826 results about "Scale space" patented technology

An imaging acquisition system that generates a depth map from two pictures of a three dimensional spatial scene is described. According to one aspect of the invention, the system generates the depth map based on the relative blur between the two pictures and the absolute blur contributed by the system. According to another aspect of the invention, the system calculates the depth map directly from the relative blur between the two pictures.

A system and process for identifying corresponding points among multiple images of a scene is presented. This involves a multi-view matching framework based on a new class of invariant features. Features are located at Harris corners in scale-space and oriented using a blurred local gradient. This defines a similarity invariant frame in which to sample a feature descriptor. The descriptor actually formed is a bias/gain normalized patch of intensity values. Matching is achieved using a fast nearest neighbor procedure that uses indexing on low frequency Haar wavelet coefficients. A simple 6 parameter model for patch matching is employed, and the noise statistics are analyzed for correct and incorrect matches. This leads to a simple match verification procedure based on a per feature outlier distance.

An imaging acquisition system that generates a depth map for a picture of a three dimension spatial scene from the estimated blur radius of the picture is described. The system generates an all-in-focus reference picture of the three dimension spatial scene. The system uses the all-in-focus reference picture to generate a two-dimensional scale space representation. The system computes the picture depth map for a finite depth of field using the two-dimensional scale space representation.

Point cloud data sets representing overlapping physical areas are received. Surface feature points are extracted from the point cloud data sets and coordinates are determined for each point of the point cloud data sets. The point cloud data sets are aligned based on matching surface feature points between the point cloud data sets. Matched surface feature points are moved toward the coordinates of corresponding matched surface feature points and the remaining points are moved based on the moved matched surface feature points and the determined coordinates for each point.

A system and method for automatically generating a hierarchical table of contents or outline for indexing a document and identifying clusters of related information in the document. The document may comprise text, audio, video, or a multimedia presentation. The invention employs a unique and novel combination of latent semantic indexing techniques to identify related blocks and major topic changes within the document with scale space segmentation techniques to respectively identify self-similar blocks within the document and to thus find topic changes of various sizes at block edges. The invention then produces a visual presentation of the semantic structure of the document.

A method and system is provided for finding stable keypoints in a picture image using localized scale properties. An integral image of an input image is calculated. Then a scale space pyramid layer representation of the input image is constructed at mulitple scales, wherein at each scale, a set of specific filters are applied to the input image to produce an approximation of at least a portion of the input image. Outputs from filters are combined together to form a single function of scale and space. Stable keypoint locations are identified in each scale at pixel locations at which the single function attains a local peak value. The stable keypoint locations which have been identified are then stored in a memory storage.

A system and method for identifying critical features in an ordered scale space within a multi-dimensional feature space is described. Features are extracted from a plurality of data collections. Each data collection is characterized by a collection of features semantically-related by a grammar. Each feature is normalized and frequencies of occurrence and co-occurrences for the feature for each of the data collections is determined. The occurrence frequencies and the co-occurrence frequencies for each of the features are mapped into a set of patterns of occurrence frequencies and a set of patterns of co-occurrence frequencies. The pattern for each data collection is selected and distance (similarity) measures between each occurrence frequency in the selected pattern is calculated. The occurrence frequencies are projected onto a one-dimensional document signal in order of relative decreasing similarity using the similarity measures. Wavelet and scaling coefficients are derived from the one-dimensional document signal using multiresolution analysis.

The invention discloses a behavior recognition method based on sparse spatial-temporal characteristics. The method comprises the steps as follows: step 1, convolving an input video with an original input video by using space-time Gabor to establish a scale space; step 2, using the expressions of different scales as values of different channels of a space-time depth belief network and associatively learning characteristics of multi-scale; step 3, recognizing and classifying the behavior characteristics. The behavior recognition method of the invention inputs depth network to associatively learn the characteristics of multi-scale by establishing the scale space to improve performance of behavior recognition. The behavior recognition method of the invention introduces the thought of spatial pyramid in terms of information loss problem of a pooling operation and performs multilevel expansion to pooling output, and combines with sparse coding to fuse pyramid multilevel characteristic, thereby improving characteristic dimension output by a pooling layer, further improving the performance of the original network and improving behavior recognition rate.

The invention provides a feature extraction and matching method and device for a digital image based on PCA (principal component analysis), belonging to the technical field of image analysis. The method comprises the following steps of: 1) detecting scale space extreme points; 2) locating the extreme points; 3) distributing directions of the extreme points; 4) reducing dimension of PCA and generating image feature descriptors; and 5) judging similarity measurement and feature matching. The device mainly comprises a numerical value preprocessing module, a feature point extraction module and a feature point matching module. Compared with the existing SIFI (Scale Invariant Feature Transform) feature extraction and matching algorithm, the feature extraction and matching method has higher accuracy and matching speed. The method and device provided by the invention can be directly applied to such machine vision fields as digital image retrieval based on contents, digital video retrieval based on contents, digital image fusion and super-resolution image reconstruction.

The invention discloses a target automatically recognizing and tracking method based on affine invariant points and optical flow calculation, which comprises the following steps: firstly, carrying out image pretreatment on a target image and video frames and extracting affine invariant feature points; then, carrying out feature point matching, eliminating mismatching points; determining the target recognition success when the feature point matching pairs reach certain number and affine conversion matrixes can be generated; then, utilizing the affine invariant points collected in the former step for feature optical flow calculation to realize the real-time target tracking; and immediately returning to the first step for carrying out the target recognition again if the tracking of middle targets fails. The feature point operator used by the invention belongs to an image local feature description operator which is based on the metric space and maintains the unchanged image zooming and rotation or even affine conversion. In addition, the adopted optical flow calculation method has the advantages of small calculation amount and high accuracy, and can realize the real-time tracking. The invention is widely applied to the fields of video monitoring, image searching, computer aided driving systems, robots and the like.

The invention discloses a method for remote sensing monitoring and early warning fire in sylvosteppe, belonging to the field of remote sensing technology application. The invention makes full use of the information expressive force of satellite remote sensing data on a large-scale space, comprises fire risk evaluation on the sylvosteppe ecological system in a large-scale range and detection of abnormal high-temperature points in the sylvosteppe on the basis of the remote sensing data, and generates the fire remote sensing monitoring and early warning information for the sylvosteppe in a point-sphere integrated manner by assimilating and processing the detection data (point) on the abnormal high-temperature points and data of fire risk evaluation (sphere), thus providing decisive instruction information for fire protection and fire emergencies of the sylvosteppe. The invention is characterized by comprising four parts of a data pre-treatment unit, an abnormal high-temperature point detection unit, a fire risk evaluation unit and an early warning information processing unit.

A scale-Space feature extraction technique is based on recursive decomposition of polyhedral surfaces into surface patches. The experimental results show that this technique can be used to perform matching based on local model structure. Scale-space techniques can be parameterized to generate decompositions that correspond to manufacturing, assembly or surface features relevant to mechanical design. One application of these techniques is to support matching and content-based retrieval of solid models. Scale-space technique can extract features that are invariant with respect to the global structure of the model as well as small perturbations that 3D laser scanning may introduce. A new distance function defined on triangles instead of points is introduced. This technique offers a new way to control the feature decomposition process, which results in extraction of features that are more meaningful from an engineering viewpoint. The technique is computationally practical for use in indexing large models.

The invention discloses an efficient image retrieval method based on an improved SIFT (scale invariant feature transform) feature, relates to the field of image processing and computer vision and belongs to a content-based image retrieval method. The method comprises the steps of establishing a gauss difference scale space, taking scale space detection extreme points as feature points, calculating the direction of the feature points and gradient module values in the direction, generating a global feature vector of the image feature points through a geographical statistical method, conducting global feature vector similarity matching to finish rough retrieval, generating feature descriptors on the basis of rough retrieval, and conducting feature descriptor matching to finish accurate retrieval. The novel image retrieval method is superior to a traditional SIFT algorithm, and is more suitable for requirements of image infringement examination in digital prints than an existing retrieval algorithm.

The invention relates to a license number matching algorithm based on digital image processing. The license number matching algorithm based on the digital image processing comprises the following steps of: twice shooting a same license through shooting equipment to obtain two images, and carrying out scale space extreme value detection, accurate characteristic point position positioning, characteristic point principal direction determination and characteristic vector generation on the two images to obtain image characteristics; and then matching the characteristic vectors of the two images. The license number matching algorithm based on the digital image processing, which is disclosed by the invention, can still complete the matching and keep good stability and robustness under the conditions of scaling, rotation, brightness variation, affine transformation and noise influence through a particular experiment verification algorithm and is suitable for carrying out fast and accurate matching in a great number of characteristic databases.

The invention discloses a remote sensing image registration method based on anisotropic gradient dimension space, which mainly solves the problem of relatively low correct matching rate under the condition of relatively great brightness nonlinear change of the remote sensing images. The implementing steps of the remote sensing image registration method based on anisotropic gradient dimension space are as follows: (1) inputting remote sensing image pairs; (2) constructing dimension space of anisotropic diffusion; (3) calculating a gradient amplitude image; (4) detecting feature points; (5) generating a main direction of the feature points; (6) generating a descriptor of each feature point; (7) matching the feature points; (8) deleting wrongly matched feature point pairs; and (9) registering a reference image and a to-be-registered image. As feature point detection, feature point main direction generation and feature point descriptor generation are carried out on the gradient amplitude image in the anisotropic dimension space, the situation of relatively great brightness nonlinear change of the images can be dealt efficiently, and the remote sensing image registration method based on anisotropic gradient dimension space can be applied to complex multisource and multispectral remote sensing image registration.

Method, system and computer program for efficiently extracting and describing scale-space interest points. It is designed towards low overall computational complexity. On one hand, the data acquired during extraction in the description phase is intensively re-used. On the other hand, an algorithmic optimization of the description that dramatically speeds up the process, is proposed. First, the image is filtered with triangle kernel at different scales. The triangle filtered images are reused for extraction of the keypoints dominant orientation and the computation of the DAISY-like descriptor

The invention discloses an improved scale invariant feature transform (SIFT) image feature matching algorithm. The algorithm comprises: step one, scale space extreme points are detected; step two, feature descriptor is generated; and step three, a K-d tree balanced binary tree is built, a nearest neighborhood feature point on the K-b tree is searched by BBF, a matched feature dot pair is judged by Euclidean distance, and secondary matching is conducted after Euclidean distance matching. According to the improved SIFT image feature matching algorithm, the descriptor of 128 dimensions is reduced to 48 dimensions, execution efficiency of the algorithm is improved by two-thirds and reaches the speed of speeded-up robust features (SURF) feature description subalgorithm based on integral, and the defects that the algorithm is not suitable for the gray level and the changing circumstances of the point view of the images are overcome.

A method for extracting and matching gesture features of image is disclosed. An input gesture image is captured, and then a closed curve formed by a binary contour image of the gesture image is determined by preprocessing the gesture image. A curvature scale space (CSS) image of the gesture image is drawn based on the closed curve. Feature parameters of a plurality of sets of the gesture image are determined by extracting first plural peaks from the CSS image as basis points, and each feature parameter of the plurality of sets of the gesture image is compared with each feature parameter of a plurality of reference gesture shapes represented as a basis point of the maximal peak, thereby determining a gesture shape corresponding to the gesture image.

The invention discloses a fast pedestrian detection method based on aggregated channel features, which comprises an early-stage position calibration process and a later-stage position screening process. During the early-stage position calibration process, multiple pieces of channel feature information in an input video or an image are aggregated, an image pyramid of the input image or the video in scale space is built, an image for each channel feature is calculated for each scale, features as pedestrian existing judgment basis are recognized, and the pedestrian position is initially extracted. During the later-stage position screening process, a convolutional neural network classifier is used for further screening each position calibrated in the early stage, pedestrians appearing in the image or the video are further detected, and a detection result is obtained. By adopting the technical scheme of the invention, in the case of a large training data amount, the classifier can automatically select features with a good recognition ability, the features can serve as pedestrian judgment basis, the method has good robustness, and the pedestrian detection precision is improved.

The invention discloses a time phase unwrapping method based on double frequency three-gray-scale sinusoidal grating fringe projection. The method comprises the steps that two sets and five ideal double-frequency sinusoidal grating fringes are generated by utilizing a computer, then three-gray-scale sinusoidal grating fringes corresponding to the ideal double-frequency sinusoidal grating fringes are generated by adopting the three-gray-scale space pulse width modulation technology, under condition that that a projector keeps the defocusing state, the five three-gray-scale sinusoidal grating fringes are projected to the surface of an object to be measured, two sets of wrapped phase images are solved, wherein the five double-frequency fringes comprise the two sets of wrapped phase images, the fringe order corresponding to the wrapped phase is solved, and unwrapping of the wrapped phase is achieved. The time phase unwrapping method based on the double frequency three-gray-scale sinusoidal grating fringe projection improves measurement precision and measurement speed.

The invention discloses a combined type device and a method for precisely and dynamically measuring spatial position and posture. The device comprises an inertia measuring unit and a total station measuring unit which are connected with a data acquisition and processing unit respectively; and the data acquisition and processing unit is connected with a computer. The inertia measuring unit measures the acceleration and the angular velocity of a target solid conjugant; a total station measures and calculates the position and the posture of the target solid conjugant; the acceleration, the angular velocity and the position and the posture are sent to the computer through the data acquisition and processing unit; according to data arrival time, measurement time delay and an inertia measurement solution principle of the total station, the computer adopts a time backtracking algorithm and KALMAN wave filtering, merges and processes two kinds of data, removes the measurement time delay of the total station and resolves out the optimally-estimated position and the posture of the target solid conjugant. The combined type device and the method improves the stability and the real-time position and posture measuring precision of the prior total station measuring system, enriches the measuring output information of the prior dynamic tracking measurement system and meets the real-time measuring requirement of high precision, high sampling rate and long-time operation of a large-scale device on large-dimension space.

The invention designs a deformable convolution hybrid task cascade semantic segmentation method based on embedded balance, which is used for realizing image target recognition and semantic segmentation, and comprises the following steps: inputting a cut image into a pre-trained neural network; mapping the two samples to the same scale space through a feature pyramid network; performing informationfusion on semantic features extracted from different hierarchies; predicting a pixel-level segmentation result by adopting a convolution layer; performing feature extraction on the input image by adopting a deformable convolutional neural network at the convolution and pooling part of the feature pyramid network to obtain a feature map; dividing the feature map into parts with the same size; inputting a feature map obtained after passing through the feature pyramid network into a regional candidate network for training the network; wherein the region candidate network comprises a target detection classifier and a candidate frame positioning classifier, the target detection classifier outputs a target recognition result and prediction accuracy, and the candidate frame positioning classifier can provide accurate positioning for candidate regions and output candidate frames of a plurality of candidate regions. According to the method, the semantic segmentation positioning accuracy and the segmentation accuracy are improved.