112 results about "Image degradation" patented technology

The invention provides a ship target detection method based on nonlinear network enhancement, a medium and equipment, and belongs to the field of artificial intelligence. The method comprises the following steps of: performing type conversion, data division and enhancement on image data in a data set by adopting the ship data set acquired in a real river channel scene; a nonlinear network enhanced YOLO ship detection model is constructed; adopting a multi-task joint loss function to train the YOLO ship detection model; and inputting a port monitoring image and a sea surface aerial image into the trained YOLO ship detection model, outputting a category number, a confidence value and bounding box coordinates of each prediction box, and forming a visual image. According to the method, a nonlinear network enhanced YOLO ship detection model is constructed, and the modeling expression capability of the network on a ship target in a complex scene is effectively improved, so that the robustness and the accuracy of target detection are enhanced, and particularly, the performance is better under the conditions of small target detection and image degradation.

The embodiment of the invention provides a model training and image processing method and device, a storage medium and a program product. In the embodiment of the invention, the target degradation parameter is obtained by optimizing the initial degradation parameter with the target that the texture loss after image degradation processing is smaller than or equal to the set texture loss threshold value, so that the target degradation parameter is obtained according to the target degradation parameter obtained through optimization. The texture loss of the input image of the model obtained by carrying out degradation processing on the target image is smaller than the texture loss threshold compared with the target image, so that a moderately degraded training sample is generated, and the degraded image still keeps some texture information to provide reliable context clues for the model; and texture reasoning can be carried out based on real observation instead of generating some artifacts, so that texture enhancement is effectively realized.

The invention relates to image processing, in particular to a complex environment-oriented power transmission line inspection image processing method, which comprises the following steps of: screening key frames with high geometric information content by quantitatively evaluating inter-frame motion amplitude and feature tracking quality; extracting features of each key frame by using a pre-trained visual model; performing intra-frame feature aggregation on the features of each key frame by using an intra-frame self-attention mechanism to obtain corresponding intra-frame feature representation; performing global feature aggregation on the intra-frame feature representations of all the key frames by using a global self-attention layer to obtain inter-frame feature representations; constructing a multi-task learning network, and carrying out end-to-end collaborative optimization on depth estimation, image defogging and high-level semantic segmentation; inputting the inter-frame feature representation into a multi-task learning network, and obtaining a restored clear image through deep fusion of scene depth information and image degradation priori; according to the method, the defect that the dual requirements of quality improvement and feature retention of the power transmission line inspection image in a complex environment cannot be met can be overcome.

The application discloses a remote sensing image super-resolution reconstruction method based on spatial variable degradation perception, and comprises the following steps: S1, acquiring a high-resolution remote sensing image dataset; S2, constructing an image degradation model; S3, constructing a spatial variable degradation perception blind super-resolution model; S4, training a remote sensing image blind super-resolution model based on spatial variable degradation perception; and S5, reconstructing a high-resolution remote sensing image through the trained blind super-resolution model. The blind super-resolution reconstruction task is decomposed into a degradation process and a reconstruction process, on the one hand, the complex degradation process of an image is simulated, and on the other hand, the degradation conditions of different spatial positions of the image are considered, the degradation kernel closer to the real world is evaluated while the perception degradation range is expanded, serious kernel estimation deviation is avoided, and more accurate remote sensing image super-resolution reconstruction is realized.

The application is suitable for the technical field of radar signal processing, and provides a deep unfolding ISAR super-resolution imaging method based on sparse-neighborhood joint constraint, comprising: firstly, constructing an ISAR image degradation model, constructing an ISAR super-resolution imaging problem based on the ISAR degradation model and the compressed sensing theory, using ADMM to solve the ISAR super-resolution imaging problem, unfolding the process of solving the ISAR super-resolution imaging problem by ADMM into a multi-level neural network, and finally inputting low-resolution ISAR echoes into the trained neural network to obtain an ISAR super-resolution imaging result. The application combines sparse constraint and neighborhood amplitude constraint, deduces a corresponding signal model and an ADMM solving algorithm, effectively improves the reconstruction ability of the algorithm to complex target details and the representation ability of the algorithm to structural information, and can realize effective super-resolution imaging based on narrowband short-aperture echoes.

This invention discloses a detail-learning-based low-light image enhancement method and apparatus for smart home sensing, relating to the field of image processing. The method includes: constructing and training a detail-learning-based low-light image enhancement model to obtain a trained low-light image enhancement model. The detail-learning attention module of the feature enhancement module in the low-light image enhancement model includes a channel detail attention module and a spatial detail attention module connected in sequence. The method involves acquiring low-light images collected by smart home sensing devices and inputting them into the trained low-light image enhancement model. The low-light images first pass through a feature extraction module to obtain initial image features. These initial image features are then input into a feature enhancement module, sequentially passing through N detail enhancement modules to obtain the Nth detail enhancement feature. The Nth detail enhancement feature is then processed by a feature mapping module and residually connected to the low-light image to obtain the corresponding enhanced image. This invention addresses the problem of prominent low-light image degradation.

The application discloses a GAN neural network multiple distortion suppression model based on a coordinate attention mechanism, obtains an original neutron radiographic image from a neutron source; proposes a brand-new neutron radiographic image degradation model; adds Gaussian blur, Gaussian noise, Poisson noise and noise obeying gamma distribution to the clear neutron radiographic image; uses real white spot noise to train the GAN neural network, simulates white spot noise and randomly adds the white spot noise to the clear neutron radiographic image; constructs a multiple distortion suppression model of the GAN neural network based on the coordinate attention mechanism; uses Huber loss to train the constructed GAN neural network based on the coordinate attention mechanism; inputs a real neutron radiographic image containing multiple distortions into the trained multiple distortion suppression model as input, predicts the original image of the real neutron radiographic image containing multiple distortions, and obtains a target result.

The invention discloses a mulberry leaf picking and positioning method and device based on machine vision, and relates to the technical field of machine vision. The method comprises the following steps: acquiring calibrated and synchronized multi-modal image data of a mulberry target area, wherein the data can be a binocular sequence or a depth camera color-depth pair; performing preprocessing including brightness color normalization and image degradation compensation on the data to obtain an enhanced frame; inputting the enhanced frame into a target detection and maturity evaluation joint network, identifying each mulberry leaf instance, and outputting an instance segmentation mask defining the contour of the mulberry leaf instance and a probability vector representing the maturity; based on the instance segmentation mask and depth information, pixels in the mask are converted into a three-dimensional leaf surface point cloud; according to the point cloud, the six-degree-of-freedom pose of each mulberry leaf target is obtained, and at least one candidate picking site is determined; and finally, performing priority ranking on the candidate picking sites according to a preset comprehensive scoring function, and generating a scheduling queue containing timestamps, six-degree-of-freedom poses and priorities.

The invention discloses an underwater target detection method based on multi-modal feature fusion and time sequence context sensing, and aims to solve the problem of low detection precision caused by image degradation, target shielding and limitation of a single sensing mode in an underwater environment. According to the method, an MFTC-Net model is constructed, and the MFTC-Net model comprises a dynamic multi-modal feature fusion module which carries out adaptive alignment and weighted fusion on optical image features and simulated sonar features through deformable convolution and a double-path attention mechanism so as to enhance perception of a fuzzy target; the lightweight time sequence context sensing module is used for carrying out alignment and time-dependent modeling based on motion information on fusion features of continuous frames by using an optical flow estimation network and a convolution gating cycle unit so as to improve the recognition robustness of an occluded target; the context decoupling detection head introduces a large receptive field context module and a detail keeping context module for classification and positioning branches through a task decoupling structure so as to optimize feature representation. According to the method, deep complementation of optical and sonar modes and effective utilization of time sequence information are realized, and the precision and stability of target detection in a complex underwater scene are remarkably improved.

The application provides a degenerated infrared image restoration method and system, which comprises the following steps: applying an image degradation factor corresponding to an image degradation mechanism and adapted to each infrared image in an infrared image sample to construct different types of degraded images based on different image degradation mechanisms; training a super-resolution processing model based on at least one type of degraded image to obtain a degenerated infrared image restoration model; inputting a current infrared image with the image degradation factor into the degenerated infrared image restoration model to perform image restoration processing on the current infrared image by the degenerated infrared image restoration model. The application corresponds to obtain different types of degraded images by multiple modeling methods, and further trains the super-resolution processing model by the combination of different types of degraded images to restore the infrared image acquired under diversified meteorological environments with high quality.

The invention relates to the technical field of image processing, and discloses an image optimization method and device, equipment and a medium, and the method comprises the steps: recognizing the degradation information of a target image, carrying out the image transformation processing of the target image according to the image degradation information and a preset image transformation process, and obtaining an optimized image, performing multi-type visual task detection on the optimized image, generating a quantized image quality index, judging whether the image reaches a quality standard by comparing a detection parameter with a preset threshold value, if not, calculating a reward signal of a reinforcement learning agent model according to a detection result, and updating an image transformation process by using the signal, so as to improve the quality of the image. And then returning to the optimization step to process the image again, if the current strategy reaches the standard, indicating that the current strategy is effective, directly processing the subsequent to-be-processed image by the agent by using the updated strategy, and finally outputting a high-quality target optimized image. According to the invention, the efficiency and precision of image processing are improved.

The invention relates to the technical field of atmosphere image degradation simulation, in particular to an image degradation simulation method for atmosphere turbulence space-time blurring and aerosol multiple scattering, and the method comprises the steps: obtaining an atmosphere turbulence data set, and generating a three-dimensional space-time refractive index fluctuation field; constructing an atmospheric turbulence influence operator, and respectively constructing a real-time turbulence influence operator and an atmospheric turbulence influence distortion operator; constructing a multi-scattering influence operator, obtaining aerosol scattering parameters, and simulating a multi-scattering process through Monte Carlo ray tracing; fusing the atmospheric turbulence influence operator and the multiple scattering influence operator to form a composite atmospheric degradation model; and inputting the clear image into the composite atmospheric degradation model, and outputting a degraded image. According to the method, the Monte Carlo multiple scattering simulation and the dynamic turbulence operator are fused, the complex atmosphere comprehensive degradation effect can be efficiently and vividly reproduced, and continuous degradation processing of the video frame can be carried out due to the fact that the turbulence operator has space-time continuity.

The application discloses a marine organism image processing method and device, electronic equipment and a readable storage medium, and is applied to the technical field of digital image processing. The method comprises the following steps: performing step-by-step down-sampling on a to-be-processed optical image by using a convolution module with different convolution kernels, so as to obtain a plurality of initial images with different scales. Global feature extraction is performed on each initial image under different scales, so as to obtain a plurality of multi-level feature maps; the multi-level feature maps are input into a Transformer encoder, and multi-level detail feature extraction is performed on the multi-level feature maps by adopting a multi-level feature deepening extraction fusion mode. The output features of the Transformer decoder and the multi-level detail features are up-sampled, so as to obtain an enhanced optical image. The application can solve the problem that the underwater image degradation phenomenon is serious or the details are blurred, and effectively improve the image enhancement effect of the marine organism image.

This disclosure provides an image detection method, apparatus, electronic device, storage medium, and program product. Specific implementations include: acquiring an image to be detected; processing the image to be detected using an image detection model; determining image quality quantification information of the image to be detected, image degradation regions in the image to be detected, and textual description information associated with the image degradation regions. The image quality quantification information includes information characterizing whether a predetermined anomaly exists in the image to be detected; the image degradation regions include areas in the image to be detected where predetermined anomalies exist; and the textual description information includes descriptive information about the predetermined anomalies corresponding to the image degradation regions. The image detection method provided by this disclosure improves the accuracy of image detection by processing the acquired image to be detected using an image detection model. By using the image quality quantification information output by the image detection model, the image degradation regions in the image to be detected, and the textual description information associated with the image degradation regions, the image quality quantification information, the anomalies, and the degradation regions of the image to be detected are determined, achieving accurate multi-dimensional image detection.

The invention relates to an intelligent aiming design method and system based on deep learning. According to the method and system, infrared and visible light bimodal video streams are synchronously collected, and weather types and grades are recognized in real time in combination with meteorological sensor data; carrying out image enhancement by adopting a physical constraint self-adaptive denoising network; constructing a target trajectory by using a target detection and tracking algorithm; high-precision prediction of a target motion track is realized based on a double-flow neural network and a time sequence prediction model; fusing the real-time meteorological data and the trajectory model to carry out trajectory compensation calculation; finally, the dynamic aiming point coordinates are calculated; according to the method, the technical problems of serious image degradation and misalignment of target prediction in severe weather are effectively solved, and the first hit rate and task efficiency of an aiming system in a complex environment are remarkably improved.

The invention provides an image enhancement method and device, equipment and a medium, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining a to-be-processed video stream carrying a foreground mask; the foreground mask comprises a first foreground mask of the robot and a second foreground mask of the task object; the second foreground masks of other frames of to-be-processed images except the first frame of to-be-processed image are obtained by taking the second foreground mask of the first frame of to-be-processed image as prior information; and according to the foreground mask, separating an original foreground image and an original background image in each frame of to-be-processed image in the to-be-processed video stream: respectively carrying out image enhancement on the original foreground image and the original background image, and carrying out image degradation operation on the foreground and background fusion image after image enhancement to obtain a target image, therefore, the stability of foreground segmentation is realized, the difference between an enhanced image and an image acquired in actual deployment in visual details is reduced, and the generalization ability of imitation learning of a robot in a real environment is improved.

The invention discloses a multi-weather target detection method based on degraded category perception image restoration, and relates to the field of computer vision and artificial intelligence. The method comprises the following steps: (1) adopting a training normal form of restoration and detection combined learning, simultaneously carrying out restoration and detection decoupling by utilizing features extracted by a single feature encoder, and extracting clear image features from images of various weather types when a model is trained by using various different types of degraded images, therefore, different weather conditions can be adapted; (2) image degradation information is introduced, coding features of the clear image and the degradation image are distinguished, only the degradation image is subjected to image restoration, and mutual influence between the clear image and a restoration module is avoided; and (3) introducing a degradation category predictor to predict the degradation category of the image, and making a detector have the degradation category perception capability through degradation classification loss.

The application provides an industrial CT image coupling artifact multi-scale generative adversarial network correction method, and belongs to the field of digital image processing. The method aims at the problem that image degradation caused by the coupling of various artifacts in industrial CT imaging is difficult to correct effectively, and provides a multi-scale generative adversarial network correction method. The feature pyramid structure is used to help the network capture more comprehensive artifact feature information, and a multi-scale discriminator is used to build a generative adversarial network framework, so that the generated artifact-removed image is clearer and more real. The method has strong coupling artifact correction capability, and can better restore the detail information of the image, thereby ensuring the accuracy of the geometric size and contour edge of the part in the CT image.

The invention discloses a retina eye movement tracking and OCT (Optical Coherence Tomography) scanning compensation method and system, and the method comprises the steps: firstly dividing a fundus SLO image into a plurality of sub-regions, and calculating the stability index weight of each sub-region, thereby adaptively selecting a stable sub-region which is most suitable for being used as a tracking feature; and performing cross-correlation matching on the selected sub-regions to estimate local displacements, and performing weighted fusion on a plurality of local displacements based on a stability index weight to obtain a global eye motion vector of the current frame. According to the global eye motion vector, the drop point position of an OCT scanning light beam is adjusted in real time, and dynamic compensation of B-scan scanning is achieved. According to the method, stable and accurate retina positioning can be kept under the conditions of eye movement, illumination change and local image degradation, the alignment quality and the scanning definition of the OCT image are improved, automatic positioning of the same retina position can be achieved for examination of the same user at different moments, and the accuracy of retina positioning is improved. Regional repeated scanning can be directly carried out to observe focus changes at the same position, and the optical coherence tomography device is suitable for high-speed OCT imaging and clinical fundus examination.

The application provides a vector map construction method and system based on prior map and multi-stage diffusion reasoning. A unified vector encoder is used to encode multiple prior maps to generate prior features. Real-time data obtained is converted into bird's-eye view features, and multi-stage diffusion reasoning is performed to map the bird's-eye view features of the degradation domain to diffusion features of the normal domain. The prior features and the diffusion features are fused to obtain fused features. The fused features are input into a constructed map decoder to generate an online vector map. In the scheme, multiple prior maps are introduced, and a unified vector encoder is used to efficiently encode different prior maps, thereby enhancing the robustness and accuracy of online map construction. Through multi-stage diffusion reasoning technology, image details are gradually optimized, which not only uniformly processes multiple image degradation problems, but also provides high-quality image restoration and enhancement effect in complex environments, thereby providing strong support for an automatic driving perception system.

A non-transitory computer readable storage medium acquires transmission information indicating a transparent state of each pixel of an image to be subjected to image processing and color information of each pixel of the image, acquiring a ratio of a region of a specific color to an entire region of the image by referencing the acquired color information, executing a first image processing based on color information of a first number of pixels in the image when the acquired ratio is less than a first threshold, and executing a second image processing based on color information of a second number of pixels, the second number obtained by deleting at least a portion of the region of the specific color in the image and less than the first number, when the acquired ratio is greater than or equal to the first threshold and is less than a second greater than the first.

The invention discloses a training method and device of an image super-resolution model, equipment, a medium and a product. Inputting the first image and the second image into a first image super-resolution model, wherein the first image super-resolution model comprises a first initial image degradation sub-model, an initial image super-resolution sub-model and a second initial image degradation sub-model; performing degradation processing on the first image by using the first initial image degradation sub-model, and then performing super-resolution processing to obtain a fourth image; performing super-resolution processing on the second image by using the initial image super-resolution sub-model, and then performing degradation processing to obtain a sixth image; under the condition that the loss value does not meet the condition, adjusting parameters of the model, and returning to input the first image and the second image into the first image super-resolution model; and obtaining a second image super-resolution model under the condition that the loss value meets the condition. Even in the face of unknown degradation, accurate high-resolution images can be constructed, and actual application requirements are met.

The invention discloses a decoupling cooperative enhancement method for an underwater image degradation factor, and belongs to the technical field of underwater image enhancement. Aiming at the problem that an underwater imaging environment is complex and changeable, and light propagation is influenced by wavelength selective absorption and suspended particle scattering to cause image degradation, a degradation factor is approximately decomposed into relatively independent scattering degradation and color cast degradation, and a binarization strategy is introduced to compress the storage and calculation amount of data, so that the image degradation is realized. And global information is compensated in back propagation through decoupling learning, and data compression precision loss is recovered. According to the decoupling collaborative optimization method, scattering and color problems are distributed to a physical compensation module and a spectrum compensation module, learning information is isolated through a gating mechanism, and degeneration decoupling is achieved; the loss function is combined with the subtasks and the global loss to compensate the binarization precision loss of the submodules, and finally, the modules are cascaded to process and recover the degraded image.

The present application relates to the field of computer image processing, in particular to a feature enhancement and self-adaptive regression tea leaf picking point positioning method and system. The basic principle of the method is: firstly, the collected tea bud image is preprocessed for clarity; then, a two-stage model architecture of detection first and then positioning is adopted, an improved YOLOv5 network is used in the detection stage to robustly detect multi-scale buds and suppress background interference; then, the target suitable for picking is screened out and cut into a single bud image; in the positioning stage, an improved YOLOv11-Pose network integrated with an adaptive convolution kernel module is used to accurately regress the picking point coordinates; finally, the coordinates are mapped back to the original image and output. The core technical effect of the present application is: through the synergistic optimization of the two-stage process and the targeted improvement of the model components, the problem of inaccurate picking point positioning and poor robustness caused by image degradation, multi-scale targets, complex background and variable bud morphology in the natural environment is effectively solved, providing a high-precision solution for tea leaf automatic picking.

The present application belongs to the technical field of image enhancement in coal mine, and specifically discloses a coal mine image enhancement method, system, computer device and storage medium. Firstly, a glow imaging degradation model is established to obtain a glow image, and a layer separation method is used to remove the glow to obtain a de-glow image; then, on the basis of a low-illumination enhancement SRLLIE method, a target function of the low-illumination enhancement SRLLIE method is improved according to the de-glow image, the improved target function is iteratively solved, an illumination graph with suppressed overexposure and retained structural details and a de-noised reflection graph are obtained; then, the brightness of the illumination graph is adjusted by using an S-shaped gamma correction function to obtain an optimized illumination graph; finally, the optimized illumination graph and the reflection graph are point multiplied according to the Retinex theory to obtain an enhanced image. The present application can not only effectively improve the brightness of the image and suppress over-enhancement, but also restore the image detail information and improve the visual effect of the image.

The invention discloses a traffic road segmentation method in severe weather based on multi-domain feature alignment. According to the method, a segmentation network named as MDFANet is provided for solving the problem that the road segmentation performance is reduced due to image degradation and low visibility under the conditions of rain, snow, fog, low illumination at night and the like. The core of the network comprises three innovative modules: a dynamic adaptive weighted spatial pyramid pooling module (DAWASPP), and the adaptability of the model to complex weather is enhanced by replacing a traditional static fusion strategy with multi-scale cavity convolution and dynamic weight fusion; a layered attention mechanism: suppressing noise and enhancing road structure features in a layered manner through a geometric attention module (GAB) and a channel-space attention module (CSAB); a progressive decoder (PFD) solves the problem of semantic and detail information segmentation through multi-stage feature alignment and fusion, and improves the boundary segmentation precision. Experiments show that the method is obviously superior to a mainstream model in a self-made data set and a public data set, and has higher segmentation accuracy, robustness and real-time performance in severe weather.

The present application relates to the technical field of image processing and generation, and particularly relates to a low-light image degradation simulation method and system based on conditional offset diffusion trajectory, comprising the following steps: S1, obtaining a diffusion model trained on a clear image dataset, and using a deterministic reverse sampling process thereof to define a reference trajectory from noise to a clear image; S2, constructing a conditional offset network, taking a trajectory state, a time step and a degradation condition vector as input, outputting a disturbance vector, and completing training under frozen diffusion model parameters; S3, inputting a clear image, generating an initial state through forward noise addition, iteratively sampling along the reference trajectory and superimposing the disturbance, and finally outputting a low-light image meeting the degradation condition through a decoder. The present application introduces a conditional disturbance mechanism to offset the reference trajectory of the diffusion model, realizes efficient and controllable modeling of the low-light image degradation process, and has the advantages of stable training, accurate degradation, high-quality generated data and the like.

The application discloses a kind of vehicle-mounted image super-resolution reconstruction method, system, device and storage medium, wherein method includes: obtaining the low-resolution vehicle-mounted image to be reconstructed;Using the trained vehicle-mounted image super-resolution reconstruction model to the low-resolution vehicle-mounted image is reconstructed, obtains high-resolution vehicle-mounted image, completes image reconstruction task;Wherein, the vehicle-mounted image super-resolution reconstruction model is trained by using unsupervised degradation enhancement GAN network;In the training process, the feature information of vehicle-mounted image degradation is extracted, and the extracted feature information is fitted into the vehicle-mounted image super-resolution reconstruction model, to improve the reconstruction ability of model.The application fully extracts the feature information of noise, artifact, interference and the like leading to vehicle-mounted image degradation in the training process, and the extracted feature information is fitted into the model, thereby improving the reconstruction performance of the model.The application can be widely applied in the field of image processing technology.

The invention discloses an image denoising model training method and device, an image denoising method and device, equipment and a medium, and relates to the technical field of artificial intelligence, image processing and deep learning. The method comprises the following steps: determining sample image data; training a conditional generative adversarial network by adopting the paired image, the label image and the initial noise to obtain a noise generation model; performing noise synthesis on the paired image and the label image by adopting a noise generation model to obtain synthesized noise; performing joint training on the image recovery network and the convolutional neural network by adopting the label image, the synthetic noise and the time sample to obtain a first image recovery model; using a first image restoration model to restore the non-paired image to obtain a pseudo label image; performing image degradation on the pseudo label image by adopting a noise generation model to obtain a degraded image; and training the first image restoration model according to the degraded image and the non-paired image to obtain a target image restoration model.

The invention provides an image-based text recognition method and device and a medium, and relates to the technical field of image recognizing.The method comprises the steps that a text image is preprocessed to obtain a corresponding normalized RGB tensor; generating an image quality vector based on the normalized RGB tensor and an image quality sensing network; obtaining an initial feature map based on the normalized RGB tensor and the image visual feature extraction network; based on the image quality vector, adopting a FiLM mechanism to modulate the initial feature map so as to obtain an intermediate feature map; performing element-by-element multiplication on the spatial attention mask and the intermediate feature map to obtain a target feature map; performing sequence modeling and decoding on the target feature map to obtain a text recognition result; the negative influence of image degradation on the text recognition performance is effectively relieved, the character error rate is remarkably reduced, and the accuracy of a text recognition result is improved.