5725 results about "Reference map" patented technology

The invention relates to the technical field of mode recognition, in particular to a transformer substation equipment temperature measurement method and system based on visible light guidance. Firstly, a visible light image, an infrared thermogram and vibration spectrum data are synchronously collected, a dynamic compensation coefficient matrix is constructed through illumination intensity analysis, the dynamic compensation coefficient matrix is applied to the infrared thermogram, and illumination interference compensation is achieved. And then, inputting the corrected infrared thermogram and visible light image into a pre-trained heterogeneous atlas matching model, and identifying the category and three-dimensional coordinates of the target equipment. And based on the equipment category information, calling a standard temperature field template in the digital twin model library, establishing a heat conduction path model in combination with the vibration spectrum data, and generating a temperature field prediction reference map of the equipment. And finally, through pixel-by-pixel deviation analysis, inputting the temperature deviation value, the image features and the vibration data into the probability graphic model, and outputting equipment fault information. The method can realize high-precision temperature measurement and fault diagnosis.

Systems and methods are described for video coding using generalized bi-prediction. In an exemplary embodiment, to code a current block of a video in a bitstream, a first reference block is selected from a first reference picture and a second reference block is selected from a second reference picture. Each reference block is associated with a weight, where the weight may be an arbitrary weight ranging, e.g., between 0 and 1. The current block is predicted using a weighted sum of the reference blocks. The weights may be selected from among a plurality of candidate weights. Candidate weights may be signaled in the bitstream or may be derived implicitly based on a template. Candidate weights may be pruned to avoid out-of-range or substantially duplicate candidate weights. Generalized bi-prediction may additionally be used in frame rate up conversion.

The invention relates to the field of image comparison, and discloses a multi-layer PCB alignment deviation detection system and method based on image comparison, and the method comprises the steps: obtaining high-resolution image data before and after lamination of a multi-layer PCB, carrying out the unified registration preprocessing of an original image through the combination of a multi-mode image fusion algorithm and a geometric distortion correction technology, and obtaining a multi-layer PCB alignment deviation detection result; constructing a standardized image registration input set; key alignment feature extraction is carried out on the image registration input set, and a multilayer structure graph model is constructed based on a graph neural network; in combination with the alignment reference map, dynamically adjusting an image comparison window and a search region by adopting a region attention mechanism and a local adaptive matching algorithm, and constructing an alignment deviation mapping map; constructing a deviation evolution model by using a time sequence behavior recognition network according to the constructed alignment deviation mapping graph and historical process deviation data; and performing comprehensive evaluation on the image registration input set based on the deviation evolution model and the generated early warning information. The method has the advantage of improving the accurate detection level.

The invention relates to the technical field of industrial machine vision detection and image processing, in particular to an automobile part production mold surface smoothness detection system based on image enhancement, which comprises a data acquisition module used for acquiring an original grayscale image of the surface of an automobile part mold; performing low-pass filtering processing on the original grayscale image to eliminate imaging thermal noise; the manifold reconstruction module is used for constructing a structure tensor field; reversely deducing a pseudo-curvature field of the mold surface; the adaptive enhancement module is used for generating a corrected image; constructing a texture orthotropic diffusion model; generating a texture reconstruction reference image; the surface metering module is used for calculating the difference between the corrected image and the texture reconstruction reference image and generating a defect saliency image; calculating the surface roughness value of the mold surface; according to the method, the problem that design textures and abnormal scratches are difficult to distinguish in the prior art is effectively solved, and the technical bottleneck that micro defects are easily missed in a complex geometric structure in traditional visual detection is overcome.

The invention relates to a multi-stage filtering road spilled object detection method based on dynamic difference analysis, which is suitable for automatic identification of unstructured foreign matters in video monitoring. The method comprises the following steps: firstly, extracting a reference image road mask, eliminating vehicle and pedestrian interference by using YOLOv8 detection, and extracting a motion candidate area through a frame difference method and background modeling; and then context expansion and super-resolution reconstruction are carried out on the candidate region, the candidate region is converted into an HSV space, multi-dimensional features such as color similarity, structural similarity and shadow determination are synthesized for screening, false detection is further removed in combination with inter-frame time sequence consistency, and finally a stable detection result is output. The method provided by the invention has the advantages of strong anti-interference capability, high adaptability, high detection precision and the like, and is suitable for the intelligent recognition task of the expressway thrown objects in a complex environment.

Wireless sensing using classifier probing and refinement is described. In one example, a described method comprises: computing output analytics by a classifier based on input data constructed based on raw measurement data; mapping each output analytics to a respective mapped outcome; identifying at least one reference input data each associated with a reference output analytics and a reference mapped outcome; each reference input data being one of the input data for the classifier for which a respective reference outcome is available and is different from the reference mapped outcome; and for each reference input data for the classifier: computing a respective plurality of perturbed output analytics by the classifier, generating selected perturbed input data based on the plurality of perturbed output analytics; and re-training the classifier based on the selected perturbed input data and the associated reference outcome.

The invention discloses a transformer substation positioning and navigation method and system integrating positioning and visual auxiliary positioning. The method comprises the following steps: establishing a background electromagnetic field distribution baseline graph associated with a three-dimensional reference map; electromagnetic field data of the transformer substation in an emergency response mode is collected through the unmanned aerial vehicle, and the abnormal field intensity of a space electromagnetic field and a fault source pointing vector pointing to a fault source are calculated; constructing a risk adaptive elliptical forbidden zone; calculating a passage cost index Cto of the unmanned aerial vehicle passing through each three-dimensional grid in the three-dimensional reference map by using the fault source pointing vector and the risk adaptive elliptical forbidden zone; and searching a path which is from the current position of the unmanned aerial vehicle to a preset target point, bypasses the risk adaptive elliptical forbidden zone, faces the fault source and has the lowest total passing cost index Cto by adopting a preset path search algorithm. According to the method, the risk self-adaptive elliptical forbidden zone is constructed and is used as the three-dimensional grid forbidden assignment processing, so that the dynamic global path avoidance of the sudden dangerous zone is realized.

The invention relates to the technical field of printed matter quality detection and computer vision and the technical field of artificial intelligence systems in the production field, in particular to a color box printed matter printed pattern visual detection algorithm. The invention discloses a visual detection algorithm for a printing pattern of a color box printed matter. The visual detection algorithm comprises the following steps of image acquisition, preprocessing, feature extraction, reference comparison, defect evaluation and defect classification. According to the scheme, the accuracy and the real-time performance of complex printing pattern defect detection are improved. A light-weight multi-scale attention convolutional neural network and a self-learning mechanism are combined, defects in various printing patterns such as block colors, fine characters and graph gradient are effectively detected, and human intervention and misjudgment are reduced. A defect saliency map is generated through color normalization, geometric correction preprocessing and reference map comparison, an online misjudgment cache and incremental learning strategy is introduced, self-adaptive optimization of a model along with time is kept, and real-time high-precision quality monitoring of an offset printing color box production line is achieved.

The invention discloses a power transmission line small sample abnormal data generation method and system based on a diffusion model and feature migration, and belongs to the technical field of abnormal detection. The problem that stable and high-fidelity abnormal data can be generated only when the demand quantity of abnormal samples is much larger than that of actually obtained abnormal samples is solved. The method comprises the steps that an abnormal defect library and an equipment main body library are constructed based on defect reference images in a data set and corresponding labels, and the data set at least comprises one defect reference image of each type; training an abnormal defect identification baseline model based on the data set and verifying to obtain a low-quality image; comparing the abnormal defect library and the equipment main body library with low-quality images by using an image comparison technology to obtain an abnormal defect reference image and a base image to be generated; generating a to-be-migrated region mask based on the to-be-generated base map; and based on the to-be-generated base map and the to-be-migrated region mask, using a diffusion model to migrate the abnormal region features of the low-quality image to the to-be-generated base map for anomaly generation. The method is used for generating the small sample abnormal data of the power transmission line.

Example devices, methods, and computer-readable media are disclosed. An example device includes one or more processors configured to determine a reference picture list for a current block of a current picture of the video data. The one or more processors are configured to determine that a reference picture of a collocated block is listed in the reference picture list for the current block, the collocated block being collocated with the current block in a collocated picture. The one or more processors are configured to select, in a condition where the reference picture of the collocated block is listed in the reference picture list for the current block, a reference picture for the current block to be the reference picture of the collocated block. The one or more processors are configured to and encode or decode the current block based on the reference picture for the current block.

An example device includes one or more processors configured to determine a first search area in a first reference picture for a current block of the video data. The one or more processors are configured to determine a first initial reference block in the first search area. The one or more processors are configured to apply a local illumination compensation model to the first search area to generate a refined first search area. The one or more processors are configured to apply template matching to the refined first search area to determine a first candidate motion vector having a lowest template matching cost for the refined first search area. The one or more processors are configured to decode the current block based on the first candidate motion vector.

The invention relates to the technical field of image processing, and provides a personalized image generation method and device, electronic equipment and a storage medium, and the method comprises the steps: associating image data, text description and instance segmentation masks through a cross-modal automatic labeling system, and generating an image-text-instance segmentation mask associated data set; constructing a corresponding mixed training data set based on the associated data set, and performing data enhancement on the mixed training data set; performing model training on a preset diffusion model based on the mixed training data set after data enhancement; and generating a target image through the trained diffusion model by adopting a hybrid guidance strategy of time step size control. According to the method, the problems that in the prior art, the generation effect is not ideal, user operation is complex, and the generated image and the reference image are inconsistent in feature can be solved, and high-quality, flexible and accurate personalized image generation is achieved.

The embodiment of the invention provides an information processing method and device, a storage medium and a program product. In the embodiment of the invention, the picture generation instruction can be directly initiated for the first picture displayed on the first page of the target application, the picture backstepping model can be automatically called based on the picture generation instruction to obtain the picture generation cue word and the picture generation parameter information of the first picture in a backstepping manner, and the first page is automatically skipped to the second page; in addition, a plurality of corresponding picture generation information items on the second page can be automatically filled with the picture generation prompt word, the picture generation parameter information and the first picture serving as the reference picture, the picture generation process is simplified in the automatic process, and the picture generation efficiency is improved. Furthermore, on the basis of the information in the information item, the image generation model is called to generate at least one second image similar to the first image in style, so that the image generation efficiency and quality are further improved, and the style similarity between the second image and the first image is improved.

The embodiment of the invention provides a special effect template generation method and device, electronic equipment and a storage medium, special effect generation data is obtained by responding to a first operation, the special effect generation data comprises configuration parameters representing a special effect style of a self-defined special effect, and the configuration parameters comprise a first prompt word and / or a first reference graph; displaying an effect picture of the user-defined special effect according to the special effect generation data and the user personalized image; in response to the second operation, the effect picture and the construction data corresponding to the effect picture are packaged, a special effect template of the user-defined special effect is generated, the effect picture is used for displaying the visual effect of the user-defined special effect based on the user personalized image on the resource platform, and after the special effect template is called, a corresponding special effect image is generated through the construction data. The flexibility and convenience of creating the self-defined special effect by the user and the propagation attribute of the special effect template are improved.

The invention relates to a three-dimensional model generation method based on new visual angle texture correction, and the method comprises the steps: generating a basic three-dimensional patch model through a cross-domain diffusion model, and initializing three-dimensional Gaussian splash representation on the three-dimensional patch model; sampling the three-dimensional Gaussian splash representation from different new view angles, and obtaining a rendered image and a correction mask of the three-dimensional Gaussian splash representation under the new view angles; the rendering image of the three-dimensional Gaussian splash representation under the new visual angle is fused into reference image information by using a diffusion model for correction, and a corrected image is obtained; projecting the reference image to the new view angle by using the three-dimensional patch model to obtain a reference image projection; and jointly optimizing three-dimensional Gaussian splash representation by utilizing the corrected image and the reference image projection and combining loss functions of different visual angles and a self-adaptive density control strategy, and obtaining a three-dimensional model. Compared with the prior art, the three-dimensional model with rich texture details generated by the scheme of the invention has excellent visual effect and keeps geometric rationality of the model at the same time.

The invention provides a three-dimensional reconstruction method based on guided filtering and Mama geometric feature fusion in the technical field of three-dimensional reconstruction, and the method comprises the steps: S1, collecting a multi-view RGB image, carrying out the calibration of a camera, obtaining a projection matrix and an internal reference matrix, and obtaining a reference image; s2, extracting image features from each RGB image; s3, sampling on the reference image to obtain a depth hypothesis plane, performing homography transformation on each image feature, and aggregating with the depth hypothesis plane to obtain a cost body; s4, sequentially performing filtering, regularization and exponential normalization on the cost body to obtain a probability body, and constructing a depth map and a confidence map based on the probability body; s5, performing frequency domain filtering and up-sampling on the depth map and the confidence map, and then performing feature fusion to obtain a visual angle depth; and S6, executing three-dimensional reconstruction operation based on the visual angle depth, the projection matrix and the internal reference matrix. The method has the advantage that the precision and robustness of three-dimensional reconstruction are greatly improved.

In an example embodiment, a method carried out by an uncrewed aerial vehicle (UAV) may involve receiving a reference map of a cluster of charging pads from a server. The cluster may include a layout of charging pads and fiducial markers distributed across the layout, the reference map representing the layout and fiducial markers. The UAV may fly to the cluster and acquire an image of charging pads and observed fiducial markers near the charging pads. The image may capture an observed constellation of fiducial markers at apparent positions and orientations relative to the charging pads. A reference constellation of fiducial markers at reference positions and orientations relative to reference charging pads may be identified in the reference map. Identities of the reference charging pads and a match of the reference constellation to the observed constellation may be used to disambiguate a particular charging pad from among the charging pads.

The present invention discloses an optical remote sensing image change detection method based on deep learning, belonging to the field of remote sensing image change detection. First, optical remote sensing image data is acquired and preprocessed; then, a two-step progressive structure is adopted to combine the Resnet18 network and the Vision Transformer model to extract the features of the images before and after the change; next, the extracted features are input into the channel attention module to obtain features with different attention sizes for different channels; then, the features before and after the change are sent into the difference information extraction module to obtain the difference feature map of the two remote sensing images; finally, the difference feature map is sent into the classifier to obtain the final change detection binary image, and it is compared with the change reference map. The present invention effectively optimizes and improves the problems existing in the existing change detection methods, such as low detection accuracy, large probability of false detection and missed detection, and poor universality of the change detection algorithm on different data sets.

The invention belongs to the technical field of circuit board quality detection, and provides an AOI-based circuit board visual quality inspection system and method, and the system comprises a feature extraction module which is used for carrying out the feature extraction of a set circuit board element reference diagram and a to-be-detected circuit board element input diagram based on a feature extraction model, and obtaining a feature extraction result; obtaining first feature data corresponding to the circuit board element reference diagram and second feature data corresponding to the circuit board element input diagram to be detected; and the feature comparison module is used for performing defect category identification on the first feature data and the second feature data based on a defect identification model to obtain a defect category identification result. According to the method, through accurate feature extraction and efficient defect identification, the detection first pass yield of the AOI equipment can be improved, and the cost of manual reinspection is effectively reduced.

The invention discloses an interactive footwear design and platemaking optimization system based on AI, which relates to the technical field of digital interaction and comprises an auxiliary recommendation module, an AI interactive design module, a platemaking simulation module and a parameter optimization module. The auxiliary recommendation module is used for performing feature extraction processing on the shoe design related data by using a machine learning method to obtain recommended shoe design parameters; the AI interactive design module is used for generating a design reference drawing of the shoes, performing visual display and contour extraction on the design reference drawing, generating designed shoes with corresponding shoe sizes according to the three-dimensional shoe style data and the shoe size data, and performing local parameter modification according to the try-on feedback data; the plate-making simulation module is used for splitting the designed shoes, determining sewing procedures, outputting 3D printing parameters and driving the corresponding production line models to execute simulation production tasks to perform plate-making simulation on the obtained simulation shoe models, the design cost can be reduced, and the design quality and efficiency of the shoe products can be improved.

A mechanism for processing video data is disclosed. An indication is signaled. The indication indicates whether a picture following a dependent random access point (DRAP) picture in decoding order and preceding the DRAP picture in output order is permitted to refer to a reference picture positioned prior to the DRAP picture in decoding order for inter prediction. A conversion is performed between a visual media data and a bitstream based on the indication.

The invention discloses a monocular endoscope real-time depth estimation method and system based on self-supervised learning, and the method comprises the steps: firstly inputting a reference image and a target image, carrying out the preprocessing of the target image, and generating an image block sequence; and inputting the image block sequence into the depth map generation network, and outputting a depth map. And extracting feature graphs of the reference graph and the target graph by using an encoder, and respectively inputting the feature graphs into the appearance flow graph generation network and the optical flow graph generation network to obtain an appearance flow graph and an optical flow graph. Inputting the feature map into a body motion estimation network to obtain relative motion transformation of the endoscope, and inputting the relative motion transformation into a camera internal reference decoder to obtain a camera internal reference matrix; and reconstructing a target map based on the depth map, the relative motion transformation and the internal reference matrix, carrying out brightness calibration in combination with the reference map and the appearance flow map, and supervising and optimizing the depth map generation network through the optical flow map and the visibility mask to realize the real-time depth estimation of the monocular endoscope. According to the invention, depth information can be rapidly output in a medical scene with a high real-time requirement, and accurate three-dimensional visual assistance is provided.

The invention relates to the technical field of defect detection, and particularly discloses a high-precision defect detection system based on image difference and threshold processing, which comprises a registration image acquisition module, a fixed mask conversion module, an image difference detection module, a correction difference graph generation module and a defect set output module, the method comprises the following steps: firstly, completing affine registration by shape features, changing a fixed mask into a dynamic mask, and performing edge robustness in the mask to obtain a preprocessed image; local movement difference is carried out on the preprocessed image and the multiple reference images, fusion of a reference domain and a time domain is carried out to generate a difference response image, and candidate areas are extracted in combination with a dynamic mask; estimating and correcting a displacement field in the candidate range, synchronously mapping to obtain a correction difference graph and a correction mask graph, and determining a peripheral rejection area; according to the method, false alarm and missing detection are effectively reduced, and the method is suitable for online high-precision detection.

1. Name of the Design Product: Graphic User Interface for Enabling Aiming Mark of Electronic Device. 2. Use of the Design Product: An electronic device. 3. Design Key Points of the Design Product: Lies in the graphic user interface. 4. Picture or Photo Most Indicating the Design Key Points: The front view of Design 1. 5. Design 1 is designated as the basic design. 6. Use of the Graphic User Interface: For enabling an aiming mark for images collected in real time. 7. Human-Machine Interaction Method of the Graphic User Interface: The front views of each design are graphic acquisition interfaces. Different aiming marks can be enabled by clicking on the second aiming mark icon from top to bottom in the upper left corner of the interface. For example, click on the aiming mark icon in the upper left corner of the reference view of the usage state of Design 1 interface and select 1 to switch to the reference view of the usage state of Design 1 interface in Figure 2. The interaction methods of the reference views of the usage states of the interfaces of Design 2, 3, and 4 are the same as that of Design 1.

The present invention discloses a multi-source heterogeneous large-scene remote sensing image matching method for global optimal solution, including: S1, preprocessing the input image information; S2, extracting the reference image point group features by constructing a sub-pixel point group feature detection algorithm; S3, constructing a multi-mode matching algorithm based on wavelet analysis for sub-pixel matching; S4, constructing a point group global registration algorithm for forward and inverse projection, hierarchical elimination, and obtaining the global optimal solution; S5, judging whether the sub-pixel requirement is met. If it is judged that the requirement is not met, return to step S4, otherwise output the result. According to the present invention, a high-precision sub-pixel is obtained, the consistency of image intensity and phase is ensured, an accurate matching result is obtained, and an algorithm system with good compatibility is provided.

The present invention provides a method, a system and a storage medium for ensuring the arrival consistency of multiple robots. The method steps include: Step S1: Establish a calibration site and calculate the transformation matrix of the pose of each robot under the reference map; Step S2: Establish a working site and calculate the pose of the robot and the working site and their transformation matrices in the reference map coordinate system map respectively; Step S3: When the robot navigates to the working site, each robot generates a new end point for navigation and commands the robot to go. Thereby, multiple robots can complete the precise docking between the robot docking mechanism and the site docking mechanism under the condition of using the same map.

The invention discloses a story-driven role and scene image generation method. The method comprises the steps that a natural language story text input by a user is received and preprocessed; through predefined role description structure constraints, enabling the language understanding and generation model to output a structured role description information structure under template constraints; generating a role image according to the structured role description information text, and extracting image features for consistency control; establishing a mapping table of structured role description information and image feature representation, and realizing the consistency of the appearance of roles in multiple scenes; automatically disassembling the complete story text into a plurality of scene nodes, and generating structured scene description information for each scene; and generating a complete story picture in combination with the scene description information and the role reference diagram. The invention provides a story-driven role and scene image generation method, which is used for automatically generating a story text to a role image and a scene image through semantic understanding, information description structured generation and image consistency management.

The invention is suitable for the technical field of graph data, and relates to a graph association rule mining method and device, equipment and a medium. The method comprises the following steps: acquiring graph data of a service system, and determining a target attribute corresponding to a target task and a target node tag corresponding to the target attribute; obtaining a reference graph mode according to the target attribute and the target node label; determining target sub-graph data matched with the reference graph mode from the graph data; obtaining graph features of the target sub-graph data, and performing decision tree training according to the graph features by taking the target attributes as training tags to obtain a reference decision tree; and performing association rule extraction on the reference decision tree to obtain a target association rule. The mining efficiency of the graph association rule can be improved.

Methods and systems are described for end-to-end video coding and decoding using deep learning. Proposed fusion schemes for reference pictures and motion vector features allow single-prediction models for deep-learning video coding (DLVC) to be extended to support multi-reference prediction coding as well. Example high-level syntax to support a variety of deep learning models for bi-prediction is described as well.

A method for video processing includes receiving an input image via an initial feature extraction portion, where the initial feature portion includes a result of RPR up-sampling, a low resolution (LR) reconstructed frame, and an LR predicted frame; processing the input image by a first convolution layer; and processing the input image by a multi-level semantic information mapping portion to generate a combined image, where the multi-level semantic information mapping portion includes one or more residual spatial and channel attention blocks (RSCBs). Each of the RSCBs includes two or more residual spatial blocks (RSBs) and a channel attention (CA) block. Each of the RSBs includes a spatial attention (SA) block.