674 results about "Source image" patented technology

The invention discloses a multi-source image data fusion-based fine decoration air crack seepage quality problem detection method and device, and solves the technical problem of how to carry out comprehensive, high-precision and intelligent detection on the fine decoration surface air crack seepage quality problem. Comprising the following steps: 1) receiving visible light image data, thermal infrared image data and three-dimensional laser point cloud data of a target refined decoration surface acquired from visible light acquisition equipment, thermal infrared imaging equipment and three-dimensional laser scanning equipment respectively; 2) carrying out feature extraction on the visible light image data, the thermal infrared image data and the three-dimensional laser point cloud data; 3) obtaining point cloud projection coordinates, and then performing association fusion on the first two-dimensional feature and the second two-dimensional feature with corresponding three-dimensional features to generate fusion point cloud data containing multi-source features; and 4) based on the fused point cloud data, carrying out defect classification identification and spatial positioning to obtain a detection result. And comprehensive and high-precision detection of the quality problem of air crack seepage of the finely-decorated surface is realized.

The invention relates to the technical field of rural image inspection and recognition, and discloses a multi-source image collaborative inspection and recognition analysis system and method for a digital rural area, and the method comprises the steps: collecting multi-source image data in real time; obtaining a plurality of characteristic parameters corresponding to each image data item in the image data set, and carrying out space-time registration and multi-scale fusion processing on the plurality of characteristic parameters of each image data item; performing target detection and identification analysis on the plurality of feature parameters in the fusion feature parameter set, and constructing an abnormal point identification model; setting a plurality of abnormal point change thresholds according to the inspection coordinate data set for classification processing to obtain a plurality of abnormal point categories; and setting a corresponding co-processing scheme according to the plurality of abnormal point categories, and setting early warning information corresponding to the change trends of the plurality of abnormal point categories based on the co-processing scheme. According to the invention, the intelligent degree and response efficiency of rural inspection are improved, and the safety and sustainable development of digital rural construction are effectively guaranteed.

The invention discloses a parallel U-Net-based dual-domain collaborative infrared and visible light image fusion method. The method comprises the following steps: firstly, extracting initial features of a source image by using dense connection blocks; then, parallel frequency domain branches and spatial domain branches are constructed, the frequency domain branches are combined with discrete wavelet transform and fast Fourier convolution to decompose and enhance multi-scale global frequency domain features, and the spatial domain branches capture long-distance spatial dependence with linear calculation complexity by using a convolutional layer and Mama based on a selective state space model; dynamic interaction and weighted fusion of double-domain information are realized through an adaptive feature fusion module; and finally, generating a fused image through an image reconstruction module. According to the method, the problems of high calculation overhead and video domain information negligence in the prior art are solved, and infrared heat radiation maintenance and visible light texture enhancement are effectively considered.

The invention discloses an intelligent fire detection and quick response method based on image recognition. The method comprises the following steps: S1, synchronously acquiring and preprocessing multi-source image data; s2, carrying out multi-modal feature fusion and early fire identification; s3, intelligent grading response based on deep reinforcement learning: constructing a fire behavior grade dynamic evaluation network, inputting fused multi-modal features, outputting fire behavior grade probability distribution and a crisis index, judging a fire behavior grade according to a dynamic threshold value, and triggering a predefined grading response strategy; s4, performing cross-scene adaptive transfer learning; and S5, carrying out system self-diagnosis and dynamic optimization. According to the invention, through time-space synchronous acquisition and deep fusion of visible light, infrared and smoke multi-mode data, vision, thermal radiation and smoke characteristics of flames are integrated, and common interferences such as real fire behavior and lamplight, light reflection, a moving heat source, water mist and dust are effectively distinguished by using an improved lightweight CNN, a self-adaptive background temperature model and an environment airflow model.

The invention discloses a construction scene dynamic obstacle avoidance method and system based on multi-source image fusion, and belongs to the technical field of computer vision and image processing, and the method comprises the steps: mapping multi-source image data to a symmetric positive definite matrix manifold space, carrying out the high-precision registration based on Riemannian geometric measurement, achieving the self-adaptive feature fusion through geometric flow optimization, and achieving the dynamic obstacle avoidance of a construction scene. According to the method, spatial topological features of obstacles are extracted through topological data analysis, and probability trajectory prediction is carried out through a variational inference method. Compared with the prior art, the method has the advantages that the obstacle avoidance success rate is increased by 35%-50%, the false alarm rate is reduced by 40%-60%, the similarity of the technical scheme is lower than 20%, and the method has the advantages that the method is suitable for large-scale popularization and application. And the accuracy, the reliability and the self-adaptive capability of dynamic obstacle avoidance in the construction scene are remarkably improved.

The invention discloses a multi-mode AI collaborative industrial visual defect detection system, and relates to the technical field of industrial visual defect detection and machine learning, and the system comprises a multi-source image collection module which is composed of a plurality of cameras and sensors and supports multi-mode image parallel collection; the image preprocessing module optimizes the image quality through enhancement, denoising and semantic segmentation; the multi-mode AI model collaboration module integrates three types of core models to realize feature collaboration fusion; the defect hierarchical classification module is used for constructing a secondary classification system association knowledge base; the defect positioning and quantifying module is used for accurately positioning and quantifying defect parameters; the detection result intelligent output module supports multi-form output and alarm; and the closed-loop optimization module is used for updating the model and the knowledge base based on feedback iteration. According to the invention, through multi-mode data fusion and multi-mode AI cooperation, the defect detection precision and scene adaptability are improved; the method has cross-scene rapid adaptation, real-time response and closed-loop optimization capabilities, and adapts to multi-industry detection requirements.

The invention discloses a high-quality AIGC art two-dimensional code generation method and system based on intelligent area positioning. The method comprises the following steps: scanning a source image uploaded by a user, and dividing the source image into a plurality of candidate areas; for each candidate area, calculating a multi-dimensional image characteristic value, generating an area suitability score, eliminating an unsuitable area through a smoothness threshold elimination rule, and determining the candidate area with the highest area suitability score as a target area; extracting a target area image segment from the source image, and inputting the image segment and to-be-coded two-dimensional code data into a generative image fusion process to generate an artistic two-dimensional code segment; and backfilling the artistic two-dimensional code segment to a corresponding position of the source image to obtain a complete artistic two-dimensional code image. According to the method, the optimal fusion area can be automatically identified and positioned before generation, the generation success rate and the scanning identification rate of the AIGC art two-dimensional code are remarkably improved, the calculation cost is reduced, and the unification of the visual effect and the functionality is ensured.

Systems and methods for image processing (e.g., image extension or image uncropping) using neural networks are described. One or more aspects include obtaining an image (e.g., a source image, a user provided image, etc.) having an initial aspect ratio, and identifying a target aspect ratio (e.g., via user input) that is different from the initial aspect ratio. The image may be positioned in an image frame having the target aspect ratio, where the image frame includes an image region containing the image and one or more extended regions outside the boundaries of the image. An extended image may be generated (e.g., using a generative neural network), where the extended image includes the image in the image region as well as generated image portions in the extended regions and the one or more generated image portions comprise an extension of a scene element depicted in the image.

The invention discloses a remote consultation and live broadcast director method and system for a multi-source medical image, and belongs to the technical field of medical information, and the system comprises a multi-source image collection and adaptation module, a streaming media processing and live broadcast director module, and an intelligent consultation cooperation module. The multi-source image acquisition and adaptation module is used for acquiring original image data streams from different medical image devices in real time through at least two different interface protocols, and performing standardization processing on the original image data streams to generate multiple paths of image data streams; the streaming media processing and live broadcast director module is in communication connection with the multi-source image acquisition and adaptation module and is used for receiving multiple paths of image data streams and carrying out parallel high-speed hardware coding on each path of image data streams; and the intelligent consultation cooperation module is in communication connection with the streaming media processing and live broadcast director module and is used for providing a consultation interaction interface. According to the invention, by integrating the multi-source heterogeneous image data, cross-device and cross-region real-time collaborative consultation is realized.

The invention discloses a generative multi-modal image fusion detection method based on a state space model, and belongs to the technical field of multi-modal image processing and target detection.The method comprises the steps that a network model comprising a generator and a discriminator is adopted, the generator is composed of a feature flow, a fusion flow and a reconstruction flow, extracting low-level features of the source image by using a convolution module, a Mangbar module and a guide type Mangbar module in the feature flow; performing shallow layer and deep layer fusion by using a cross-modal interaction fusion module based on Mangban in the fusion stream, and guiding deep layer feature extraction by using a shallow layer fusion result; and generating a fusion image through up-sampling in the reconstruction stream, and integrating a target detection module to carry out end-to-end detection. Wherein the Mangbar module utilizes the linear complexity characteristic of a state space model to realize global perception modeling of image features. According to the method, the quality and information richness of multi-modal image fusion are effectively improved, and the accuracy and robustness of target detection in a complex scene are enhanced.

The invention relates to a tooth health primary screening method and device based on image recognition, equipment and a medium. The method comprises the following steps: based on acquired tooth multi-source image data, carrying out image preprocessing on the tooth multi-source image data to obtain preprocessed tooth image data; extracting multi-scale tooth features from the preprocessed tooth image data to obtain tooth feature map data; performing multi-feature fusion processing on the tooth feature map data to obtain fused tooth feature data; on the basis of the fused tooth feature data, combining a preset tooth anatomical structure template, performing feature region division and mapping to obtain tooth region feature data; performing health condition analysis on the tooth region feature data to obtain tooth health condition data; and generating a tooth health screening result according to the tooth health condition data. By adopting the method, multi-modal information can be fully mined, and the comprehensiveness and accuracy of preliminary screening are improved.

The invention relates to the technical field of remote sensing and computer vision, and discloses a multi-source image time domain super-division method and system for giant constellations, and the method comprises the steps: obtaining multi-source remote sensing image data which comprises an SAR image, an infrared image and a visible light image; constructing a cross-modal feature space, mapping multi-source remote sensing image data to a unified semantic space, and realizing cross-modal feature alignment; the SAR image or the infrared image is converted into a visible light modal image based on a neural Schrodinger bridge model, and the neural Schrodinger bridge model is decomposed into a plurality of Markov chain sub-problems and solved through adversarial learning and a staged optimization strategy; two-way semantic constraints are introduced into the neural Schrodinger bridge model, and the two-way semantic constraints comprise visual feature matching constraints and text guiding constraints, so that semantic consistency of the generated image in a CLIP feature space is optimized; and the visible light image sequence after time domain super-division is output, and the time resolution is improved.

A method and apparatus for training a monocular depth estimation (MDE) network, including: obtaining a source dataset including a first source image and a first ground truth depth map corresponding to the first source image; obtaining a target dataset comprising a first target image and a second target image; generating an estimated first source depth map corresponding to the first source image using the MDE network; generating an estimated target depth map corresponding to the first target image using the MDE network; generating an estimated relative pose based on the first target image and the second target image using a pose network; and training the MDE network and the pose network by performing mixed supervision training, wherein the performing the mixed supervision training includes performing fully-supervised training based on the estimated first source depth map and the first ground truth depth map, and performing self-supervised training based on the estimated target depth map and the first estimated relative pose

The embodiment of the invention discloses an indoor equipment intelligent control method based on human body induction and human body induction equipment. A specific embodiment of the method comprises the following steps: performing gait feature recognition on a target vibration signal; starting a heat source sensing assembly to collect a heat source image sequence aiming at the target area, and performing heat source identification on the heat source image sequence; starting a heat source positioning assembly to collect a millimeter-wave radar signal for the target area, and performing radar signal feature extraction on the millimeter-wave radar signal; generating object description information according to the gait features, the heat source features and the radar signal features; and controlling the intelligent indoor equipment corresponding to the target linkage information in the target area through the wireless communication component. According to the embodiment of the invention, accurate personnel identification and intention identification are realized, and the effectiveness and timeliness of intelligent indoor equipment control are further ensured.

The invention discloses a YOLO-based avalanche monitoring self-improvement system and method, and belongs to the technical field of geological disaster intelligent monitoring. According to the method, multi-source image data are acquired through an unmanned aerial vehicle and satellite remote sensing, and initial avalanche detection is realized through YOLO; screening high-confidence pseudo labels by using a time sequence consistency verification mechanism, and constructing an enhanced training set in combination with a dynamic time decay weight strategy; when pseudo labels are accumulated or the performance is reduced, the model is automatically triggered to be retrained, and a self-learning closed loop is formed through updating or rollback of a performance evaluation decision model. According to the method, YOLO detection is innovatively and deeply fused with automatic pseudo label distillation, physical parameter inversion and edge end spreading deduction, and full-chain self-improvement of detection-deduction-early warning-optimization is achieved. According to the method, the detection precision is continuously improved, the manual annotation amount is reduced, complex terrains and extreme environments such as strong light reflection and dynamic fuzzy are supported, and the early warning delay is reduced to a millisecond level.

The invention relates to the technical field of photogrammetry, and discloses a photogrammetry space reconstruction method based on multi-source image fusion, which comprises the following steps: extracting an interior orientation element and an initial exterior orientation element, and establishing an image beam intersection constraint intensity model in an object space to calculate a constraint weight; monitoring a weight space change rate to identify a heterogeneous image connection area and extracting a high-weight image bundle as a geometric anchor point; a differential regularization item is applied to geometric anchor points to construct a stable geometric skeleton, a weight damping factor is utilized to restrain a low-weight image beam to execute conformal mapping correction, the image beam constraint strength is represented, the condition number of an adjustment method equation matrix is improved, numerical oscillation caused by difference of different-source sensors is restrained, geometric step of a fusion interface is eliminated, and the conformal mapping correction accuracy is improved. And geometric closing and seamless stitching of a multi-source image under a unified framework are realized.

Techniques are disclosed relating to exchange of images in networked computing applications. In particular, the disclosure relates to exchange of gain curves that are used to represent imaging and / or video in such applications. A gain curve may define a mathematical transformation that relates values from a source image domain to a destination image domain. The image and its associated gain curve(s) may be published to destination devices for consumption. When a destination device consumes the image, the destination device may apply a transform to source image content according to the gain curve(s) published with the image. For example, the destination device may apply a gain curve to an associated image directly, or it may derive another transform from the gain curve and additional information known to the destination device.

The invention relates to the technical field of identification, in particular to a laboratory dangerous behavior intelligent identification method and system. The initial working parameters of the multi-source image acquisition equipment are adjusted through the real-time illumination parameters of the multiple preset monitoring areas, different illumination environments can be adapted in real time, the image acquisition quality is remarkably improved, the image definition and contrast are improved by dynamically fusing the features of different image sources, and the image quality is improved. Through feature enhancement and generation of an optimized behavior feature set, behavior features and laboratory dangerous behavior recognition tasks are effectively associated, so that the model can accurately recognize dangerous behaviors in different situations, and based on association analysis of multi-modal features, the recognition accuracy of the model on various dangerous behaviors is enhanced, and the recognition efficiency of the model is improved. The intelligent level of laboratory safety management is improved, and real-time monitoring and recognition of dangerous behaviors become more efficient and accurate based on the powerful data association capability and the deep learning technology.

A computer system generates descriptions of image-text misalignments. The system includes one or more processors and models for generating textual and visual descriptions of misalignments between a source text string and a source image. The textual description identifies misaligned text segments, while the visual description may include bounding boxes indicating the location of the misalignment. This system automatically generates synthetic image-text misalignment training examples and feedback, which includes generating misalignment captions and visual bounding box labels.

The invention belongs to the field of computer vision, particularly relates to a riverway garbage recognition and positioning method, system and equipment based on deep learning, and aims to solve the problems of low efficiency, high cost, inaccurate positioning, low intelligent degree and the like. The method comprises the following steps: collecting multi-source image data of fixed monitoring and unmanned aerial vehicle inspection, and extracting physical characteristics such as water surface flow velocity; and inputting the preprocessed data into a double-track deep learning model in parallel, fusing a detection result and a feature vector through a space-time correlation and coordinate mapping mechanism, and outputting accurate geographic coordinates, refined classification attributes and continuous drift trajectories of the garbage in combination with a water surface flow velocity field. According to the invention, all-weather, automatic and high-precision monitoring and tracking of the floating garbage in the river channel are realized, and the intelligent level and decision-making efficiency of water environment treatment are remarkably improved.

The invention discloses a cross-modal anti-attack method and system based on mask weight and random cutting, and belongs to the technical field of deep learning. The method comprises the following steps: extracting a saliency mask of a source image; disturbing the source image by using the disturbance variable to generate an adversarial sample; randomly sampling K cutting frames in combination with a significance mask; respectively extracting clipping characteristics of the clipping frame on the adversarial sample and target embedding selected on the target image according to an alignment mode; calculating the cutting loss of the cutting frame according to the cutting feature and the target embedding; calculating the weighted sum of the cutting loss according to the significance mask to obtain the total loss; performing back propagation according to the total loss so as to update the disturbance variable; and re-perturbing the source image by using the perturbation variable based on the updated perturbation variable until the maximum step number or total loss convergence is achieved. The method gives consideration to attack success rate, mobility and disturbance concealment, and can adapt to quality evaluation of different large language models.

Using the superposition principle of linear systems, a series of images of a surface, captured under different illumination conditions (e.g., different patterns or directions of illumination) can be registered to one another based on an additional, composite illumination image that is captured while illuminating the surface under all of the constituent illumination conditions, e.g., with directional illumination from all directions concurrently or with concurrent illumination using a number of different illumination patterns. Additional images may also be obtained under various combinations of illumination conditions, and used with illumination multiplexing techniques to obtain super-resolution or noise-reduced images of the surface. The individual super-resolution images may be used, in turn, to derive super-resolution or noise-reduced three-dimensional reconstructions based on the improved source images.

The invention discloses an image fusion method based on a double-branch feature decoupling auto-encoder, and belongs to the field of computer vision and image processing. According to the method, for the problems of modal pollution and structural distortion in infrared and visible light image fusion, structural semantic information and high-frequency texture details of a source image are extracted respectively by constructing a content feature coding module and a detail feature coding module, and feature decoupling is achieved. And performing deep fusion on the decoupled features by using an adaptive feature weighting mechanism, and reconstructing a fused image with infrared target saliency and visible light detail definition through a shared decoder. According to the method, an end-to-end two-stage training strategy is adopted for optimization, complex prior or post-processing is not needed, the effects of improving the fused image contrast, edge preservation and target recognition performance are remarkable, and the method is suitable for the fields of weak light monitoring, intelligent perception, unmanned system navigation and the like.

Methods and systems are disclosed for using machine learning models to perform pixel-based deformation of fashion items. The methods and systems receive one or more images depicting a first person in a first pose and receive a source image depicting a target fashion item worn on a portion of a body of a second person in a second pose. The methods and systems process, using one or more machine learning models, the one or more images together with the source image to generate a flow field indicating existence and location of each pixel of the one or more images in the source image and modify, based on the flow field, a portion of the one or more images to overlay the target fashion item on the first person including one or more portions of the target fashion item that extend beyond a body of the first person.

The invention provides an image processing method, an image processing device and projection equipment, and the image processing method comprises the steps: obtaining a first image, the first image being shot towards a projection medium after an infrared light supplement lamp is turned on, and the projection medium displays a projection image; judging whether the projection image is shielded by the target object based on the first image; when it is determined that the projection image is shielded by the target object, a shielded first area of the projection image and a second area in the shielded source image are determined, and the second area is an area, corresponding to the first area in the source image, in the projection image. Because a user usually uses the projection equipment to watch a film in a dark environment, after the infrared light supplement lamp is turned on, the imaging quality of the CMOS module on the target object can be improved, and an intelligent eye protection function can be realized based on the CMOS component. And compared with projection equipment only carrying a CMOS module, only a small amount of cost is increased.

The invention provides a semantic segmentation method, system and equipment based on a traditional village multi-source image. The method comprises the following steps: acquiring a multi-source image and preprocessing the multi-source image; performing pixel-level labeling on the multi-source image to obtain a corresponding label graph; calculating the class imbalance weight of each class based on the tag graph, and obtaining a normalized class weight; inputting the label graph into a multi-scale context-aware coding-decoding network for feature extraction and fusion to obtain a prediction probability, constructing weighted multi-classification cross entropy loss based on a normalized category weight and the prediction probability, constructing a composite loss function in combination with boundary sensitive loss, and carrying out iterative updating on the network until convergence, so as to obtain a multi-scale context-aware coding-decoding network; through the setting, a synergistic effect is formed in four levels of data construction, a network structure, a training strategy and result output, so that objective indexes of traditional village street view scale semantic segmentation are improved, and the labor burden in an engineering use scene is remarkably reduced.

The invention discloses an image feature matching method based on cross-direction double-flow interaction and spatial enhancement. The method comprises the following steps: respectively extracting a coarse scale feature map and a fine scale feature map from an input source image and a target image; after the two coarse scale feature maps are spliced, feature extraction is carried out according to multiple scanning directions, and a sequence tensor is generated; performing state space model updating on the first part of sequence in the sequence tensor; splicing and fusing the second part of sequence in the sequence tensor and the reverse feature of the updated sequence to generate a compensation sequence; performing state space model updating on the compensation sequence; forming a multi-direction output sequence by the sequences updated twice; and reconstructing a two-dimensional coarse scale feature map based on the multi-direction output sequence, and executing coarse-to-fine matching optimization in combination with the fine scale feature map to obtain a final matching result. According to the method, multi-direction structure compensation, local geometric consistency enhancement and state updating of multi-source power item driving under the condition of low calculation amount can be effectively realized.

The invention discloses a scraped car broken object double-source image fusion method based on material semantic perception and dynamic compensation, and belongs to the field of image processing. The method comprises the following steps: synchronously acquiring infrared and visible light images, extracting an optical flow field through a dynamic motion compensation module, and realizing pixel-level space-time alignment of double-source images; acquiring a pixel-level material proportion label by utilizing a semantic pre-recognition and multi-material unmixing module; dynamically generating an infrared and visible light weight map based on material characteristics to realize material adaptive weighted fusion; and image enhancement is further carried out through edge constraint and texture constraint dual mechanisms, the detail integrity of the fused image is ensured, artifacts are eliminated through post-optimization processing, and final fused image generation is completed. The fusion result directly supports an automatic sorting system, and the image recognition precision and sorting efficiency of materials such as metal, plastic and rubber are remarkably improved.

The invention discloses an intelligent forest wild animal monitoring method and system based on multi-source data fusion, and relates to the field of wild animal monitoring and ecological protection, and the method constructs an integrated monitoring system by arranging a ground infrared camera network and an unmanned aerial vehicle multi-sensor inspection platform and accessing satellite remote sensing data. And the system analyzes infrared data in real time by using the edge computing node and triggers emergency response of the unmanned aerial vehicle, so that quick tracking of key events is realized. Pixel-level space-time alignment and cross-modal feature fusion are performed on multi-source image data, and a target detection model is input, so that species and behaviors can be accurately identified. Meanwhile, after the multispectral data and the geographical environment data are subjected to gridding alignment, the data are input into a habitat analysis model, and a habitat suitability probability graph is generated. And automatically generating protection decision support information through comprehensive correlation analysis of species distribution and habitat quality. According to the invention, all-weather, full-space and intelligent wild animal monitoring and protection response are realized.

The invention provides a stealth wave-absorbing material defect segmentation method based on multi-source image fusion, and belongs to the technical field of stealth wave-absorbing material defect identification. Visible light and infrared images of a defect area are synchronously acquired through an industrial camera and an infrared thermal imaging device, after image registration and data enhancement, features are extracted by using parallel ResNet18, and the defect of the defect area is identified by using a multi-source image fusion method. And cross-modal feature fusion is realized based on a conditional generative adversarial network, and finally, a DeepLabv3 + network is adopted to complete pixel-level segmentation and classification of defects. According to the invention, texture and temperature information are effectively fused, accurate identification of four defects of broken holes, corrosion, perforation and scratches is realized, and the detection efficiency and accuracy are improved.