83 results about "Texture extraction" patented technology

Disclosed herein is a 3D urban modeling apparatus and method. The 3D urban modeling apparatus includes a calibration unit for calibrating data about a translation and a rotation of at least one capturing device at a time that input aerial images and terrestrial images were captured. A building model generation unit generates at least one 3D building model based on the aerial images and the terrestrial images to which results of the calibration have been applied. A terrain model generation unit generates a 3D terrain model by converting an input digital elevation model into a 3D mesh. A texture extraction unit extracts textures related to the building model and the terrain model from the aerial images and the terrestrial images. A model matching unit generates a 3D urban model by matching the building model with the terrain model, which are based on the textures, with each other.

A system and method for real-time texture mapping for an augmented reality system are described. A viewing device includes an optical sensor to capture an image of a real-world object. A texture extraction module extracts a texture of the image of the real-world object. A recognition module identifies the real-world object based on the captured image. A texture mapping module retrieves a virtual object corresponding to the identified real-world object, maps the texture to the virtual object, dynamically updates the texture to the virtual object in real time, and generates a visualization of the virtual object in a display of the viewing device.

The invention discloses a remote sensing image processing method combined with the shape self-adaption neighborhood and the texture feature extraction for image preprocessing. The method includes subjecting compressed image to a gray level co-occurrence matrix calculation; subjecting the generated gray level co-occurrence matrix to S coefficient modification of an SAN (Storage Area Networking) irregular object window to obtain a regular matrix; calculating a new co-occurrence matrix according to the modified regular matrix and selecting texture descriptors with obvious feature and low correlation; extracting texture feature map in the SAN irregular images; and calculating to obtain accurate images with combination feature which is overall comprehensive feature of neighborhood. According to the method, the overall classification accuracy based on a shape self-adaption neighborhood method can be improved by 4%. The method can not only extract the texture feature in the SAN irregular images of remote sensing images completely, but also process the extraction of mixed pixel feature of the fuzzy edge of earth surface objects, and is applicable to texture extraction of earth surface objects in natural states.

The invention, which belongs to the technical field of construction of an urban three-dimensional model, especially relates to a dense-point-cloud-based rapid construction method of an urban three-dimensional model. With the method, technical problems of low efficiency and the like in the prior art can be solved. The dense-point-cloud-based rapid construction method comprises: A, carrying out adaptive blocking on point cloud data; B, carrying out three-dimensional TIN model construction on the divided blocks; C, selecting an image with the best quality from a multi-view slanted image for each surface of the three-dimensional TIN model and using the image as a texture data source; D, after selection of the texture data sources with the best qualities, carrying out texture extraction; E, carrying out substitution of textures of shielded parts to complete vulnerability repairing of the textures; and F, pasting the preferably selected textures to the surface of the three-dimensional TIN model to generate a three-dimensional model with a lifelike effect. The method has advantages: firstly, the construction efficiency is high; and secondly, the texture quality is improved and the lifelike effect is realized.

The invention discloses a building texture extraction method based on a UAV low-altitude aerial survey system. The method comprises the first step of automatic registration between a building contour line and an image, the second step of semi-automatic registration between the building contour line and the image on the basis of multi-view image linear matching, the third step of building facade texture extraction and the fourth step of building facade texture geometric correction. The building texture extraction method has the advantages that the method for automatic registration between the building contour line and the image is provided for solving the problem that registration is not achieved between a vector line and an image during building texture mapping; in consideration of the complexity that the mapping relation between a building space contour line and the image linear feature is built in multiple images, the image linear feature semi-automatic matching strategy based on multi-view geometric constraints is provided.

The invention discloses a video-processing-based crowd density and distribution estimation method, which belongs to the technical field of computer video processing and is particularly suitable for a flowing crowd with a high / medium density grade. The method comprises the following steps of: detecting sparse characteristic points by utilizing a three-dimensional Hessian matrix; performing dynamic texture extraction based on a time-space local binary pattern; performing spectral analysis based on Fourier transform; estimating overall crowd density based on a support vector machine; and mapping a frequency value to a jet color map, and performing Gaussian smoothing to obtain displayed spectrum-based local crowd density. Sparse time-space local binary dynamic texture characteristics of an image are extracted, and video time-space characteristics are extracted based on the symmetry of a time-space domain, so that robustness is improved, a calculated amount is decreased, and the method is suitable for the estimation of a large-scale moving crowd.

The application provides a method and system for smart community face recognition. The method of the application includes: acquiring a face image of a target human body; performing edge detection on the face image, and dividing the face image into a plurality of regions according to a closed edge; performing texture extraction on a key region in the plurality of regions to generate a texture feature vector of the key region, wherein the key region being a face organ region determined according to a clustering rule; and comparing and recognizing the key region based on the texture feature vector, and then determining the identity information of the target human body. The method and system of the invention are a few in collected facial features, small in amount of calculation, simple in algorithm, and high recognition speed, thereby improving the speed of face recognition, adapting to actual conditions such as distributed low-configuration hardware resources, mass recognition objects andhigh-frequency triggering in a smart community scenario, and realizing face recognition in an efficient, reliable, fast and low-cost manner.

The invention provides an automatic optic inspection method for surface defects of metal cylindrical workpieces and belongs to the automatic optic inspection methods. The automatic optic inspection method is particularly suitable for detecting surface defects of special metal cylindrical workpieces and comprises the following steps: a linear array CCD acquires the unrolled images of a metal cylindrical workpiece, the local binary pattern (LBP) and the local variance (LVAR) are combined, that is, the image variance intensity serves as the weight value to adjust the local texture extraction and measurement result of the LBP, not only the LBP texture space structure mode but also the texture intensity contrast mode is drawn into consideration, finally, the photoelectric image fault features of metal cylinders are accurately extracted, the sizes, the positions, the ranges, the severities and the like of the defects are determined, and the automatic intelligent detection of the workpiece production quality is realized. Therefore, nonuniform lighting caused by the metal materials of the workpieces per se is avoided, the accuracy in detecting tiny defects is improved, the misjudgement rate is lowered, and automatic optic inspection of the metal cylindrical workpieces is achieved.

The invention is an edge- and texture- data synthesized pyramid image fusion method, firstly establishing a pyramid structure based on edge and texture features, considering the linear relations between binomial Gauss filter and texture extraction filter as well as edge extraction filter, adopting singular value decomposition to find the corresponding coefficient of the edge and texture images, then utilizing these features of each scale image to express each layer of the decomposed image, and finally make fusion by the fusion policy based on similarity measure and remarkability measure. It largely improves the quality of the image by a large margin and has an important significance and practical value in the follow-up treatment and image display of various application systems.

The invention discloses a key frame extraction method for RGBD 3D reconstruction. The method includes at first, dividing a plurality of data frames that are temporally adjacent to each other into a group for the RGBD data flow acquired by a camera and the camera pose estimated by a visual odometer; and for each group of data, projecting each frame of depth image to a first frame of depth image according to the camera pose and the camera parameters; projecting each group of first frame of RGB image to the remaining frames of image to obtain the gray value of each projected RGB image by linear interpolation; estimating the blur degree of each frame of RGB image, and combining with the weight of the corresponding projected depth image to obtain the weight of the projected RGB image; and fusing the intra-group projected depth images and the projected RGB images according to the obtained weights to obtain the RGBD key frame. The key frame extraction method reduces the void and noise of data acquired by a depth camera to obtain clear depth images and RGB images, and provides reliable data source for other works in the 3D reconstruction such as the camera pose global optimization and texture extraction.

A 3D model appearance texture extraction method and a 3D model appearance texture extraction device of the invention relate to the technical field of 3D models. The method comprises the following steps: getting a preset initial proxy model, and getting a 3D input image input by a user; registering and fitting the initial proxy model according to the 3D input image and the initial proxy model to form a fitted proxy model; and performing intrinsic image decomposition on the 3D input image according to the fitted proxy model, generating the geometric shape of the 3D input image after intrinsic image decomposition and the albedo information of the 3D input image, and determining the secondary geometric texture and the material texture of the 3D input image. The secondary geometric texture and the material texture are appearance textures to be extracted. The problem that it is hard to realize fast and effective extraction of the appearance textures of a 3D model in the prior art is avoided.

The invention provides a method and device for extracting a biological feature, a biological identification method and system. The method for extracting the biological feature comprises the following steps: utilizing a multi-scale multidirectional iteration filter to filter an ROI (region of interest) of an original palm print image and the ROI of an original palm pulse image, thereby extracting texture; performing competitive encoding according to the maximum response of the filter; and respectively taking the obtained encoded images as an identification feature of palm print and the identification feature of palm pulse. According to the method for extracting the biological feature provided by the invention, the calculation volume for extracting the biological feature in the biological identification technology is efficiently reduced, the time for extracting the biological feature is saved and the identification efficiency of the whole biological identification system is increased.

The invention belongs to the technical field of three-dimensional modeling, and discloses a novel city three-dimensional modeling control system and method. The system comprises an image input unit, acamera calibration unit, a parallel moving and rotating data input unit, a digital elevation model input unit, a landscape model generation unit, a texture extraction unit, an architectural model generation unit, an architecture model storage unit, a model matching unit and a display unit. According to the system, by using the digital elevation model (DEM) and images acquired from an aerial vehicle and the ground, modeling of three-dimensional appearance of city architectures and three-dimensional landscape is achieved, and realistic textures are extracted from the images; three-dimensional architecture models can be generated by using the images taken from the air in vertical and inclined directions except for the DEM, textures of the related three-dimensional models can be extracted, the three-dimensional models generated by the images acquired from the ground can be simulated more accurately, or the textures with higher definition can be extracted.

The invention discloses a surface crack texture extraction method based on fusion of a seepage algorithm and an adaptive Canny algorithm. The method comprises the following steps: getting and preprocessing a surface crack image; processing the preprocessed surface crack image by using an adaptive Canny algorithm and a seepage algorithm respectively to get a first crack image and a second crack image; extracting the low-frequency component and high-frequency component of the first crack image and the low-frequency component and high-frequency component of the second crack image through wavelet transform; superimposing the extracted low-frequency components to get a fused low-frequency part, and superimposing the extracted high-frequency components to get a fused high-frequency part; and getting a reconstructed image through inverse wavelet transform, wherein the reconstructed image is a fused crack image. A more real crack image is obtained by the method. The method is used in a product appearance quality detection assembly line. By using the method, the detection efficiency is improved, time is saved, and the human cost is reduced.

An image processing device includes a texture extraction unit to extract a texture component of a G corrected image in which blur or defocus of a structure component of a G image including a G component of an input image is corrected, a mask generation unit to generate a mask image in which the synthesis amount of the texture component of the G corrected image to an R corrected image and a B corrected image is decreased for a region in which at least one of correlation between a variation of the G component and a variation of the R component or correlation between the variation of the G component and a variation of the B component is weak, and a synthesis unit to synthesize the texture component of the G corrected image to the R corrected image and the B corrected image using the mask image.

The invention relates to a method for extracting multi-dimensional texture of nodi from medical images. The images are segmented by using a region growing method, content texture characteristic parameters of nodi are extracted by adopting a gray level co-occurrence matrix method, and edge texture characteristic parameters of the nodi are extracted by adopting a Curvelet conversion method. The method can enhance the edge information of the nodi and maintain the edge information thereof better so as to improve the accuracy of judging the properties of the nodi.

The invention refers to an image processing device, image processing method and program. the image processing device includes a texture extraction unit to extract a texture component of a g corrected image in which blur or defocus of a structure component of a g image including a g component of an input image is corrected, a mask generation unit to generate a mask image in which the synthesis amount of the texture component of the g corrected image to an r corrected image and a b corrected image is decreased for a region in which at least one of correlation between a variation of the g component and a variation of the r component or correlation between the variation of the g component and a variation of the b component is weak, and a synthesis unit to synthesize the texture component of the g corrected image to the r corrected image and the b corrected image using the mask image.

The invention discloses a seismic data reconstruction method based on a texture constraint U-net network, and the method comprises the steps: S1, segmenting a training data set through a K-means algorithm, and obtaining a texture label of training data; S2, training a U-net network as a texture extraction network through the training data set and the texture label to obtain optimized texture extraction network parameters; S3, connecting the trained texture extraction network with a reconstruction network in series, extracting texture information of label data and reconstruction seismic data byadopting optimized texture extraction network parameters, and obtaining texture loss; and S4, reconstructing the network through texture loss optimization, and then reconstructing the seismic data. The characteristic that the seismic data has rich texture information is combined, and the reconstruction accuracy and continuity of the seismic event can be improved under the condition of limited samples by strengthening texture learning.

The invention discloses a picture character positioning method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining a target picture which comprises a to-be-recognized character; inputting the target picture into a texture extraction model to obtain a plurality of texture feature layers of the target picture, wherein the texture extraction model is obtained through analysis according to texture features in a historical image and used for extracting the texture feature layers in the input picture; screening out a basic texture feature layer from theplurality of texture feature layers; performing feature superposition on the basic texture feature layer to obtain a texture feature layer of the to-be-recognized character; and obtaining the position of the to-be-recognized character in the target picture according to the texture feature layer of the to-be-recognized character. The method is short in time consumption and high in accuracy.

The invention discloses an automatic insulator identification and analysis system. The system performs the following steps: performing picture preprocessing, extracting an insulator in a picture, detecting the defects of the insulator, extracting the insulator in the picture by the step of extracting the insulator in the picture after the picture preprocessing is performed on the picture, then the defects in the picture are detected by the step of detecting the defects of the insulator, the picture preprocessing comprises an image collection module and global linear transformation, the global linear transformation is connected with the image collection module and mainly converts the picture into a gray scale map, so that the contrast and the quality of the picture are improved, the insulator in the extracted picture comprises Sobel edge separation, morphological filtering, insulator making and noise point separation, the insulator defect detection comprises texture extraction, classification processing, self-destruction, gap locating and stain detection, pooling work of the insulator is realized, and the operation safety of a power system is ensured.

Camera-based texture extraction in Augmented Reality (AR) systems is enhanced by manipulating projected patterns. One or more fine line patterns are projected onto a textured surface, a Moire interference pattern measured, and different properties of the projected pattern(s) adjusted until the Moire interference pattern measurements indicate that a similar texture pattern to that of the three dimensional target is being projected. Thereby, the target texture may be more closely matched even as sub-pixel resolutions, variable lighting conditions, and / or complicated geometries.

The invention provides a gradient correction and multi-direction texture extraction-based sketch generation method and device. The method comprises the following steps of: carrying out grayscale transformation on an input image to obtain a grayscale image; processing the grayscale image to obtain a pre-processed image, and segmenting the pre-processed image to obtain a segmented image; carrying out gradient processing on the grayscale image to obtain gradient information; determining a texture direction of each segmented region of the segmented image according to the gradient information; generating a white noise image according to the input image; combining the texture direction of each segmented region of the segmented image and the white noise image to generate a texture map; processing the input image to obtain an outline map; and fusing the texture map and the outline map to generate a sketch. According to the method and device, multidirectional textures are used, so that the generated sketch is close to a real freehand sketch and the sketch effect is richer and more alive.

The invention provides a detection method for a frozen lake in a remote sensing image. The method comprises a texture extraction step, an energy analysis step, and a result output step. At the texture extraction step, texture feature extraction is carried out on a to-be-detected remote sensing image and a fractal dimension graph is obtained. At the energy analysis step, an energy characteristic of a local image block of the fractal dimension graph is analyzed by using a regional contrast ratio and a local entropy feature. At the result output step, a detection result is outputted based on the analyzed energy characteristic. In addition, the invention also provides a detection system for a frozen lake in a remote sensing image. According to the detection method and system, the specific texture characteristics of a frozen lake are analyzed and an interference source is extracted from a background, so that the effect in a low-noise-ratio / signal-to-clutter-ratio scene is good. Moreover, the method can be operated quickly and is easy to program; and the method and system can be applied and operated in various systems in a cross-platform manner.

The invention relates to the technical field of coal flow recognition. The invention relates to a conveyor belt, in particular to a method and system for identifying coal flow in the conveyor belt based on video monitoring. ANALYSIS BY REAL-TIME VIDEO PICTURE, a belt picture in the video picture is extracted and separated from coal in the belt; according to the coal flow detection method, the coalflow detection device and the coal flow detection system, extraction of coal flow pictures is achieved, the real-time coal flow is visually and clearly displayed to a user, in the process, abnormal behaviors such as belt deviation and sudden overexposure can be recognized, the real-time width of the coal flow can be displayed, and overload and no-load behaviors can be simply and visually detected; the method comprises the following steps that S1, a system is built; s2, acquiring a real-time picture; s3, extracting and converting an analysis area; s4, performing image sharpening processing; s5, texture extraction; S6, coal-belt separation; s7, calculating a coal flow proportion; and S8, overload and no-load recognition.

The embodiment of the invention provides a building texture extraction method and a device based on an inclined image. The method comprises the following steps: obtaining coordinates of top corner points of a building based on a stereoscopic image pair containing the building, and constructing a building skeleton model based on the top corner points; selecting the top texture image source from vertical image based on distance constraint and location constraint, and selecting the elevation texture image source from inclined image based on angle constraint, location constraint and area constraint, based on the edge detection method and the transform detection straight line method, extracting the top texture from the top texture image source, extracting the facade texture from the facade texture image source, and mapping the top texture and the facade texture to the building skeleton model. Using multi-angle simultaneous image acquisition method, we can obtain more abundant and comprehensive building side texture, so as to improve the model texture information collection, processing method, and improve the speed and quality of modeling.

The invention belongs to the technical field of ultrasonic image processing, and particularly relates to a meat quality detection system and method based on ultrasonic signals. The meat quality detection system based on the ultrasonic signals comprises an ultrasonic image collection device, a texture extraction module and an image sampling module, wherein the ultrasonic image collection device is used for collecting an ultrasonic image of meat to be detected; the texture extraction module is used for extracting the texture characteristic of the ultrasonic image through the texture extraction algorithm to obtain a texture image; the image sampling module is used for calculating the pixel mean value and the pixel variance value of the texture image and judging the texture evenness of the texture image according to the pixel variance value of the texture image. By the adoption of the bi-dimensional empirical mode decomposition and the bi-dimensional spectral characteristic, the texture image is obtained, the texture evenness of the texture image is judged according to the pixel variance value of the texture image and serves as an index for evaluating the quality of the meat, the system is simple, operation is convenient, and the quality of the meat is reflected visually, rapidly and accurately.

The invention discloses a building texture extraction method based on oblique photogrammetry. The method comprises the following steps of acquiring a vertical image and an oblique image of a building;screening out an optimal top texture source of the building from the vertical image of the building according to a dual-constraint model, wherein the dual-constraint model comprises whether the texture region is in the center of the vertical image and whether the texture region is complete; screening out the optimum texture source of the vertical surface of the building from the inclined image ofthe building according to the multi-combination constraint model, wherein the multi-combination constraint model comprises the visibility of the facade texture surface on the inclined image and whether the inclined image of the building completely comprises the facade texture surface and whether the projection area of the facade texture surface on the inclined image is maximum or not; and detecting and extracting the texture contour line of the building from the screened optimal texture source of the top of the building and the optimal texture source of the building facade. According to the method, the optimal texture image source can be accurately positioned, and the accurate position of the building texture information can be accurately positioned and extracted.

The invention provides a 4D video rendering method comprising steps of camera selection, visible patch calculation, visible patch texture extraction, fusion weight map calculation, edge blending, andsynthesis rendering. With the method, texture chartlet rendering is carried out on a three-dimensional model of a main scene body obtained by 4D capturing from shot multi-camera image data efficientlywith high quality, so that the three-dimensional model is able to represent a 4D video effect vividly.

The invention relates to a dynamic texture waterfall modeling method combining with multiple physical attributes, belonging to the technical field of virtual reality science. The dynamic texture waterfall modeling method comprises the following steps of: (1) measuring on the spot to acquire physical attribute parameters of real waterfalls in different environments, such as flow velocity, flow rate, head drop, terrain and sunlight which influence the forms of the real waterfalls, and also recording video textures in corresponding states by using a video camera; (2) extracting a plurality of control parameters from the dynamic textures of a waterfall model; (3) reasonably reorganizing acquired data in a database, and building distribution models for the physical attributes through correlation analysis; (4) determining demands according to scenes, and calculating real data inside the database by utilizing a mapping law to obtain the dynamic textures of the waterfalls, which approac to real effect; and (5) rendering the waterfall scenes by using the dynamic textures. According to the invention, a statistic law of multiple physical attributes and the dynamic textures of the waterfalls is extracted, high-lifelike waterfall scene modeling is carried out on the waterfalls in any environmental form by utilizing statistical distribution and the traditional data, the defect of model texture distortion brought about by ignoring physical factors in the traditional waterfall modeling is overcome and the real simulation of the surface textures of waterfall flows driven by the physical attribute parameters is realized.