259 results about "Color texture" patented technology

An image retrieval system performs both keyword-based and content-based image retrieval. A user interface allows a user to specify queries using a combination of keywords and examples images. Depending on the input query, the image retrieval system finds images with keywords that match the keywords in the query and / or images with similar low-level features, such as color, texture, and shape. The system ranks the images and returns them to the user. The user interface allows the user to identify images that are more relevant to the query, as well as images that are less or not relevant to the query. The user may alternatively elect to refine the search by selecting one example image from the result set and submitting its low-level features in a new query. The image retrieval system monitors the user feedback and uses it to refine any search efforts and to train itself for future search queries. In the described implementation, the image retrieval system seamlessly integrates feature-based relevance feedback and semantic-based relevance feedback.

The invention provides a plastic profiled bar decorative layer combination material with natural texture for co-extrusion molding, formed by mixing 15-80wt% of plastic substrate, 15-80wt% of composite pigment and 0-50wt% of inorganic powder, wherein the composite pigment comprises pigment or dye with natural color texture effect; inorganic pigment and inorganic powder in the composite pigment need to be respectively and actively processed by a surface conditioning agent or a coupling agent; the combination material provides the surface of the plastic profiled bar with a high-polymer material co-extrusion composite layer with stronger adhesion and endows the surface of the plastic profiled bar with the metal-like texture effects, such as the colors of gold, silver, copper or other natural metal materials. The decorative layer has a certain thickness, high hardness, strong scratch resistance and longer product service life and can be used for industries, such as electronic appliance shells, plastic color windows and doors, plastic profiled bars, furniture sheets, and the like.

A method of forming at least one three dimensional (3D) color image of at least one object in a target space. The method comprises projecting, each of a plurality of projection cycles, a sequence comprising a plurality of gray coded light patterns, each colored in one of red green or blue, on a target space, capturing a plurality of two dimensional (2D) images of the target space during a plurality of acquisition cycles, each the acquisition cycle being timed to correspond with the projection of at least a sub sequence of the sequence, the sub sequence comprising red, green, and blue gray coded light patterns of the plurality of gray coded light patterns, extracting range data and color texture information of at least one object in the target space from the plurality of 2D images, and forming a 3D color image of the range data and color texture information.

A method and apparatus is described here that categorizes images by extracting regions and describing the regions with a 16-dimensional subscene feature vector, which is a concatenation of color, texture, and spatial feature vectors. By comparing the spatial feature vectors in images with similarly-obtained feature vectors in a Gaussian mixture based model pool (obtained in a subscene modeling phase), the images may be categorized (in a subscene recognition phase) with probabilities relating to each region or subscene. Higher probabilities are likelier correlations. The device may be a single or multiple core CPU, or parallelized vector processor for characterizing many images. The images may be photographs, videos, or video stills, without restriction. When used real-time, the method may be used for visual searching or sorting.

An algorithm for estimating which pixels belong to the background of a scanned page. The algorithm is particularly designed to handle situations in which the page background varies in color / intensity, as is case when bleed-through artifacts from the reverse side of the page appear in the background. In determining background regions, including properly classifying bleed-through artifacts regions as such, the algorithm uses multiple local and global criteria for making the determination. In addition to being able to find large connected pieces of background, the algorithm is also able to find isolated islands of background by analyzing transition characteristics of neighboring regions. Regions are identified on the basis of similar local features and also by the nature of transitions between foreground regions that do not directly share a boundary. An adaptive white-point adjustment technique based on identified background regions improves the perceived quality of the printed output.

A method for reconstructing a 3D shape model of an object by using multi-view image information, includes: inputting multi-view images obtained by photographing the object from multiple viewpoints in a voxel space, and extracting silhouette information and color information of the multi-view images; reconstructing visual hulls by silhouette intersection using the silhouette information; and approximating polygons of cross-sections of the visual hulls to a natural geometric shape of the object by using the color information. Further, the method includes expressing a 3D geometric shape of the object by connecting the approximated polygons to create a mesh structure; extracting color textures of a surface of the object by projecting meshes of the mesh structure to the multi-view image; and creating a 3D shape model by modeling natural shape information and surface color information of the object.

The invention discloses an RGBD camera-based three-dimensional human body rapid modeling system. The system is executed through the following steps of: establishing a scanning array by utilizing a plurality of RGB-D cameras; calibrating the scanning array by using an automatic calibration technology, so as to obtain a calibrated matrix and automatically guide the calibrated matrix into a scanning system; enabling the RGB-D cameras in the array to independently scan a target object after the scanning is started, so as to obtain point cloud data of corresponding poses; processing the acquired point cloud data by utilizing a point cloud optimization technology, so as to obtain high-quality point clouds which can be used for triangular meshing; realizing reconstruction from the point clouds to a triangular patch model by utilizing a rapid triangular meshing algorithm; and obtaining a colored three-dimensional human body model by utilizing a color mapping technology. According to the system disclosed by the invention, a technology which is used for constructing the scanning matrix on the basis of the plurality of RGB-D cameras, rapidly acquiring human body feature data and realizing the construction of the human body three-dimensional model with high precision and refined color texture is provided.

The invention discloses an object-level high-resolution remote sensing image change detection method based on multi-scale fusion. A color texture segmentation algorithm JSEG is introduced to change detection and multi-scale feature extraction and analysis on an object is realized based on a J-image image sequence generated in the segmentation process of the JSEG algorithm. The algorithm makes comprehensive use of the shape, the size and the spectral and texture features of the object, introduces two different multi-scale fusion strategies based on a D-S evidence theory and weighted data fusion and further improves divisibility of changing categories and non-changing categories so as to construct a complete set of object-level change detection frames based on multi-scale fusion. The feasibility and effectiveness of the method are verified by respectively carrying out experiment on a high-resolution aerial remote sensing DOM image and an SPOT5 satellite remote sensing image and comparing detection effects of change detection methods of different object levels and pixel levels.

The invention provides a color three-dimensional laser scanner and a three-dimensional color point cloud scanning method. A three-dimensional color point cloud scanning system of the invention uses the combined type technology of an ordinary light and a laser for scanning the outline structure and color texture of a space or object, so as to obtain and convert spatial coordinate information and color texture information of the space or object surface into a digital model containing both the coordinate information and the color information, and the digital model can be applicable to industries of machinery manufacture, home decoration design, construction and civil engineering, cultural heritage conservation, game development and the like and other related industries. The invention is designed for one time molding, requiring no post-processing, 360-degree panoramic modeling can be done within a minute, the whole process is automatic and quick, and the accuracy of measurement data is also ensured.

A graphical processing unit (GPU) and methods for rendering a three-dimensional (3D) scene generated in a field of view having in-focus and out-of-focus regions on a two-dimensional (2D) screen region of pixels are described. One method includes initially rendering the scene to create color and depth texture maps and creating mip-map layers for the color texture map. The method further comprises subsequently rendering the scene by, for each pixel: creating a mip-map layer selection value as a function of a depth of the pixel from the depth texture map, generating a color value by interpolation using color values from at least one of the mip-map layers chosen according to the mip-map layer selection value, and setting a color of the pixel to the generated color texture.

The invention provides a method for carrying out Bayes judgment on modeling for a video image based on color-texture dual characteristic vectors so as to realize moving target detection. The method comprises the following steps: 1. obtaining a video frame; 2. preprocessing an image; 3. background modeling and carrying out motion detection by utilizing Bayes judgment, wherein the background modeling comprises image color-texture dual characteristic vectors, respectively adopting different characteristic vectors according to the variation of interframe difference of a pixel point, expanding the original Bayes judgment to be two-dimensional due to the independence of the characteristic vectors, setting model parameters and updating control parameters so as to adapt to the change of the state of a moving object; 4. shadow detection, using a texture model to detect a shadow candidate point and eliminating the shadow; 5. processing after the detection, using horizontal projection and vertical projection to determine a foreground area and perfecting the detection result in a divided area through block analysis; and 6. updating a background image, respectively updating according to a detection classification result and setting self-adaptive updating rate parameters so as to adapt to illumination change.

A method of rendering a plurality of triangles into a color buffer defined by a plurality of pixel locations, utilizing a triangle identification buffer and a depth buffer. A relatively unique identifier is assigned to each of the triangles to be rendered. Before color and texture mapping, each triangle is depth compared on a per pixel basis. If a pixel of a current triangle is in front of any existing pixel at that point, the current triangles identifier is over-written into a triangle identification buffer. Color texture data is only retrieved for each triangle that appears in the identification buffer once all triangles have been compared.

The invention provides a free-moving type three-dimensional scanning method, belongs to the field of optical detection, and relates to a full three-dimensional skeleton surveying method, a handheld measuring device and a multi-angle splicing method based on texture, the curvature of surface and normal directions. Mobile handheld high-precision three-dimensional scanning is a problem of the scanning field, according to the free-moving type three-dimensional scanning method, a three-frequency color fringe projection method is utilized to project three-frequency color fringes of 1 : 4 :12 to the surface of a tested object through a projection device, and a camera is used for shooting a deformed fringe pattern in a single-frame exposure mode. On one hand, a three-frequency phase unwrapping method is adopted by the deformed fringe pattern to calculate out high-frequency carrier frequency to unwrap a phase, and system calibration parameters are used for obtaining a three-dimensional point cloud of the tested object; on the other hand, a texture pattern of the surface of the tested object can be recovered through the deformed fringe pattern. Handheld scanning has a higher requirement for multi-angle splicing, and by means of the free-moving type three-dimensional scanning method, non-mark splicing of multi-angle point clouds is achieved according to the color texture characteristics, measurement of the curvature of surface and restraint of the normal directions.

A background (sky box) in a video game or other virtual reality environment dynamically changes as a function of a parameter, producing a more realistic display. Luminance-only texture and color texture data are employed to create the background. The color texture data are provided as a lookup table in which one axis of the lookup table corresponds to time or another variable, and the other axis indicates the colors of pixels in the display as a function of the altitude or azimuth of the pixels. The luminance-only texture data are thus modulated by the chrominance values from the color texture data for pixels in the background. Optionally, one of a plurality of two-dimensional color textures can be selected as a function of the variable and applied to the luminance-only texture data.

The invention discloses a point cloud and image automatic high-precision registration method based on mutual information. The method comprises the steps of shooting a close-range optical image of a to-be-detected object; collecting ground laser scanning point cloud of the to-be-detected object, and generating an RGB color image and a reflection intensity image; calculating coarse registration parameters by using the close-range optical image and the point cloud RGB color image; extracting Harris operator feature points from the reflection intensity image by grids; a Forstner positioning operator is used for accurate positioning, rough registration parameters are used for calculating the initial positions of corresponding homonymous points of the feature points, the mutual information theory is used for searching for accurate matching point positions of the feature points in the close-range optical images in the initial positions, and accurate homonymous feature point pairs of the two images are obtained; calculating a fine registration parameter; and generating an accurate color point cloud model of the to-be-tested object. The method is not influenced by non-homologous data projection and radiation difference, and meets the requirement for reconstruction of a high-precision three-dimensional color texture model in the cultural relic protection industry.

Embodiments of the invention provide systems and methods for generating combined image data representing a volumetric image of a subject and a color textured external surface of the subject. According to one embodiment, two-dimensional image data of a surface of the subject can be obtained from a two-dimensional imager. Volumetric data of the subject can be obtained from a volumetric scanner. The two-dimensional imager can have a known optical property and physical position relative to the volumetric scanner. External surface mesh data can be generated based on the volumetric data. The two-dimensional image data can be mapped to the external surface mesh data based on the known physical position of the two-dimensional imager and the volumetric scanner. Display image data including the two-dimensional image data mapped to the external surface map data combined with the volumetric data can be generated and displayed.

The invention relates to a feature fusion living body detection method, and belongs to the field of image processing. The method deeply combines the advantages of the two types of images, simply and effectively improves the classification accuracy of a real person face and prosthesis faces such as a photo, a video, a three-dimensional mask and the like, and the robustness is stronger. According tothe invention, rich color texture information in the visible light image and reflection information in the near infrared image are fully utilized; an end-to-end deep learning method is adopted for automatic global learning, features of the two types of images are fused and classified, the advantages of the two types of images are combined, and the classification accuracy of the real human face and the prosthesis face is simply and effectively improved; attacks of photos, videos and three-dimensional masks can be identified at the same time, and the robustness of cross-data and attack types ishigh; and user interaction and cooperation are not needed, non-inductive living body detection is realized, and the user experience is good.

A method of forming at least one three dimensional (3D) color image of at least one object in a target space. The method comprises projecting, each of a plurality of projection cycles, a sequence comprising a plurality of gray coded light patterns, each colored in one of red green or blue, on a target space, capturing a plurality of two dimensional (2D) images of the target space during a plurality of acquisition cycles, each the acquisition cycle being timed to correspond with the projection of at least a sub sequence of the sequence, the sub sequence comprising red, green, and blue gray coded light patterns of the plurality of gray coded light patterns, extracting range data and color texture information of at lea one object in the target space from the plurality of 2D images, and forming a 3D color image of the range data and color texture information.

The present invention provides a wireless capsule endoscopy redundant image screening method based on multi-feature fusion. The method comprises the steps of extracting the color feature vector and the texture feature vector of an image through an HSV color feature histogram and a gray level co-occurrence matrix, a step of calculating the color feature normalized mutual information amount of an adjacent image and a texture feature mean square error value as similarity measurement, a step of considering the robustness of the method, a step of proposing the mean value method setting adaptive similarity determination threshold based on a W parameter, a step of dividing an WCE image with certain time correlation and color-texture similarity into a same sub image segment through comparing the similarity measurement value of the adjacent WCE image and a judgment threshold value orderly, and a step of using an adaptive K-means clustering algorithm to carry out key frame extraction on each sub image clip to achieve the purpose of filtering a redundant image. According to the method, the redundant image can be effectively filtered, and thus the efficiency of doctors is improved.

The invention discloses a license plate detection method based on a color texture analyzer and machine learning. The detection method comprises the following steps: step 1, carrying out texture feature extraction on a picture after preprocessing; step 2, carrying out transmission conversion on a candidate license plate graph after the texture feature extraction to correct a license plate inclination angel; step 3, determining an under color of the license plate, and carrying out color feature extraction on a blue or yellow license plate twice to lock a target position; step 4, carrying out secondary texture feature extraction on a white license plate; and step 5, carrying out machine learning determination on a candidate license plate after feature extraction. The detection method has advantages of high accuracy and broad application prospect and the like.

The invention discloses an image content retrieval system which is characterized by comprising 1) a segmentation subsystem, 2) an image area characteristic extraction unit and 3) a calculation and comparison subsystem, wherein the segmentation subsystem is used for respectively segmenting the training image and testing image into n examples and comprises A) an FCM segmentation unit and B) a pyramid segmentation unit, wherein the FCM segmentation unit is used for respectively performing area segmentation on the training image and the testing image by employing a fuzzy C-means and the pyramid segmentation unit is used for performing image segmentation on the training image and the testing image on the basis of a space pyramid matching mechanism by employing a cvPrySegmentation function; the image area characteristic extraction unit is used for extracting the color, texture, CENTRIST characteristics and each type of mean characteristic vectors of each example; and the calculation and comparison subsystem adopts a multi-example learning and diverse density sparse learning method, describes the image contents from the multiple nonlocal form characteristic angles and fulfills the retrieval aim. The method is easily realized through programming, has a certain artificial intelligence and provides algorithmic reference for the image understanding.

The invention discloses a hierarchical fast image global positioning system (GPS) position estimation method which includes: performing global feature clustering on images according to color texture global features for an on-line database, distributing the images obtained through clustering to second-stage small centers according to the shooting geological positions, and finally performing representative image selection on each clustering center. In a line system, a first-stage candidate class is selected according to the color texture features, the distance calculation is performed and a second candidate class is selected in the second-stage small centers contained in the first-stage candidate class according to the color texture features, and finally GPS position estimation is performed on eventually input images through a local feature confirmation method. By means of the hierarchical fast image GPS position estimation method, image retrieval speed can be effectively improved, and retrieval accuracy can be improved.

The invention belongs to the technical field of image processing, and relates to a pedestrian re-recognition method based on multi-channel consistency features. The method includes the following stepsthat N to-be-matched image pairs of training data and test data and labels ln corresponding to the image pairs are input, wherein n equals to 1,...,N; semantic feature representations of the input image data and color texture spatial distribution feature representations are extracted; consistency feature representations of the semantic feature representations and the color texture spatial distribution feature representations are obtained through multi-scale feature matching; a binary classifier is built for the obtained consistency feature representations, and a probabilistic representation for the same target is output. The method has the advantages that pedestrian recognition is carried out through comprehensive pedestrian image semantic attributes and color distribution features, and the method is high in precision, stable in performance and suitable for solving the problem of pedestrian re-recognition in a complex scene.

One embodiment of the invention provides a process and related apparatus for obtaining quantitative data about a 2-dimensional, 3-dimensional image, or other dimensional image. For example, the invention is capable of classifying and counting the number of entities an image contains. Each entity comprises an entity, structure, or some other type of identifiable portion of the image having definable characteristics. The entities located within an image may have a different shape, color, texture, or other definable characteristic, but still belong to the same classification. In other instances, entities comprising a similar color, and texture may be classified as one type while entities comprising a different color, and texture may be classified as another type. An image may contain multiple entities and each entity may belong to a different class. Thus, the system embodying the invention may quantify image data according to a set of changing criteria and derive one or more classifications for the entities in the image. Once the image data is classified, the total number of entities in the image is calculated and presented to the user. Put simply, embodiments of the invention provides a way for a computer to determine what kind of entities (e.g., entities) are in an image and counts the total number of entities that can be visually identified in the image. Another aspect of the invention is that the information utilized during a training process may be stored and applied across different images.

The invention provides a glass cover plate, a manufacturing method thereof and a mobile phone. The manufacturing method comprises the following steps: performing etching treatment on a first surface;and performing UV (Ultraviolet) transfer printing treatment on a second surface. According to the manufacturing method provided by the invention, the first surface of the glass cover plate is subjected to the process treatment of frosting, etching, polishing and the like in sequence, so that the structure of the first surface of the glass cover plate becomes a microstructure being similar to the structure of metal / wood / rock / cloth / rubber and the like. Then, the second surface of the glass cover plate is subjected to UV transfer printing, printing or ink spraying and the like to achieve the color texture three-dimensional effect of the metal / wood / rock / cloth / rubber, so that the selection range of materials of the glass cover plate is extended for a user through superposed feeling of touch andvision.

The invention discloses an improved image segmentation method based on picture and color texture features, and relates to the technical fields of image processing and computer vision. The method comprises the following steps: (1) extracting the texture features and color features of 30 sample pictures, constructing a matrix by using the obtained texture features and color features, and solving the covariance of the matrix for serving as a unified coefficient of color information and texture information of the 30 sample pictures; (2) constructing a weighted undirected graph of a picture to be segmented; (3) extracting the texture feature and color feature of the picture to be segmented, and constructing a texture color feature descriptor of the picture to be segmented; (4) calculating the weights of sides in the weighted undirected graph of the picture to be segmented by using the uniform coefficient obtained in the step (1) and the texture color feature descriptor of the picture to be segmented obtained in the step (3); (5) merging and classifying nodes in the weighted undirected graph constructed in the step (2). The method has the advantages of high segmentation accuracy and little wrong segmentation and over-segmentation phenomena.

The present invention relates to a texture map of paint colors in which a plurality of paint colors are arranged to be able to identify their textures; and its production method, production program, production system and data structure. The texture map of paint color production method of the present invention is a method of producing a paint color texture map in which a plurality of paint colors are arranged on a two-dimensional plane using coordinate axes of two parameters that represent the color and texture of the paint colors, the method comprising: determining at least three characteristic quantities, using a plurality of spectral reflectances measured at a plurality of light-receiving angles, for each of the plurality of paint colors (S1 to S3); subjecting a data group including the at least three characteristic quantities to principal component analysis so as to determine a First Principal Component and Second Principal Component expressed as linear expressions, using the characteristic quantities (S4); calculating values of the First Principal Component and Second Principal Component for the plurality of paint colors; and positioning the paint colors on a two-dimensional plane that has the First Principal Component and Second Principal Component as the two parameters and has values of the First Principal Component and Second Principal Component as coordinates (S5).

The invention relates to the three-dimensional texture information measurement technology in the field of three-dimensional face recognition and especially relates to a three-dimensional texture photography system, a calibration method and an imaging method. The invention proposes the concept of non-coupling three-dimensional texture information acquisition for the first time, multiple measuring units are used to obtain texture images from different visual angles in the invention, different shooting angles are determined by measured complete three-dimensional face surface triangular normals to shoot the weight of two-dimensional texture information to three-dimensional texture information construction, and the acquisition of three-dimensional face surface data and a three-dimensional color texture in no-coupling or low-coupling real space meaning can be realized in one time. A reliable face shooting method is provided for the three-dimensional face recognition technology.

The invention discloses a line structured optical three-dimensional measurement system and a three-dimensional texture image construction algorithm. The system comprises a CCD camera, a two-dimensional shifting fixed support, a linear laser device and a two-dimensional adjusting platform. The CCD camera is fixed to one end of the two-dimensional shifting fixed support; the linear laser device is fixed to the other end of the two-dimensional shifting fixed support; the two-dimensional adjusting platform is arranged on the two-dimensional shifting fixed support and is used for containing a measured object. According to the three-dimensional texture image construction algorithm based on the system, a new color texture mapping method is adopted to construct a distortionless three-dimensional outline curved surface formed by real image outlines. According to the core concept, binarized contour lines without any color are made to have color information (or grey information) through color texture mapping. According to the line structured optical three-dimensional measurement system and the three-dimensional texture image construction algorithm, a three-dimensional texture model with the real surface textures can be constructed.

The invention provides a binocular vision-based target tracking method for a hot-line working robot. A target area is described with color and texture fused features, a color and texture fused feature histogram is used as a feature histogram of a target model, the target area of a left camera image is tracked with a mean shift-based target tracking method, and the position and dimension information of a target in a current-frame image of a left camera are acquired; the position of the target in a right camera image is acquired with a template matching method, and coordinates of the target in a world coordinate system are calculated according to the binocular stereo vision three-dimensional measurement principle; by the aid of the dimension information of the target in the current-frame image of the left camera, the dimensions of the target in the left camera image are estimated with a dimension estimating method based on the image forming principle and moving constraints, and dimension correction is performed on a target area, required to be tracked, in a next-frame image of the left camera. With the method, the target identification rate of the hot-line working robot during working can be increased.