66 results about "Texture recognition" patented technology

The invention discloses a panoramic image real-time splicing algorithm and system based on multi-sensor fusion, and particularly relates to the technical field of panoramic image real-time splicing, and the algorithm comprises the following steps: constructing a structured fusion sequence based on multi-source images, postures and position information, optimizing a matching effect through high-density feature extraction and repeated texture recognition, and obtaining a multi-source image fusion sequence; a dynamic foreground and a static background are distinguished by using sparse optical flow so as to improve the visual angle estimation precision, pose fusion optimization is realized in combination with a multi-mode residual error, and the continuity and stability of a spliced image are improved through edge smoothing, brightness tuning and color correction; according to the method, the structured fusion sequence is constructed through multi-source data alignment, so that the data synchronization and splicing stability is improved; identifying repeated regions based on texture direction features, and optimizing feature matching accuracy; and through edge smoothing, brightness harmonizing and color consistency processing, the visual coherence and output quality of the panoramic image are enhanced.

The invention relates to the technical field of intelligent identification, in particular to a highway pavement damage intelligent identification and evaluation method, which comprises the following steps: detecting a structure continuity and texture density region based on a pavement image, extracting crack, pit slot and track contour marks, grouping crack, pit slot and track texture to generate an identification graph, and evaluating the pavement damage. And extracting a crack area texture and gradient change to generate a difference layer, judging trend offset to generate an evolution label, and adjusting a classification boundary to evaluate a damage level. According to the method, the structure continuity and the texture density area are detected, the positioning precision is improved, misrecognition is reduced, direction gradient and texture changes are collected under multiple scales, the crack directions are clustered and grouped, the texture recognition stability is enhanced, the difference of the crack area and the periphery is analyzed, boundary jump is recognized, difference perception is enhanced, and the dynamic recognition depth is improved. Evaluation fuzzy errors are reduced, the overall process improves morphological representation precision and trend tracking and classification boundary stability, and the method adapts to various road condition changes.

The invention discloses an aviation equipment detection and maintenance method and equipment based on target detection and a medium, and relates to the technical field of aviation equipment detection, and the method comprises the steps: inputting a correction image set into a pre-trained target detection network, carrying out the multi-scale feature analysis and edge texture recognition, and generating a part detection result; establishing a component topological relation according to a component detection result, and comparing the component topological relation with a standard configuration by using a graph structure analysis method to generate topological consistency information; performing defect positioning and quantitative analysis on the topological consistency information, extracting a defect position, a defect type and a defect quantitative index, tracking an extension track of the defect quantitative index along with time through continuous time frame difference, and generating a defect parameter; and performing risk assessment on the defect parameters, generating and executing a maintenance scheme, verifying the maintenance effect through real-time image detection, and forming a detection and maintenance closed loop. According to the invention, the accuracy and integrity of aviation equipment detection are finally improved.

The invention relates to the technical field of beef muscle texture recognition, and discloses a multi-modal image AI recognition method and system for beef muscle textures, and the method comprises the steps: collecting a beef image and spectral data; carrying out denoising, enhancement and normalization processing; using a variational auto-encoder to extract fusion features; optimizing a modal fusion strategy through reinforcement learning; reconstructing and enhancing a texture image by using the generative model; the key texture features are subjected to self-adaptive weighted reinforcement; the input classifier outputs a texture recognition result; the system comprises an image acquisition module, a spectrum acquisition module, a data preprocessing module, a multi-modal feature extraction module, a feature fusion module, an image generation module, a texture feature analysis module and an identification decision module. According to the invention, a bimodal joint modeling mechanism of image and spectral information is introduced, deep feature mapping is carried out on two types of modals by means of a variational auto-encoder, a cross-modal expression relationship is established through a submerged space, and the texture expression integrity and resolution are enhanced.

The invention discloses a visual tactile sensor fused with an optical waveguide and a preparation method thereof, and belongs to the technical field of intelligent tactile sensing. The visual tactile sensor comprises a fixed shell, and the top end of the fixed shell is open; the composite tactile sensing layer is arranged at the opening position of the fixed shell, and the composite tactile sensing layer is sequentially composed of a shading layer, a reflecting layer, a red-green spaced fluorescent layer, an elastic deformation layer and a supporting layer from top to bottom; and the optical lighting and sensing component comprises a blue LED light bar, an array light intensity sensor and a camera, the blue LED light bar and the array light intensity sensor are arranged on the two corresponding sides of the composite touch sensing layer respectively, and the camera is arranged at the bottom in the fixed shell. Through fusion of rapid and stable light intensity detection characteristics of optical waveguide sensing and pressure distribution detection capability of the visual tactile sensor, high-resolution space and texture identification and high-sensitivity mechanical response detection are simultaneously realized in a single sensor architecture.

The application discloses a flexible tactile sensing and feedback system and method based on texture recognition, which comprises an electrostatic flexible tactile sensor, a feedback actuator and a tactile feedback control circuit, the electrostatic flexible tactile sensor comprises a tactile interaction film for acquiring texture features and a pressure sensor for texture feature conversion; the feedback actuator completes tactile feedback of the texture feature signal acquired by the flexible sensor in the form of vibration; the tactile control system is mainly connected with the electrostatic flexible tactile sensor and the tactile feedback actuator through the feedback control circuit; the tactile feedback actuator has a stimulation reinforcement structure on the contact surface with the human body, which can further enhance the fine feedback of the tactile feeling. The application can effectively identify the texture features and complete the feedback of the fine tactile feeling in the form of vibration, provides effective assistance for the tactile recovery of the disabled, and has important application value in the game industry, especially in the development of virtual reality technology.

A texture recognition module. The texture recognition module includes: a substrate (1); a driving circuit layer (9) located on one side of the substrate, including a plurality of arrayed driving transistors (12); a first insulating layer (5) located on the side of the driving circuit layer away from the substrate, including a plurality of first vias (13) penetrating the thickness of the first insulating layer; a plurality of photoelectric conversion units (15) located on the side of the first insulating layer away from the driving circuit layer, which are in one-to-one contact with the first electrode of the driving transistor through the first vias; a second insulating layer (16) located on the side of the first insulating layer away from the substrate, including a plurality of second vias (17) corresponding to the photoelectric conversion units; the distance between the edge of the second via and the edge of the surface of the photoelectric conversion unit away from the substrate is less than or equal to a first preset value; a plurality of first electrodes (18) located on the side of the photoelectric conversion unit away from the driving circuit layer; each first electrode covers the photoelectric conversion unit in the second via.

This invention discloses a method and system for detecting coating defects on the surface of rollers based on image generation, belonging to the field of industrial product quality inspection technology. The method includes the following steps: acquiring continuous image data of the roller during a stable rotational speed phase and a sudden increase in rotational speed, recording the start time of the speed change, and generating rotational speed change identification information; organizing the continuous images in chronological order based on the rotational speed change identification information, comparing the brightness distribution of the images before and after the speed change, extracting areas of sudden brightness increase, and generating a set of bright areas. This invention achieves precise correspondence between image acquisition and roller operating state by introducing rotational speed change identification information, eliminating interference from sudden changes in illumination on texture recognition; and through dynamic backtracking of the texture real image set and peeling risk information, it achieves continuous tracking and adaptive detection of coating defects, improving recognition stability and detection reliability.

The invention relates to the technical field of computer vision and artificial intelligence, and particularly discloses a bird's nest texture AI identification verification system and method, and the system comprises an image collection module, an AI processing module, a database module, and a verification interaction module. The image acquisition module acquires a texture image by using high-precision equipment; the AI processing module extracts features, trains a model and matches the features; the database module stores data in a cloud distributed manner; and the verification interaction module supports multi-terminal verification. The method comprises the steps of production end collection and filing, AI model training optimization and terminal verification operation. The system takes the natural texture of the cubilose as an identifier, avoids counterfeiting, does not need a paper label, is high in verification accuracy, is simple to operate by consumers, and is suitable for multiple scenes.

The application provides a large-scale satellite image automatic registration and enhancement method based on adaptive multi-source fusion, comprising: performing orthorectification and multi-resolution unified projection on a multi-source original data set obtained to obtain a coarse registration image set; calculating a local tensor structure of the coarse registration image set, performing repeated texture recognition, generating a final structure representation and a structure reliability; performing regional registration to obtain a full-image dense deformation field, then performing image transformation to be aligned and registration uncertainty estimation to obtain a high-precision registration image and structure registration uncertainty estimation; combining a pseudo-change probability of the high-precision registration image and the structure registration uncertainty estimation to obtain a change reliability weight; under the constraint of the change reliability weight, performing consistency enhancement on the high-precision registration image to output a quantitative and reliable enhanced image. The application realizes high-precision, low-false alarm and quantifiable and reliable automatic registration and enhancement of multi-source large-scale satellite images.

A texture recognition device and a display device are provided. The texture recognition device includes a backlight element, configured to provide first backlight; a light constraint element, configured to perform a light divergence angle constraint process on the first backlight to obtain second backlight with a divergence angle within a preset angle range, the second backlight being transmitted to a detection object; and a photosensitive element, configured to detect the second backlight reflected by a texture of the detection object to recognize a texture image of the texture of the detection object.

The present application relates to the field of electronics. Provided in the embodiments of the present application are a detection assembly, a touch display screen, an electronic device and a control method therefor, which are used for providing a detection assembly integrating functions of bioelectric signal detection and identity recognition signal detection. The detection assembly comprises a texture recognition module, a first conductive layer and a bioelectric signal acquisition module. The first conductive layer is arranged on an acquisition side of the texture recognition module and serves as a biological sampling electrode. Fusing the texture recognition module and the first conductive layer serving as a biological sampling electrode with each other allows one detection assembly to have both functions of identity recognition and bioelectric signal acquisition. Thus, when users place parts to be tested such as hands on detection assemblies, identity recognition signals and bioelectric signals can be obtained.

The invention relates to the field of textile production recognition, in particular to a textile production positioning recognition system based on visual inspection, which comprises a texture recognition end, an image detection end, a cutting position generation end, a first execution end and a knitting positioning end. The method comprises the following steps: processing lines for detecting a target detection image to obtain an image texture value of a unit element, then combining the image texture value with an image with a standard line to obtain a target cutting position, then identifying a cutting piece, generating a region image, marking a standard stitching point in the region image to obtain a region marking image, and finally marking the region marking image to obtain the target cutting position. And overlapping and processing all the area marking images to obtain a target distance value, and obtaining a sewing positioning point according to the target distance value, so that the overall spinning point of the textile is determined according to sewing positioning identification, and the production quality of the textile in the production process is kept consistent.

The present application relates to the technical field of fabric texture detection, and particularly relates to a fabric defect detection method, comprising: collecting a fabric image; inputting spatial domain maps into spatial domain encoders of a ViT-S model respectively; inputting frequency domain maps into frequency domain encoders of the ViT-S model respectively; inputting a spatial domain first layer block vector, a spatial domain second layer block vector and a spatial domain third layer block vector into a TPP module; inputting a frequency domain first layer block vector, a frequency domain second layer block vector and a frequency domain third layer block vector into the TPP module; performing Concat operation on a fusion first layer feature, a fusion second layer feature, the spatial domain third layer block vector, the frequency domain first layer block vector, the frequency domain second layer block vector and the frequency domain third layer block vector, and then inputting the result into a classifier to output a classification result. The present application overcomes the problem that the recognition accuracy of traditional methods for complex fabric texture needs to be further improved.

The application discloses a control system of a six-degree-of-freedom hydraulic mechanical arm for dismounting and mounting a cylinder, relates to the technical field of cylinder dismounting and mounting, and is characterized in that a multi-angle polarized light source is configured at a light jump area on the surface of the cylinder, a brightness gradient sequence is established, a multi-exposure time chain is constructed, the light intensity is controlled in time, the reflection area is formed into a continuous light level, spot energy diffusion is performed to form a continuous texture transition zone, a pose feature track is extracted, a motion edge is generated by combining light intensity variation to identify an optical flow clue, a visual guide channel is constructed according to the clue, a hydraulic actuator is driven to perform pose fine adjustment, and synchronous compensation and pressure regulation of the alignment process are realized. The application realizes the continuous transition of the light on the surface of the cylinder through the multi-angle polarized light and the multi-exposure time chain, improves the imaging accuracy and the texture recognition stability, and realizes the pose fine adjustment and the pressure balance control by combining the optical flow guide, so that the hydraulic mechanical arm can complete adaptive assembly in a reflection interference environment, and the dismounting and mounting accuracy and safety are significantly improved.

The invention belongs to the technical field of image processing, and particularly relates to an image recognition system and an image recognition sensor applied to a camera, and the system comprises an image recognition module, a preprocessing module, a texture recognition module, an optical flow track module and a camera tracking module which are in communication connection in sequence. FPGA hardware acceleration is combined with redundancy reduction of particle screening, so that an image recognition system can stably perform full-link processing on a high-speed camera, and the non-delay tracking requirement of a high-speed moving target is met; through a texture channel fused by multiple edge channels, the edge integrity rate in a high-speed motion fuzzy scene is improved, and the tracking loss rate of light field particles is reduced; through a designed double-constraint optical flow mechanism, the optical flow estimation error of a high-speed moving target is reduced, the inter-frame position deviation of camera tracking is reduced, and accurate positioning of the high-speed target is realized, so that the problems of image precision reduction and tracking breakage caused by high-speed movement of the target in high-speed camera image recognition are solved.

The invention discloses a GPU context awareness-based code editor dynamic text rendering acceleration method, which comprises the following steps of: establishing a multi-dimensional semantic tag system comprising variables, functions, errors and structures, extracting semantic sub-dimensions of characters, generating character semantic tags through operation, caching the character semantic tags to a semantic tag SSBO, constructing a mapping rule of the semantic tags and rendering styles, and carrying out text rendering on the rendering styles according to the mapping rule. The CPU encodes the semantic rules into semantic rule textures; identifying structural units and layout attributes thereof, encoding the structural units into structural textures by the CPU, and uploading the structural textures to a structural layout SSBO of the GPU; the calculation shader reads two types of SSBO, respectively allocates four types of weights of semantics, interaction, structure and effect for characters, and generates a pattern weight texture after determining a fusion coefficient; the vertex shader calculates the final screen position of the character in parallel according to the structure unit to which the character belongs; and the fragment shader reads the two types of SSBO, each texture and the final screen position output by the vertex shader, hierarchical fusion is carried out according to a weight sequence, and character pixel-level rendering is completed.

The invention discloses a bronze ware texture recognition method based on multi-principle data annotation and deep learning, and the method comprises the steps: S1, obtaining an image containing bronze ware texture, and constructing an original image data set; s2, according to a preset multi-principle data labeling strategy, labeling textures in the original image data set, and generating a high-quality labeling data set; s3, inputting the high-quality annotation data set into a preset deep learning target detection network model for training and evaluation to obtain a mature model capable of identifying the texture of the bronze ware; and S4, processing the new bronze ware image by using the mature model, and outputting types and position information of textures contained in the image. According to the method, the recognition accuracy can be remarkably improved, the multi-scale and detail problems are effectively solved, the model generalization ability can be enhanced, and meanwhile, the method has universality.

This invention discloses a GPU-based context-aware code editor dynamic text rendering acceleration method. It establishes a multi-dimensional semantic tagging system including variables, functions, errors, and structures. Semantic sub-dimensions of characters are extracted and processed to generate character semantic tags, which are then cached in semantic tag SSBOs. A mapping rule between semantic tags and rendering styles is constructed, and the CPU encodes the semantic rules into semantic rule textures. Structural units and their layout attributes are identified, and the CPU encodes them into structural textures and uploads them to the GPU's structural layout SSBOs. The compute shader reads two types of SSBOs and assigns four weights—semantic, interactive, structural, and effect—to the characters. After determining the fusion coefficients, it generates style-weighted textures. The vertex shader calculates the final screen position of the character in parallel based on the structural unit to which the character belongs. The fragment shader reads the two types of SSBOs, each texture, and the final screen position output by the vertex shader, and fuses them layer by layer according to the weight order to complete pixel-level character rendering.

The invention discloses a wood texture identification method based on multi-scale feature fusion. The method comprises the following steps: firstly, extracting primary features of a wood texture image by using a ResNet50 backbone network; and then respectively inputting into a global branch and a local branch. The global branch performs down-sampling of different scales, then constructs a channel high-similarity attention mechanism under each scale, screens out low-correlation items in a channel similarity matrix through a dynamic threshold mechanism, performs weighted aggregation, performs residual connection with input primary features, and finally fuses the features of all scales, so as to obtain a channel high-similarity attention mechanism; and outputting the global features. The local branch calculates the spatial correlation between the positions of the pixel points in the primary features, screens out low correlation items through a dynamic threshold mechanism, and connects the low correlation items with input feature residuals after weighted aggregation to obtain local features. And finally, carrying out adaptive dynamic weighted fusion on the global features and the local features, accurately highlighting an area most related to classification, and remarkably improving the perception capability of the model for subtle differences.

This invention discloses a method, system, and device for generating fruit pattern codes based on fruit spot texture recognition. The method includes: acquiring a top image of a fruit; segmenting the fruit image from the image using a convolutional neural network model and identifying all spots on it; generating an outer rectangle for the fruit image and using its center as the origin; calculating the center coordinates of the outer rectangle for each spot to form a coordinate array; calculating the distance between the center of the outer rectangle for each spot and the origin to form a distance array, and arranging the distance array and its corresponding coordinate array in ascending order; using the straight line between the first coordinate in the coordinate array and the origin as a guideline, calculating the angle between the line connecting the current coordinate and the origin and the guideline to form an angle array; desensitizing the distance array and the angle array; concatenating the desensitized distance array and the angle array and using a hash function to obtain their hash code. This invention realizes the electronic encoding of fruits with spots and patterns, providing a foundation for subsequently attaching commercial information about the fruit.

The invention relates to a garbage cleaning method and system based on texture recognition and a medium. The method comprises the steps of obtaining a ground image of a target area; performing texture identification and image analysis on garbage in the ground image, and judging whether liquid garbage exists in the target area or not; determining a cleaning strategy of the target area according to whether liquid garbage exists in the target area or not; and cleaning the target area according to the cleaning strategy. On the basis of the short-wave infrared camera, the ground image can be collected in real time in the working process of the cleaning equipment, the cleaning strategy of the target area can be accurately determined and executed in combination with the texture recognition and image analysis technology, and compared with traditional intelligent cleaning equipment, the cleaning effect is better, the process is simple, and the execution efficiency is high.

The invention relates to a garbage cleaning method and system based on texture recognition and a medium. The method comprises the steps of obtaining a ground image of a target area; performing texture identification and image analysis on garbage in the ground image, and judging whether liquid garbage exists in the target area or not; determining a cleaning strategy of the target area according to whether liquid garbage exists in the target area or not; and cleaning the target area according to the cleaning strategy. On the basis of the short-wave infrared camera, the ground image can be collected in real time in the working process of the cleaning equipment, the cleaning strategy of the target area can be accurately determined and executed in combination with the texture recognition and image analysis technology, and compared with traditional intelligent cleaning equipment, the cleaning effect is better, the process is simple, and the execution efficiency is high.

The application discloses an aviation equipment detection and maintenance method and device based on target detection, and a medium, relates to the technical field of aviation equipment detection, and comprises the following steps: inputting a correction image set into a pre-trained target detection network, performing multi-scale feature analysis and edge texture recognition, and generating component detection results; establishing component topological relations according to the component detection results, and comparing the component topological relations with a standard configuration by using a graph structure analysis method to generate topological consistency information; performing defect positioning and quantitative analysis on the topological consistency information, extracting defect positions, defect types and defect quantitative indexes, tracking the expansion trajectories of the defect quantitative indexes with time through continuous time frame difference, and generating defect parameters; performing risk assessment on the defect parameters, generating a maintenance scheme, and executing the maintenance scheme, verifying the maintenance effect through real-time image detection, and forming a detection and maintenance closed loop. The application finally improves the accuracy and integrity of aviation equipment detection.

The invention discloses a wall aging damage intelligent diagnosis method and system based on an image recognition algorithm, and the method comprises the following steps: obtaining multi-source data of a building facade, the multi-source data comprising an unmanned plane oblique photography image, a handheld camera detail image, laser point cloud data and on-site environment metadata; time registration, space registration and standardization processing are carried out on the multi-source data, and a hierarchical labeling data set containing damage types and size information is constructed; according to the invention, through multi-source image fusion and standardization processing, the robustness of the model to multi-illumination, multi-climate and multi-material conditions is enhanced; unrelated areas such as vegetation, window frames and metal components are removed through intelligent extraction of wall body areas, and the recognition efficiency is improved; three-dimensional structure features are adopted to participate in identification decision, and the problem that a two-dimensional image cannot reflect deep damage is solved; through collaborative decision-making of the structure analysis model, the texture recognition model and the semantic segmentation model, the recognition capability of complex damage is improved.

The invention discloses a pork quality evaluation method and system based on texture recognition, and relates to the technical field of texture recognition analysis, and the method comprises the steps: marking pork, using a hyperspectral camera to shoot a pork multispectral image, and carrying out the preprocessing; constructing an image pyramid to perform image layering, and performing optimization processing on each layer of image based on the pixel relation of each layer of image to generate a multi-resolution image; re-fusing the multi-resolution images into an optimized multi-spectral image, and evaluating the pork quality based on the optimized multi-spectral image; the pork quality evaluation result is displayed and stored according to the mark number. According to the method, the intensity of detail information in the pork multispectral image is improved, help is provided for subsequent pork quality evaluation, the texture information is extracted and combined with the pixel state of the multispectral image to be input into the convolutional neural network for pork quality evaluation, and the accuracy and efficiency of evaluation are improved.

The invention discloses a diffuse interstitial lung disease texture recognition model and an implementation method thereof, and relates to the technical field of lung disease texture recognition, the diffuse interstitial lung disease texture recognition model comprises a PACS database, a training server, an application server, a plurality of data processing units and a user side; the training server is connected with the PACS database through the Ethernet, and meanwhile, the training server interacts with the user side through the Ethernet to sketch a volume of interest (VOI) for model training; the training server and the application server are connected with the plurality of data processing units through the Ethernet, and the data processing units are connected with the image reconstruction work stations one by one through the Ethernet; according to the invention, on the basis of a fully-supervised prior lung texture segmentation training set, classification fineness which cannot be reached by an unsupervised clustering scheme is realized; compared with an existing full supervision scheme, the method achieves the improvement of classification accuracy and segmentation precision through a deep learning network instead of a lung density quantitative segmentation scheme.

The application discloses a shale oil reservoir texture identification method based on wavelet decomposition and relates to the technical field of oil and natural gas exploration. The application is based on full borehole covering electric imaging button conductivity array data, extracts the local abrupt point position of the original conductivity signal through one-dimensional wavelet change, then determines the maximum and minimum values by using the first derivative and the second derivative, and finally realizes the pickup of the sandy laminations and shale foliation of the shale oil reservoir by using the least square method to fit a sine function. The example result shows that the method has good texture pickup effect, the model can be developed into a software system, the required parameters are simple, and the method has good popularization advantages.

This invention provides a pattern and texture recognition method and apparatus based on neural networks, relating to the field of texture recognition technology. Specifically, it includes: constructing a texture recognition model; extracting multi-scale features from the input image using convolutional kernels of different sizes; weighting and fusing the features to generate a dynamically fused feature map; using a channel attention mechanism to extract global channel information and generate channel attention weights; utilizing a spatial attention mechanism, extracting spatial information and generating spatial attention weights by combining average pooling and max pooling of the feature map with a spatial feature interaction formula; fusing channel and spatial attention weights to weight the feature map; performing global average pooling on the optimized feature map to generate a global feature vector; and combining an improved regularized classification formula to generate the final classification result. This method achieves efficient pattern and texture recognition through multi-scale feature extraction, attention mechanism fusion, and regularized classification.

The invention discloses a coal sample bag real-time detection and violation replacement early warning method. The method comprises the steps of sample initialization and monitoring starting, sample circulation monitoring and violation early warning and authorized unsealing and monitoring termination. The coal sample bag is provided with a flexible electronic tag and an anti-counterfeiting texture; the radio frequency radar and the camera are deployed at key nodes of a circulation path; the reasoning server is in signal connection with the radio frequency radar and the camera and is used for operating a reasoning algorithm; the personnel identity verification device and the anti-counterfeiting texture recognition device are arranged at the unsealing use node; according to the invention, through three core modules of anti-counterfeiting mark enhancement, multi-dimensional state perception and intelligent risk research and judgment, the tamper-proofing and replacement-preventing capabilities in the sample circulation process are significantly improved.