392 results about "Multi-image" patented technology

The invention discloses an image denoising method based on double-branch feature fusion, and the method specifically comprises the steps: 1, carrying out the collection of an image, and carrying out the preprocessing of the collected image; 2, a denoising network with double-branch feature fusion is built, the network comprises a generation network and a discrimination network, and authenticity discrimination is carried out on a denoised image generated by the generation network and a real image through the discrimination network; step 3, adversarial training is carried out on the generative network and the discriminant network to obtain a trained dual-branch feature fusion denoising network; and step 4, performing denoising processing on the image by using the trained generative network. The method is remarkably superior to a traditional denoising method in denoising quality, model generalization ability and operation time, different types of noise can be effectively removed, and more image details and structure information are reserved.

The invention belongs to the technical field of image processing, and discloses a multi-image processing method based on multi-modal entity alignment, which comprises the following steps: acquiring a plurality of images; retrieving rich semantic information of the entity from an external knowledge base; selecting the most representative image for each entity in the multi-image scene by using semantic information; encoding the original input of each mode; performing first-layer fusion between a visual mode and a text mode by using cross-divergent attention, then performing second-layer interaction with a structured mode, and finally aligning an entity representation of an image by using contrast loss; and outputting multiple images fusing the text modality and the visual modality. According to the invention, hierarchical interaction fusion is utilized to enhance multi-modal interaction; enhancing entity text representation by integrating the external attribute value and the context information; and the most representative image is selected by using the semantic text, so that the influence of uncorrelated images is minimized.

The invention discloses a surface defect detection method and system based on multi-light source fusion and a related product. The method comprises the following steps: collecting multiple groups of images of a sample to be detected under different light source conditions; marking image defects to obtain a marked data set, and combining the same group of multi-light-source image masks; establishing a multi-light-source fusion segmentation model with multi-image input, multi-feature fusion and single-image output; training the fusion segmentation model based on a multi-light-source data set, and optimizing a training process; the trained model is converted into an acceleration model for production line deployment, and nondestructive, real-time and automatic detection of product surface defects is achieved. According to the method, key information of various defects can be comprehensively captured through the changing light source, and the problems that a traditional industrial quality inspection method is low in efficiency and insufficient in precision are solved.

The invention belongs to the technical field of Internet of Things information security, and particularly relates to a radio frequency fingerprint identification method based on multi-image feature fusion. According to the method, multiple images are extracted from original IQ data, the limitation of single image features is made up, different basic networks are selected according to different types of image data, the basic networks are optimized, SA-D-ResNet18 and SA-ConvNeXt-T networks are provided, a cross attention mechanism module is introduced between the networks, and image feature information is exchanged. In the final stage of feature extraction, a Concat mode is used to splice and fuse the features of a plurality of images. Therefore, the method can still maintain high recognition accuracy in a more complex environment. According to the invention, a common cross attention mechanism in multiple modes is applied to radio frequency image feature fusion, an image-based classification technology is also applied to the field of radio frequency fingerprint identification, and the radio frequency fingerprint identification method based on multi-image feature fusion is researched.

The invention discloses a water level identification method and device based on image fusion and electronic equipment, and relates to the field of image water level identification or other related fields, and the method comprises the steps: employing a target instance segmentation model to carry out the segmentation processing of a target image in which a water area exists, outputting a global contour segmentation map including the water area and an initial mask map for completing preliminary segmentation of the water area; performing depth estimation on the target image through an improved target depth estimation model to generate a depth map; inputting the initial mask graph, the global contour segmentation graph and the depth graph into a multi-image fusion network model, and outputting a water area segmentation mask graph for complementing the sheltered water area part; extracting a water level line for calibrating the water area mask area in the water area segmentation mask graph; and outputting a water level detection report based on a comparison result of the water level line and the warning line. According to the invention, the technical problem of low water level detection precision caused by object shielding and wrong segmentation of a water area in the prior art is solved.

The invention discloses a printing machine equipment operation optimization process and system based on multi-vision analysis, and the method comprises the steps: obtaining multiple groups of image data in a printing process through employing a multi-vision collection scheme in a preset production period; main feature data are obtained through texture and contour feature extraction and a PCA analysis method, the recognition rate of solder paste printing is calculated through the recognition model, the recognition rate is judged, and an acquisition scheme is adjusted. And fusing the multi-image feature set by adopting a DCA feature fusion algorithm to generate a plurality of groups of feature maps, and processing the feature maps through a convolutional neural network to obtain a plurality of groups of fused feature data. And meanwhile, equipment operation parameters are obtained, defect identification is conducted on the fused feature data through the identification model, operation evaluation and optimization are conducted on the real-time equipment operation parameters based on defect information, and the printing quality and the production efficiency are ensured. According to the invention, equipment operation analysis and printing defect evaluation can be effectively carried out from multiple visual features, the intelligence of equipment operation is realized, and the production benefit is improved.

The invention aims to provide a multi-condition diffusion model-based Shengtuo calligraphy rubbing fusion restoration method, which comprises the following steps of: firstly, acquiring a plurality of Shengtuo calligraphy rubbing images from different sources, preprocessing the images, and constructing a standardized training data set; designing a multi-image feature extraction and alignment module to obtain a fused structural feature tensor; constructing a core network structure of the multi-conditional diffusion model, and supporting guidance of a plurality of conditional vectors; jointly inputting the obtained fusion structure feature tensor and a plurality of condition guide items into a diffusion model to generate a fused and repaired potential image; and finally, restoring the potential image obtained by reconstruction into a final output image through a reverse mapping module. According to the method, the problems of blurring, pen lacking, breakage and non-uniform styles during fusion and restoration of the rubbing image in the prior art are solved.

In an example, a method for assembly of a modular cable connector includes, at an end effector of a robotic insertion system, holding a cable insert module for insertion into a module slot of a module retainer. Two or more images are captured of the module retainer. Image processing is performed to identify a segmented image region corresponding to the module retainer and a virtual plane parallel to a face of the module retainer. The end effector is moved to an alignment pose determined based, at least in part, on the segmented image region and the virtual plane. The end effector is moved from the alignment pose toward an insertion pose. Upon insertion of the cable insert module into the module slot of the module retainer, the cable insert module is released from the end effector.

The invention provides a bridge structure health diagnosis agent construction method based on a thinking chain and a domain tool set. According to the method, a language model with a causal reasoning capability is introduced, and a complete task chain process is constructed by combining target detection, segmentation and geometric analysis models for image processing, so that automation and intelligence from image perception to structure diagnosis are realized. The method supports multi-image input and multi-task cooperative processing, has good generalization ability and interpretability, and can effectively improve the efficiency and accuracy of structural health assessment.

A system to treat a patient comprises a user interface that allows a physician to view an image of tissue to be treated in order to develop a treatment plan to resect tissue with a predefined removal profile. The image may comprise a plurality of images, and the planned treatment is shown on the images. The treatment probe may comprise an anchor, and the image shown on the screen may have a reference image marker shown on the screen corresponding to the anchor. The planned tissue removal profile can be displayed and scaled to the image of the target tissue of an organ such as the prostate, and the physician can adjust the treatment profile based on the scaled images to provide a treatment profile in three dimensions. The images shown on the display may comprise segmented images of the patient with treatment plan overlaid on the images.

The invention relates to the field of computer vision and intelligent transportation, and provides a zero-training vehicle re-identification method based on a visual language large model, which comprises the following steps of: generating a dynamic multi-granularity text, analyzing visual features of a vehicle by using the visual language large model to generate structured hierarchical description, and constructing a hierarchical generation framework. Generating basic semantic tags of vehicle types, colors and visual angles and key detail area descriptions guided by local semantics in a layered manner, and dynamically adjusting description levels according to confidence; adaptive feature fusion: realizing adaptive fusion of visual-text features for rough sorting of vehicle similarity; and performing combined comparison reasoning, dividing the TopN images of the vision-text rough sorting list into N / 2 comparison groups, and performing multi-image conjoint analysis by using a visual language large model to realize fine sorting. According to the method, the multi-level fine-grained text description of the vehicle image is generated, and vehicle re-identification in an open scene is realized without training.

The invention discloses a polarization dereflection method based on pseudo 3D convolution, and the method comprises the steps: 1, constructing a real-world polarization dereflection data set of real shooting, and obtaining a synthesized polarization dereflection data set through employing a polarization imaging principle and an optical theory; 2, constructing a deep convolutional neural network using polarization images based on pseudo 3D convolution, and training the deep convolutional neural network by taking a plurality of polarization images with different polarization angles as input to obtain a multi-image de-reflection model; and 3, carrying out de-reflection processing on a polarization image to be mixed by utilizing the trained model to obtain a recovered transmission image. According to the invention, polarization de-reflection based on pseudo 3D convolution can be realized, so that the de-reflection effect in a complex and changeable scene can be effectively improved, and clearer images can be provided for many advanced visual tasks.

The invention belongs to the technical field of super-division imaging, and discloses a multi-image super-division imaging method based on the guidance of a frequency domain low-pass filter, and the method comprises the steps: converting a low-resolution source image from a spatial domain to a frequency domain, and obtaining a spectrogram; performing low-pass filtering processing on the spectrogram, and extracting structure and detail information in the source image to generate an initial weight map; positioning a boundary area, and fusing the plurality of source images; estimating the optical flow based on the global motion multi-resolution, capturing direction and displacement information between adjacent pixels in the fused image, and determining the weight of a boundary region according to the direction and displacement information between the adjacent pixels; and finally, performing weighted fusion on the plurality of source images in combination with the initial weight map and the weight of the boundary region to generate a high-resolution image. According to the invention, an efficient super-resolution imaging method is realized, and the generated high-resolution image provides richer and more accurate visual detail information.

The invention relates to the technical field of chaotic cryptography, image encryption and information security, in particular to a multi-image encryption algorithm based on a hyper-chaotic system and computer-generated holography, which comprises the following steps: S1, generating a pseudo-random sequence for encryption by using the hyper-chaotic system; s2, preprocessing the original image based on computer-generated holography; s3, combining the generated pseudo-random sequence with the preprocessed image to realize image encryption; s4, through decryption operation corresponding to the encryption process, the original image is restored after the correct secret key is input; and S5, the decryption operation comprises inverse transformation and reconstruction operation on the encrypted image. The algorithm comprises an image preprocessing module, a hyperchaotic system module, a compressed sensing coding module, a computer-generated holographic encryption module, a decryption module and a post-processing module, the security of sensitive image data in the transmission and storage process can be ensured, and efficient and reliable guarantee is provided for sensitive image encryption in various fields.

The invention discloses an intelligent identification method for illegal power utilization in a power distribution area considering multi-source measurement data quality, and the method comprises the following steps: carrying out the preliminary abnormal data point screening of voltage time sequence data and power time sequence data of each node, and obtaining an abnormal time sequence data cluster; performing correlation analysis on the abnormal time series data cluster through association rule mining to obtain data quality abnormal data; repairing the marked data with abnormal data quality based on a conditional diffusion model to obtain repaired voltage time sequence data and power time sequence data; performing multi-image fusion on the voltage time sequence diagram and the power time sequence diagram corresponding to the repaired voltage time sequence data and the repaired power time sequence data to obtain a node time sequence fusion image; and inputting the node time sequence fusion image into the trained illegal power consumption identification model to carry out image abnormal point detection to obtain abnormal pixel points, and mapping the abnormal pixel points back to node voltage time sequence data and node power time sequence data to realize source tracing and positioning of illegal power consumption in the power distribution area.

A camera apparatus includes a camera body portion including a prism unit, a lens assembly, and an image sensor unit, a gyro sensor detecting a hand-shake with respect to the camera body portion and outputting gyro sensor information, a driver circuit generating at least one of a prism driving signal for image stabilization of the prism unit and a sensor driving signal for image stabilization of the image sensor unit based on the gyro sensor information, a prism actuator disposed in the camera body portion for driving the prism unit and performing image stabilization on the prism unit in response to the prism driving signal, and a sensor actuator disposed in the camera body portion for driving the image sensor unit and performing image stabilization on the image sensor unit in response to the sensor driving signal.

The invention discloses an unmanned aerial vehicle aerial multi-image splicing algorithm based on opencv, and the algorithm comprises the following steps: S1, obtaining an aerial image data set, employing an OpenCV CUDA parallel processing technology, and carrying out the preprocessing of the image information of the aerial image data set on a GPU; s2, establishing a memory management mechanism, dynamically calculating a memory size request of the image information by combining and utilizing the parallel computing capability of CUDA in the preprocessing process, pre-defining memory blocks with different sizes in the memory management mechanism, and optimizing memory allocation; s3, grouping the preprocessed image information according to the feature information of the image information, and carrying out parallelization processing on image splicing tasks in each group by using CUDA (Compute Unified Device Architecture); by adopting the OpenCV CUDA parallel processing technology, the aerial image is preprocessed on the GPU, the image processing speed and efficiency are remarkably improved, the memory size request of the image information is dynamically calculated by utilizing the parallel computing capability of the CUDA, and the memory allocation is optimized, so that the overall performance is further improved.

Provided is a food preparation assembly (100) comprising a food preparation chamber (102), and a camera assembly (104) for obtaining one or more images inside the food preparation chamber. The camera assembly is controllable to adjust an image characteristic of the one or more images. The food preparation assembly comprises a radiation assembly (106A, 106B, 108A, 108B) controllable to adopt a radiation status in which the radiation assembly emits radiation comprising visible light inside the food preparation chamber. One or more processors (112) is or are configured to: identify the radiation status; and control the camera assembly to adjust the image characteristic using a predetermined relationship for adjusting the image characteristic according to the identified radiation status. Further provided is a method of controlling a camera assembly suitable for obtaining one or more images inside a food preparation chamber of a food preparation assembly, a computer program for implementing the method, and a method of determining a relationship for adjusting an image characteristic of one or more images inside a food preparation chamber of a food preparation assembly.

The invention provides a multi-image encryption method based on a four-dimensional memristor hyperchaotic system and a storage medium, and the method comprises the steps: constructing a new memristor model, introducing a three-dimensional chaotic system, and obtaining a four-dimensional memristor hyperchaotic system; generating hash values of a plurality of plaintext images, obtaining an initial value of a chaotic system through mathematical model conversion, and generating a chaotic sequence after iterative calculation; a cross-plane spiral traversal method is adopted, pixel positions of the image are rearranged to obtain a first scrambled image, and then a chaotic series generated by a chaotic system is utilized to realize secondary image index scrambling to obtain a scrambled image, so that correlation between adjacent pixels is reduced; performing cross-plane bidirectional diffusion modular addition operation on the scrambled image and a pseudo-random matrix generated by the chaos sequence, so that each pixel comprises information of more other pixels as much as possible, and obtaining a final ciphertext image; according to the method, the randomness of the sequence and the image processing efficiency can be improved, and meanwhile higher confidentiality and robustness are achieved.

The invention provides a heat distribution pipeline defect image processing method, relates to the field of image enhancement, and provides a mixed window attention network which utilizes a triangular window technology and a rectangular window technology to synchronously work, solves the problem that the rectangular window technology is distorted at a boundary, allows a model to access more unique screening modes, and improves the efficiency of image enhancement. An attention mechanism can be activated on more image pixels, long-distance and multi-scale features are captured, local features and global features of the image are fully extracted, then the local features and the global features are fused, complementation of the local features and the global features is achieved, and therefore the super-resolution effect of the image is improved.

The invention discloses a visible light infrared multi-source image target fusion detection method and system, and relates to the technical field of image target detection, and the method comprises the steps: carrying out the radiation correction and spatial registration of a multi-source image set, forming an aligned multi-image sequence, carrying out the coordinate calculation and position correlation through employing a PROSAC optimization algorithm, obtaining the position information of a preliminary screening target, and carrying out the recognition of the preliminary screening target; and inputting the position information of the primarily screened target into a dual-channel decoupling model, executing multi-scale context sensing and space-spectrum joint analysis by a visible light channel layer, performing thermal contour extraction and thermal radiation intensity calibration by an infrared channel layer, outputting a target decoupling feature vector, performing manifold mapping on the target decoupling feature vector, and obtaining a multi-source evidence body. According to the invention, through the dual-channel decoupling model and the D-S evidence fusion engine, the discrimination capability of target feature expression is significantly improved, and the accuracy of target detection is enhanced.

Techniques for image generation based on a multi-image set and pose data is described herein. In an example, a computer system generates, by at least using a first machine learning (ML) model, a pose feature based on pose data that indicates a target pose. The computer system generates, by at least using a second ML model, a first appearance feature based on a first image and the pose data. The computer system generates, by at least using the second ML model, a second appearance feature based on a second image and the pose data. The computer system generates a combined appearance feature based on the first appearance feature and the second appearance feature. The computer system generates, by at least using a third ML model, a third image showing an appearance of an element in the target pose based on the pose feature and the combined appearance feature.

Systems and methods for generating augmented reality environments from 2D drawings are provided. The system performs a camera calibration process to determine how a camera transforms images from the real world into a 2D image plane. The system calculates a camera pose and determines an object position and an object orientation relative to a known coordinate system. The system detects and processes a 2D drawing / illustration and generates a 3D model from the 2D drawing / illustration. The system performs a rendering process, wherein the system generates an augmented reality environment which includes the 3D model superimposed on an image of the 2D drawing / illustration. The system can generate the augmented reality environment in real time, allowing the system to provide immediate feedback to the user. The images processed by the system can be from a video, from multiple image photography, etc.

Techniques are described for automated operations to analyze visual data from images acquired in multiple rooms of a building to generate multiple types of building information (e.g., to include a floor plan for the building), such as by simultaneously or otherwise concurrently analyzing groups of three or more images having at least pairwise visual overlap between pairs of those images to determine information that includes global inter-image pose and structural element locations, and for subsequently using the generated building information in one or more further automated manners, with the building information generation further performed in some cases without having or using information from any distance-measuring devices about distances from an image's acquisition location to walls or other objects in the surrounding room.

Methods, non-transitory computer-readable storage media and computer or computer systems are described which include or relate to inputting or receiving information on one or more image characteristics from a graphical user interface, outputting one or more questions or options for additional details of the one or more image characteristics on a graphical user interface by way of a generative artificial intelligence language model performed on one or more processor, inputting or receiving the additional details from the graphical user interface, and outputting one or more images by way of a generative artificial intelligence text-to-image model performed on one or more processor based on the one or more image characteristics and the additional details.

The invention relates to a metal workpiece surface image acquisition method and device based on multiple exposure. The acquisition method comprises the following steps: determining total exposure, and determining single exposure time and corresponding weight according to the total exposure; acquiring a surface image of the metal workpiece at each exposure time, and adjusting the surface image by using the corresponding weight; and combining a plurality of adjusted images to obtain a metal workpiece surface reflection-removed image. According to the method, a strategy of multiple times of irregular exposure is adopted, multiple times of short-time exposure is used for replacing one time of long-time exposure, a strong single light reflection area is avoided, light reflection is reduced, meanwhile, more image details are reserved, and the final image quality is improved.

The invention discloses a method for detecting functional components of multiple medicinal and edible substances based on image analysis, which comprises the following steps: acquiring n multispectral images of a target medicinal and edible substance sample, and preprocessing the acquired multispectral images to obtain a standardized multispectral image set; for an image Ii in the standardized multispectral image set, outputting texture features used for representing surface structure distribution and particle arrangement, extracting spectral features reflecting sample color change and chemical component reaction capability, and outputting structural features used for identifying micro-textures, edge contours and color spot distribution; generating a fused composite feature based on a multi-image channel fusion strategy; and classifying the composite features based on a support vector machine and / or a multi-layer perceptron, and outputting a component detection result of the target medicinal and edible substance sample. Through a multispectral image acquisition and analysis technology, a detection method suitable for samples with homology of medicine and food in different types, sources and processing modes is realized.

The invention provides a polarization integrated imaging detection module applicable to a severe environment, and relates to the technical field of image detection. The system comprises an optical lens, a polarization CMOS sensor, a data transmission link, a control transmission link and an on-chip integrated processing system. Wherein the exposure consistency of the multi-polarization-angle images is ensured by performing acquisition and exposure adjustment on the images at four polarization angles and executing a brightness normalization algorithm, so that the problem of brightness unbalance during multi-image fusion of a traditional exposure algorithm is avoided; besides, through local dynamic range optimization, global gamma correction and a multi-scale fusion method, texture and brightness characteristics at different polarization angles are reserved, and clear and reasonably exposed image generation is realized. The defects in the aspects of atmospheric haze, underwater imaging and strong reflected light processing in the prior art are overcome, high-resolution and high-frame-rate real-time clear imaging is achieved, and reliable image support is provided for various application scenes.

The invention relates to the technical field of super-resolution reconstruction, and discloses a continuous shooting image super-division method based on discrete wavelet transform fusion, which comprises the following steps of: extracting features of multiple frames of low-resolution pictures, and aligning the pictures by using deformable convolution to obtain aligned low-resolution pictures; decomposing the feature map of the low-resolution picture into low-frequency components and high-frequency components in different directions to obtain preliminary fusion features; separating the preliminary fusion feature channel into a semantic branch and a spatial branch, and obtaining a fused feature map by using complementarily enhanced semantic features and spatial features; and carrying out channel expansion on the fused feature map by using a pixel shuffling algorithm, and carrying out spatial rearrangement on the features after channel expansion to realize efficient and rapid up-sampling. According to the method, the problems of difficulty in alignment of multi-target complex movement and artifacts, detail distortion, blurring and the like of super-resolution of different-scale targets in a traditional method are solved, and the multi-image super-resolution imaging effect is enhanced.

The invention discloses a multi-image-source common exit pupil overlapping display system and method, and belongs to the technical field of optical imaging and display, and the system comprises a multi-image-source module which comprises a plurality of independent image generation sources and is used for providing different types of image information; the image wavefront guiding optical module is arranged behind each path of image source and is used for carrying out collimation, shaping, regulation and control and path guiding on light waves emitted by the multi-image source module and realizing consistent superposition of a plurality of heterogeneous image sources in spatial positions, angles and wavefront parameters at a common exit pupil superposition display surface P; the common exit pupil superposition display surface P is used for carrying out space and time fusion on wavefronts of all image sources to generate a composite image; and the detector module is arranged behind the common exit pupil superposition display surface and is used for receiving the fused composite image wavefront and converting the fused composite image wavefront into image data to be output. According to the system and the method, high-uniformity, high-resolution, high-dynamic-range and multispectral large-depth-of-field fusion imaging can be realized, so that higher requirements of modern simulators, visual systems and multidimensional sensing equipment on imaging quality and system performance are met.