139results about How to "Rotation invariant" patented technology

The present invention provides a dynamic model directed angiography three-dimensional reconstruction method, pertaining to an intersectional field of digital image processing and medicine imaging, for satisfying special requests of auxiliary detection and surgery guidance for cardiovascular disease in clinical medicine. The invention includes steps of angiography image preprocessing, vascellum segmentation, vascellum skeleton and radii extraction, model directed vascellum base element recognition, vascellum matching and vascellum three-dimensional reconstruction. The invention also provides a cardiovascular dynamic model construction method including steps of cardiovascular slice data extraction, cordis static and dynamic model building, and static and dynamic model building. According to the invention, good angiography three-dimensional reconstruction result may be obtained, which effectively assists detection and surgery guidance for cardiovascular disease, thereby satisfying clinical requests.

The invention provides an electrocardiosignal R peak detection method based on waveform characteristic matching. The method utilizes waveform characteristic matching to identify R peak of electrocardiosignal, the characteristic matching method takes difference vector between points as basic characteristic, the basic characteristic has translational invariance and rotational invariance and can overcome influence of baseline drift of electrocardiosignal signal; meanwhile, logarithm polar coordinate transformation is carried out on the difference vector and partition is carried out to measure similarity of wave form, the measurement is sensitive to the adjacent morphological characteristic, can capture global outline information of wave form and has robustness to wave form ripple; besides, influence of interference signal can be eliminated by setting proper threshold, and further accurate identification and detection on R peak of electrocardiosignal are realized. The invention is applied to related electrocardiogram analyser, accurate identification on R peak of electrocardiosignal can be realized, thus being beneficial to improving detection and analysis capability of electrocardiogram analysis equipment.

The invention relates to a tooth X-ray image matching method based on SURF point matching and RANSAC model estimation. The tooth X-ray image matching method includes the steps that (1), images are collected; (2), the images are enhanced; (3), features are extracted; (4), the features are matched; (5), related matching coefficients are calculated; (6), a matching result is displayed; (7), if reference images which do not participate in matching exist in an image library, the step (1) is carried out, a new reference image is selected from the image library, and if all images participate in matching, operation is finished; (8), matching succeeds, personal information corresponding to the reference images is recorded, and then operation is finished. The feature points of an input tooth image and a reference tooth image are extracted according to an SUFR algorithm, then the feature points are matched according to an RANSAC algorithm, and finally, the matching degree of the two images is measured according to the correlation coefficient of the two matched images. Experiment shows that by the tooth X-ray image matching method, the high precision ratio and high real-time performance are achieved.

The invention discloses a pose measurement method combining initial pose measurement with target tracking, and belongs to the field of computer vision. The invention aims to realize real-time output of pose parameters of a moving target in a video sequence for an irregular moving target. The method is technically characterized by comprising the steps that initial pose measurement of a target is conducted, the initial pose measurement of the target is used for starting follow-up target tracking, and real-time output of pose parameters is achieved. The method comprises the steps of reference pose measurement, feature point extraction and matching and pose measurement based on multi-point matching. Target tracking comprises searching and matching of edge features, tracking and matching of point features and nonlinear optimization solution of pose parameters. According to the method, the initial pose of the target is combined with a target tracking method, so that real-time output of the pose parameters of the moving target in a video sequence is realized; the method has a certain adaptive capacity to the change of the environment, and the pose of the object can be measured when the object is partially shielded.

The invention belongs to the field of image processing and machine vision, in particular relates to a target tracking method based on the template matching. The template image and pixels with a matching area are arrayed into a plurality of sub windows according to the circle; and the target is guaranteed to have the translation and the rotational invariance by adopting the circular template matching principle. The edge strength value of the template and each pixel in the tracking window can be calculated by using a Kirsch operator. The summation of the gray matching valve and the edge strength matching value is used as the matching result. The position of the optimal matching value is determined as the position of the tracking target. The invention can be used for a target identifying and tracking system.

The invention relates to a matching method for three-dimensional coronary angiography reconstruction, which belongs to the crossing field of digital image processing and medical imaging, and aims to meet the special requirements on the auxiliary detection and the surgical navigation of angiocardiopathy in clinical medicine. The method comprises two steps, i.e. model guidance contrastographic picture vessel segment matching and contrastographic picture puncta vasculosa matching. The matching method uses a multi-scale vascular tree model so as to better guide vascular matching, simultaneously can well solve the problem of individual difference, and improves the matching precision of vessel segments and puncta vasculosa through the method of iterative matching and largest declination of searching potential energy. The matching method for three-dimensional coronary angiography reconstruction can obtain very good angiography matching results, thereby solving the difficult problems in the automatic three-dimensional reconstruction by the multi-view angiography map.

The invention provides a detection method for a fatigue state of a driver. Confidence coefficients are acquired through calculation by sending eye SIFT features into a classifying model; the opening-closing states of the eyes are determined according to the confidence coefficients without being affected by a human face angle; and the detection method has a rotation invariable property, high accuracy on eye detection, simple calculation process, and can detect the fatigue state of the driver in time and can satisfy real-time detection requirement. In addition, the invention also provides a detection system for the fatigue state of the driver.

The invention provides a circular center detection method based on Gaussian fitting sub-pixel edge detection and an improved RANSAC algorithm, belonging to the field of aircraft ground simulation. Themethod comprises the follow concrete steps: an original image is matched by a model and RIO processing is conducted on the image; edge pixel extraction and thinning is carried out; sub-pixel edge location based on Gaussian fitting is carried out; an improved RANSAC circle fitting algorithm is provided; the coordinates of the center of a circle are calculated. The invention roughly locates the position of the circular marker through a model matching mode, and performs RIO processing to reduce the size of image processing. Gaussian fitting improves the accuracy and stability of the algorithm. The improved RACNAC algorithm reduces the sampling times, optimizes and screens the candidate circles, reduces the running time and improves the accuracy and stability.

The invention discloses a different-source image fusion method based on salient target region extraction and NSST (Non-subsampled Shearlet Transform). The method comprises the following steps that: firstly, extracting the salient target region of an infrared image; carrying out registration on infrared and visible images to obtain an affine transform matrix; through the affine transform matrix, obtaining the position of the salient target region of the infrared image in the visible light image; carrying out NSST on the infrared and visible light image; utilizing different methods to carry outfusion on the NSST results of a target area and a non-target area of two images; and finally, through NSST inverse transform, obtaining a final fusion result. By use of the method, two different-source images of the infrared light and the visible light can be effectively fused, a fusion result can better highlight and keep the target, and the background detail information of the non-target area inthe visible light image can be more favorably kept.

The invention discloses a detection identification method for a three-X combined mark, comprising the following three steps: 1. manufacturing the three-X combined mark, and combining three standard X angular points into the B type and W type of combined marks according to certain constraint rules; 2. detecting the X angular points, carrying out symmetry calibration by taking a corresponding Hessian matrix feature vector direction as the direction of the symmetry axis of the X angular points on the basis of a Hessian matrix probing X angular point algorithm; and 3. in the detected sequence of the X angular points with direction, detecting and indentifying the three-X combined mark. Two types of three-X combined marks defined by the detection identification method have the advantages of rotation invariance and distortion resistance; and by utilizing the detection identification algorithm, different three-X combined marks can be detected in a distinguishing mode, and the difficulty of subsequent vision matching and identification is lowered.

The invention relates to a three-dimensional attitude estimation method in disordered grabbing based on SHOT and ICP algorithms. The method is used for object pose estimation in the disordered grabbing process. A good initial position is provided for ICP fine registration through an SHOT coarse matching method, an M matrix in an SHOT algorithm is constructed, eigenvalue decomposition is carried out, and then a local reference system is constructed; topological characteristics under a corresponding reference system are calculated, a result is stored in a histogram, Hough transform is performedon a characteristic extraction result corresponding to the model and the scene point cloud, wrong point pairs are eliminated, and resolving is performed to obtain a final coarse registration attitudeestimation result; and nearest point searching is carried out by using an ICP algorithm, a covariance matrix construction is constructed, and a rotation matrix and a translation vector for minimizingan objective function are calculated. The problem that the registration precision is reduced when the traditional ICP is applied to the collection of point clouds with holes due to the reflection of workpieces is solved.

The invention relates to a method for tracking a point feature based on fractional-order differentiation, comprising detecting a point feature by adopting a method based on the fractional-order differentiation; forecasting the location of the next frame point by using a Kalman method or an extension method; searching in a given area, carrying out a similarity measurement, and acquiring a corresponding tracking point if requirements are satisfied; otherwise, considering the corresponding tracking point to be absent, and for such a point, considering the tracking is lost if a corresponding matched tracking point is still absent in a range of following k frames, wherein the value of k is larger than 2; and updating the point feature if the tracking is normal. The fractional-order differentiation has advantages over integral-order differentiation in presenting areas which have abundant texture details and inconspicuous texture information. Different differential gradient images are formed for fractional-order differentiation with different directions and different orders, and convolution directional diagrams with different scales are formed by combining the different differential gradient images with Gaussian kernel convolution in different sizes respectively, so that significant changes presented by the point feature are ensured when the direction is changed, having properties of rotation invariance, translation and scale invariance.

The invention discloses a method for extracting a suspected flame region based on infrared light and visible light images, and relates to the field of computer vision. In this method, brightness threshold segmentation is performed on a collected infrared image to obtain a suspected high-temperature flame region; at the same time, the infrared image and a visible light image are image-registered toobtain an affine transformation matrix, and the position of the suspected high-temperature flame region on the visible light image is obtained by combining the suspected high-temperature flame regiontherewith; and finally, a suspected flame region is obtained by flame segmentation based on the YCrCb color space, morphological processing, and connected component labeling. The method can effectively extract the more accurate suspected flame region, can guarantee the detection of the suspected flame region in the night environment at the same time, and is conducive to the all-weather flame detection.

The invention relates to the technical field of processing of images, in particular to a real-time circular printing image defect detecting method. The method comprises the following steps: a standard template graph and an on-line real-time graph of a target to be detected, obtaining a circle point coordinate through an exponential changing and searching step-length method, calculating a circle center and semi-diameter of a circular area of each of the standard template graph and the on-line real-time graph according to a method for positioning a circle through three points, dividing each circular area into a large quantity of annular areas, counting characteristics of grey level histograms of the annular areas, and comparing the characteristics of the grey level histograms of the annular areas of the standard template graph and the on-line real-time graph to complete image defect detection. Through the adoption of the real-time circular printing image defect detecting method, the circular printing image defection speed is improved from two aspects; a practicable method is provided for on-line defect detection based on machine vision.

The invention relates to a coin image recognition method based on sparse representation. The traditional image recognition method has poor recognition effect. The coin image recognition method comprises the steps of: firstly, obtaining gray level images of the front face and the back face of a coin for preprocessing and determining a coin region, standardizing the size of a coin image; expanding a training sample image into a plurality of angle samples and selecting a main training sample, expressing a testing coin sample into sparse representation of a training sample set, solving an optimal sparse coefficient vector; reconstructing the testing coin image, determining the coin image category by using a reconstructing error; and identifying the coin by utilizing the statistical property of the sparse coefficient vector. The invention can recognize the coin image without extracting the surface characteristics of a coin with rotational invariance, and has simple implementation method.

The invention provides a GPU-based automatic generation and collision detection method for a soft tissue organ metaball model. The method comprises four steps: a metaball model generation step: performing point sampling of an original mesh, calculating a Voronoi diagram of a mesh model, acquiring a medial surface of a triangular mesh model, and placing an initial ball on the medial surface; a step for local optimization of the metaball model: according to the initial position and radius of the metaball model, utilizing the method for adjusting the radius, the method of filling and other methods to perform local optimization of the metaball model; a step for global optimization of the metaball model: utilizing a charge gravity model, and calculating the gravity between the metaball and the gap to move and adjust the position of the metaball; and a collision detection step: utilized the generated metaball model to perform collision among soft tissue organs and collision between soft tissues and surgical instruments. The invention provides a soft tissue modeling method and a collision detection method for a virtual surgery. The soft tissue modeling method and the collision detection method for a virtual surgery can use a GPU parallel computation to realize acceleration and are high in real-time.

The invention discloses a footprint pressure distribution characteristic representation method which comprises the steps of (S1) using a footprint sampling instrument to collect footprint pressure image information, and preprocessing a footprint pressure image, (S2) extracting footprint front foot and back foot pressure distribution characteristics and constructing a footprint pressure distribution characteristic map, (S3) projecting front foot and back foot pressure distribution characteristic maps along direction, finding a direction with a maximum projection value (pressure value) as a main direction, and carrying out pressure distribution characteristic map main direction standardization, and (S4) carrying out characteristic selection on the obtained footprint pressure distribution characteristics. According to the method, the extracted footprint pressure distribution characteristics can represent the distribution characteristics of foot pressure when a human body is walking, the distribution between a footprint of barefoot and a footprint with socks is not needed, and the applicability is wide.

The invention discloses an improved Online Boosting and Kalman filter improvement-based TLD tracking method, and belongs to the technical field of machine vision, artificial intelligence, man-machineinteraction and target tracking. The method comprises the following steps of: (1) initialization: initializing an improved Online Boosting classifier and a P-N learning device by utilizing an initialsample set formed through selecting a target and carrying out affine transformation; (2) image tracking: selecting a feature point, tracking the feature point for twice by using an L-K optical flow method, and comparing an error between the twice tracking with a threshold value so as to obtain a tracking result; (3) image detection: obtaining a detector result through a Kalman filter, a variance classifier, the Online Boosting classifier and a KNN classifier; (4) tracking result and detection result integration: assessing confidence coefficients of the tracker result and the detector result soas to determine which module result is finally adopted; and (5) online learning: correcting the tracker result and the detector result by using the P-N learning device, and enriching the sample set.The method is capable of effectively overcoming the shielding problems, improving the speed of original methods and effectively the precision and robustness of detectors.

The invention relates to a non-contact video displacement measuring method. The method comprises the following steps that 1, motion videos of a digital equipment photographing structure are adopted and guided into a processor; 2, the processor obtains a first frame image in the motion videos, a target point is adopted as a center selection target area; 3, the next frame image is obtained, a target point in the previous image is adopted as a center selection matching area, pixel points in the matching area obtain matching blocks equal to the size of the target area by adopting the pixel point position as the center; 4, the target area and the matching blocks are matched to determine a matching block matched with the target area, the center point of the matching blocks is the position of the target point in the frame image, the position is adopted as a novel target point, and the matching blocks are adopted as the novel target area; 5, the steps 3 and 4 are repeatedly executed to obtain the position where the target point in each frame image is located, and a displacement curve of the target point is obtained. Compared with the prior art, measuring equipment is cheap, and the measuring result is reliable.

The invention belongs to the field of nuclear magnetic resonance imaging data processing, and relates to a movie nuclear magnetic resonance image sequence motion field estimation method based on a fractional order differential. The method aims to solve the problems that the method of enhancing and directly building a light stream equation through integer order differential images in the prior art is used for estimating a movie nuclear magnetic resonance image sequence motion field, grain details of the images are lost, an estimation result is influenced by illumination changes, the rotation unchanging performance does not exist, the noise resisting performance is poor, and precision based on movie nuclear magnetic resonance image sequence motion field estimation is low. The method mainly comprises the steps that firstly, grain enhancing is carried out on a movie nuclear magnetic resonance image through the fractional order differential; secondly, monogenic signals of the image are extracted through Riesz transformation, namely, the monogenic phase, the monogenic direction and the monogenic amplitude; thirdly, the light stream equation is built through the phase vector of the monogenic signals; fourthly, the movie nuclear magnetic resonance image sequence motion field is estimated through the light stream equation. The method is used for estimating the motion of an imaging object by the movie nuclear magnetic resonance image.