrix and the key area mask matrix, and performing error compensation in combination with residual learning to obtain a quantization compression feature vector; and based on the quantized compressed feature vector, performing multi-scale similarity matching with a sample library through hierarchical feature fusion, and calculating a confidence score in combination with an attention weight. According to the method, the identification precision is ensured, and the deployment efficiency of the AI chip on the edge equipment is improved.

The invention relates to the technical field of image processing, in particular to a damper valve system adjustment method based on an artificial intelligence technology, a product, equipment and a storage medium. The method comprises the steps that parameter combinations of all valve systems of the shock absorber are collected; performing data quantization on the parameter combination to obtain a feature combination; and inputting the feature combination into a neural network model to obtain a change curve of the damping force along with the piston speed predicted by the neural network. According to the method, the external characteristic functions of the passive hydraulic shock absorber under different valve system structures are predicted through the artificial intelligence algorithm, and the number of tests in the chassis adjustment process is reduced.

The invention discloses a medical image artifact identification and elimination method based on a big data technology, and relates to the technical field of medical image processing, and the method comprises the steps: carrying out the preprocessing of collected image data based on gray normalization, and then carrying out the semantic segmentation and ROI positioning of the image content; and performing lesion segmentation on the positioned image content, selecting lesion features for quantification and fusion, performing model verification, and performing distributed deployment on the verified lesion segmentation model. According to the method, the problems of high missed detection rate of small nodules and high missed diagnosis risk of malignant lesions are solved through the lesion detection model, the false positive rate is reduced, the recall rate of the malignant lesions is improved, and through the lesion segmentation model, the segmentation adaptability to lesions of different sizes is improved, clear segmentation boundaries are obtained, surgical planning is assisted, and boundary positioning errors are reduced.

The invention relates to the technical field of remote sensing image processing, in particular to a remote sensing image processing method and device based on algorithm hardware collaborative optimization, equipment and a storage medium, and the method comprises the steps: carrying out the sea-land segmentation of a to-be-detected remote sensing image through a pre-constructed hybrid heterogeneous calculation model, and obtaining a land mask image; carrying out lightweight processing on the YOLOv5s architecture based on a mixed precision quantification strategy and a dynamic precision matching mechanism to obtain a ship detection lightweight model; regulating and controlling the ship detection lightweight model based on a dynamic parameter self-adaptive regulation and control strategy to obtain a self-adaptive ship detection model; and processing the remote sensing image to be detected by using the land mask image and the adaptive ship detection model to obtain a ship detection result. Through an algorithm hardware collaborative optimization strategy, the characteristics of a heterogeneous hardware platform are fully utilized, remote sensing image ship detection with high efficiency, low power consumption and strong stability is realized, the detection efficiency and precision are improved, the power consumption is reduced, and the system stability is enhanced.

The invention discloses an image linear target high-precision parameter detection method, and relates to the technical field of image recognition, and the method comprises the steps: carrying out the super-resolution reconstruction of a to-be-detected original image based on a perception optimization SRGAN algorithm, and obtaining a super-resolution reconstruction image; using an adaptive Canny edge detection algorithm to obtain contour information in the ROI region of the super-resolution reconstruction image; based on contour information in the ROI region of the super-resolution reconstruction image, performing linear fitting through an RANSAC algorithm for dynamically adjusting the number of iterations and a tolerant error to obtain a preliminary detection result; and constructing a multi-error-source joint model, and performing error compensation and quantization on the preliminary detection result based on the multi-error-source joint model to obtain a final straight line detection result, thereby completing detection. The method effectively improves the accuracy and stability of straight line detection in a complex environment, and is especially suitable for image processing tasks with low contrast, high noise or significant distortion.

The invention relates to the technical field of image processing, in particular to an aeronautical part identification character recognition method and system based on image processing, and the method comprises the steps: obtaining a character image of an aeronautical part, carrying out the stroke fracture reconstruction of a stroke fracture in the character image, so as to generate a reconstructed image, counting the number of connection pixel points newly added in the stroke fracture reconstruction process; performing optical character recognition on the reconstructed image to obtain a preliminary recognition result and a corresponding matching confidence coefficient; determining a stroke integrity index of the character image based on the number of the newly added connection pixel points and the total number of pixel points of the reconstructed image; the matching confidence coefficient of optical character recognition and the stroke integrity index based on reconstruction pixel quantization are combined, double judgment is carried out, it is ensured that the recognition result has the two characteristics of similar forms and reliable sources, and the accuracy and reliability of character recognition under the complex, high-light-reflection and abrasion conditions are improved.

The invention discloses a contour extraction method based on edge detection, and belongs to the technical field of image processing, and the method comprises the following steps: obtaining a plurality of images to be subjected to contour extraction, and real labeled contours corresponding to the images to be subjected to contour extraction; performing noise level quantization and edge ambiguity quantization analysis on each image to be subjected to contour extraction, and calculating to obtain an image complexity level of each image to be subjected to contour extraction; performing complexity level classification on each image to be subjected to contour extraction, and constructing a complexity classification image data set; constructing a contour extraction model based on edge detection; a trained contour extraction model is obtained; and obtaining a new image to be subjected to contour extraction and an image complexity level, inputting the image to be subjected to contour extraction into the trained contour extraction model for contour prediction, and obtaining a predicted image contour of the image as a contour extraction result. The problem that it is difficult to accurately extract the object contour in a complex image scene is solved.

The invention discloses an eye fundus image quality control method and device, a storage medium and electronic equipment, and relates to the technical field of image processing. The eye fundus image quality control method comprises the steps of determining a to-be-processed eye fundus image; based on an effective area of the to-be-processed eye fundus image, determining at least one type of quality quantification data corresponding to the to-be-processed eye fundus image, the effective area being used for representing an unshielded eye fundus structure area; and determining a quality control result of the to-be-processed eye fundus image based on the at least one type of quality quantification data corresponding to the to-be-processed eye fundus image, the quality control result comprising an image quality score and / or a quality problem corresponding to the image quality score. According to the eye fundus image quality control method provided by the embodiment of the invention, the score of the eye fundus image to be processed and the quality problem corresponding to the score can be intuitively known, so that related personnel can know the score and the quality problem, the normalization of the eye fundus image is improved, and standardization of eye fundus image data is facilitated.

The present invention relates to the technical field of sewage treatment, and particularly relates to a sewage state recognition method and device based on image processing and quantitative statistical differences. The method includes obtaining real-time sewage images during the sewage treatment process at intervals; preprocessing each real-time sewage image to obtain each sewage feature image; performing quantitative statistical differences on each sewage feature image to obtain the morphological changes of particulate matters in the sewage; and determining the sewage treatment state according to the morphological changes of the particulate matters. By performing quantitative statistical differences on each sewage feature image to obtain the morphological changes of the particulate matters, and then determining the sewage treatment state according to the morphological changes of the particulate matters, it is possible to better add treatment agents according to the sewage treatment state, reduce the usage amount of the treatment agents, and improve the sewage treatment effect.

The invention provides an image high-definition amplification method and system for generating a planning intention graph. The method belongs to the technical field of image processing and territorial space planning. The method comprises the following steps: acquiring an original planning intention map by using image acquisition equipment, and carrying out image preprocessing on the acquired original planning intention map; adopting a feature extraction network in a deep learning model to perform feature extraction on the pre-processed planning intention graph; professional knowledge in the territorial space planning field is collected and sorted and converted into rules and models which can be processed by a computer, and a planning knowledge base is constructed; constructing a feature optimization model in combination with the feature quantification result and the planning knowledge base; by combining professional knowledge in the territorial space planning field, a planning knowledge base is constructed, and the knowledge is introduced into the training of a deep learning model, so that the generated planning intention map not only is improved in visual quality, but also can better meet the planning requirements, and the reasonability and availability of each element in the image are ensured.

The invention belongs to the technical field of optical image processing, and particularly relates to an optical interference three-dimensional shape measurement method and system based on multi-frame interference image registration. Moving the linear displacement table, and acquiring interference images of samples on the linear displacement table according to a preset sampling interval to obtain an original interference image sequence; calculating an actual sampling interval between the original adjacent frames to obtain a sampling interval error accumulation amount of each frame; performing space coordinate compensation on the horizontal coordinate of each frame to obtain a corrected discrete horizontal coordinate; based on the discrete horizontal coordinates, rearranging the original interference image sequence by adopting a linear interpolation method to obtain a registered interference image sequence; extracting pixel points for recombination to obtain interference signals; and carrying out phase calculation on the interference signal to obtain the three-dimensional surface topography of the sample. According to the method, the interference image sequence is reconstructed by adopting a sampling error cumulant quantization and linear interpolation method, so that sub-pixel-level pattern registration and high-precision and large-view-field three-dimensional shape measurement are realized.

A method for bit rate allocation, an apparatus, an electronic device and a storage medium are disclosed, which relate to the field of artificial intelligence technology, such as computer vision, image processing and video encoder. The method for bit rate allocation includes: obtaining N region of interest (ROI) detection results of an image to be processed, wherein N is a positive integer greater than one, and the N ROI detection results comprise: detection results obtained by performing ROI detection on the image to be processed using N different detection operators respectively; generating a target block quantization parameter offset mask map corresponding to the image to be processed based on the N ROI detection results; performing a bit rate allocation on each image block in the image to be processed based on the target block quantization parameter offset mask map.

The invention discloses an image-driven tick detection comb fused with a deep learning model, which relates to the technical field of tick detection and comprises an initial exposure time configuration module, an image data acquisition module, a feature extraction and fuzzy quantization module, an ambiguity prediction module, an image classification module, a normal image processing module and a fuzzy image exposure adjustment module. The initial exposure time configuration module is used for firstly determining initial exposure time adaptive to the current environment and the detection requirement, and carrying out parameter configuration on the micro camera; according to the method, the image blurring degree is accurately predicted through deep learning and image processing technologies, normal and blurred images are distinguished, adaptive exposure time regulation is adopted for the blurred images, motion blurring is reduced, image definition is ensured, feature extraction accuracy is improved, the misjudgment probability is reduced, ticks are detected and removed in time, and the risk that a host infects pathogens is reduced. The robustness and reliability of the equipment in different environments are enhanced through dynamic exposure adjustment, and the detection precision is remarkably improved.

The invention relates to the technical field of image data processing, in particular to a weld joint pseudo-defect recognition and filtering system based on image processing, which comprises the following steps: acquiring a surface image of a local area of a weld joint through an image acquisition module; marking the welding seam sub-region as a photosensitive dominant sub-region or a photosensitive non-dominant sub-region through a region marking module; screening defect confidence suspected sub-regions through an initial recognition module according to a comparison result of texture feature quantized values of the photosensitive dominant sub-regions and the photosensitive non-dominant sub-regions; analyzing the spatial distribution condition of the defect confidence suspected sub-regions through a feature matching module so as to judge the defect filtering type of the local region of the welding seam; and determining a defect filtering strategy for the local area of the welding seam through an identification filtering module. Furthermore, by using the difference between the pseudo defect and the real defect in the optical response rule, a targeted regional differentiation analysis mechanism is adopted, and the pertinence of the filtering strategy is improved.

The application discloses a kind of based on data center application inference big model discovers network topology fault and false alarm self-healing visualization method and system thereof, including using grid division algorithm carries out grid division and forms DeepSeek inference network topology and chip resource mark;Visual image processing is carried out in DeepSeek inference network topology, complete network topology resource re-planning and visual data annotation, output DeepSeek inference big model inference conclusion;From network topology gateway configuration file, the transmission protocol of application is used as input parameter and carried out Deepseek inference big model inference analysis, carries out power quantization;DeepSeek inference network topology node is associated, and alarm is carried out in transmission protocol through the fluctuation of power level.The application uses microfluidic chip to manage the chip that inference big model carries out operation, chip resource mark is input into inference big model, and then uses image definition position and inference big model embedded AI algorithm matching discovers network topology, improves big model inference conclusion and visual accuracy.

The invention relates to the technical field of image processing, and discloses an AI image super-resolution reconstruction method based on a multi-scale fusion mechanism. The AI image super-resolution reconstruction method based on the multi-scale fusion mechanism comprises the following steps: establishing a binocular image system; establishing a double-end collaborative model architecture; establishing a front-end model optimization mechanism; according to the method, through a scene adaptive feature labeling decision model, three-dimensional position driven quantization and spectral response calibration replace manual presetting, and parallax-pose linkage correction is combined, so that subjective interference is thoroughly eliminated, and the misalign error suppression effect is improved by more than 40%; scale intelligent selection and cross-scale feature interaction are introduced into a dynamic interactive multi-scale attention convolutional neural network (DI-MSCNN), and dynamic definition of convolution kernel parameters is matched, so that feature extraction efficiency is improved by 30%-40%, and texture density differences can be accurately adapted; a detail hierarchical perception GAN (DLP-GAN) generates a strategy through hierarchical discrimination and detail partitioning.

The application discloses a kind of crack segmentation image post-processing optimization and quantization method, device and equipment, it is related to computer vision and digital image processing technical field, this method includes obtaining the initial binary crack mask chart corresponding to the image to be processed, and carries out geometric morphological filtering operation to eliminate non-crack area, obtains preliminary filtered mask chart;Contour detection is carried out to the preliminary filtered mask chart, and the crack-containing region of interest mask chart is identified and cropped;Break point connection processing is carried out to the region of interest mask chart to repair discontinuous crack fragment, generate continuous optimized crack mask chart, to carry out the geometric size parameter calculation of crack.This application can significantly improve the accuracy, integrity of crack segmentation result, and can realize the automation, accurate quantization of key parameters.

The invention provides an image data transmission method and device and an image data processing method and device, and belongs to the technical field of image processing. The image data transmission method comprises the following steps: acquiring to-be-processed image data; encoding the image data twice to obtain top-layer code words of the image data; inputting the top-layer code word into a quantization model to obtain a top-layer quantization code word and a top-layer index; and sending the top-layer index to a receiving end. The image data processing method comprises the following steps: receiving a top index sent by a sending end; inputting the top-layer index into the generative model to obtain a bottom-layer index corresponding to the top-layer index; on the basis of the top-layer index and the bottom-layer index, querying in a quantized codeword knowledge base to obtain a top-layer quantized codeword and a bottom-layer quantized codeword; and splicing the top layer quantization code word and the bottom layer quantization code word, and inputting the spliced quantization code word into a decoder to obtain restored image data. By extracting and transmitting the key information of the image data, the image data transmission efficiency and quality are improved.