603 results about "Corner detection" patented technology

A fully automated package identification and measuring system, in which an omni-directional holographic scanning tunnel is used to read bar codes on packages entering the tunnel, while a package dimensioning subsystem is used to capture information about the package prior to entry into the tunnel. Mathematical models are created on a real-time basis for the geometry of the package and the position of the laser scanning beam used to read the bar code symbol thereon. The mathematical models are analyzed to determine if collected and queued package identification data is spatially and / or temporally correlated with package measurement data using vector-based ray-tracing methods, homogeneous transformations, and object-oriented decision logic so as to enable simultaneous tracking of multiple packages being transported through the scanning tunnel.

The invention discloses a PTAM improvement method based on ground characteristics of an intelligent robot. The PTAM improvement method based on ground characteristics of the intelligent robot comprises the steps that firstly, parameter correction is completed, wherein parameter correction includes parameter definition and camera correction; secondly, current environment texture information is obtained by means of a camera, a four-layer Gausses image pyramid is constructed, the characteristic information in a current image is extracted by means of the FAST corner detection algorithm, data relevance between corner characteristics is established, and then a pose estimation model is obtained; two key frames are obtained so as to erect the camera on the mobile robot at the initial map drawing stage; the mobile robot begins to move in the initializing process, corner information in the current scene is captured through the camera and association is established at the same time; after a three-dimensional sparse map is initialized, the key frames are updated, the sub-pixel precision mapping relation between characteristic points is established by means of an extreme line searching and block matching method, and accurate re-positioning of the camera is achieved based on the pose estimation model; finally, matched points are projected in the space, so that a three-dimensional map for the current overall environment is established.

The invention discloses an indoor robot SLAM method. The method comprises the following steps: obtaining image data through a camera, wherein the image data comprises RGB images and depth images; using a corner detection algorithm, an LK characteristic point tracing algorithm, and a RANSAC algorithm to process the RGB images and depth images to adjust the position and angle of the camera so as to obtain the RGB image data information under a robot operation system; converting the depth images in a world coordinate system to a ground coordinate, traversing the depth images that are projected on the ground coordinate, setting the gray value of the grid where a barrier stays as a first characteristic value, carrying out traversing to obtain a 2D barrier grid map; through a single line laser radar scanning mode, searching the 2D barrier grid map, when the grid with a grey value equal to the first characteristic value is found, feeding back the distance between the grid and the camera to obtain the distance between a robot and the barrier, obtaining the depth images under a robot operation system, and obtaining an environment 2D map.

The invention discloses a parameter calibration method. The method comprises steps of acquiring a calibration template image which is obtained by photographing of a calibration template image; performing corner detection for the calibration template image so as to extract an image corner; calculating radial distortion parameters in accordance with the extracted image corner; performing radial distortion correction in accordance with the calculated radial distortion parameters so as to reconstruct a distortion correction image; and according to a perspective projection relation of the calibration template and the reconstructed distortion correction image, calculating internal parameters and external parameters so as to achieve parameter calibration, wherein internal parameters and external parameters comprise internal parameter matrixes, rotating vectors and translation vectors. The invention also provides a parameter calibration device. The method and the device can be applied to parameter calibration of imaging devices such as video cameras or cameras under the condition of high distortion, the operation is simple and the precision is high.

The invention discloses an automatically calibration method based on black and white grid corner matching. The method includes adopting black and white calibrating cloth as a calibrating model, respectively detecting image corners by an improved harris corner detection calculation, building position relations between corners and forming a corner network; performing distortion correction and matching to images; calculating corner position in non-distorting images and homographic matrix of corner real position in final virtual full view bird's-eye view; and performing view changing according to virtual bird's-eye view camera position and homographic matrix, calculating corresponding relation between image points in the coordinate system and the original images, and then forming a lookup table. The automatically calibration method adopting the improved harris corner detection calculation is simple and practical, can well judge corner position relation, and forms a correct corner network; and the automatically calibration method adopting lookup table function can generate output images after the images are calibrated, is fast, and can well meet real-time need.

The invention relates to a three-dimensional human face reconstruction method based on stereoscopic vision of two eyes. Steps for implementing the method are as follows: (1) using two cameras to respectively take a picture of a human face from two different angles; (2) respectively determining an internal parameter matrix and an external parameter matrix of each camera; (3) carrying out extreme line correction and picture conversion according to the determined data; (4) utilizing a Harris corner detection operator to extract characteristics of the human face, and carrying out initial matching by utilizing a local template window method and extreme line restriction; (5) starting from an initial matching assemblage, and obtaining a sparse matching assemblage by utilizing a seed point growing algorithm; (6) taking the sparse matching assemblage as a guiding point and carrying out a dynamic planning algorithm so as to finish dense matching; and (7) calculating three-dimensional coordinates of actual physical points on the human face according to the determined data and the matching relationship, thereby reconstructing a three-dimensional point cloud of the human face.

The invention discloses an obstacle detection and road surface segmentation algorithm based on a three-dimensional laser radar, and the algorithm comprises the steps: (1), scanning the surrounding environment through the three-dimensional laser radar to obtain the point cloud information of the surrounding environment, and transforming the point cloud information to a local right-angle coordinatesystem from the coordinate system of the laser radar; (2), extracting an interest data point of the three-dimensional laser radar; (3), extracting a laser radar scanning single line through a radar detection angle clustering method; (4), segmenting the laser radar scanning single line through neighborhood fuzzy clustering based on AIC criterion; (5), accurately locating a road edge and a road surface line end point through corner detection. Compared with the prior art, the method can achieve the real-time and effective extraction of a passable region of a road surface, is high in precision andreliability, is small in judgment error in a recognition process, and can be widely used for an actual occasion of the extraction of the passable region of a structured road based on the three-dimensional laser radar.

Disclosed is a parameter calibration method of a vehicle-mounted camera. The method includes steps of self-making a three-dimensional calibration board, and establishing a world coordinate system on the basis of the self-made three-dimensional calibration board; establishing a camera calibration model on the basis of the three-dimensional calibration board, acquiring world coordinates of chessboard corners of the three-dimensional calibration board, and extracting pixel coordinates of the chessboard corners on the basis of a corner detection algorithm; and using a least square method to calibrate inner and outer parameters of a camera in combination of the corner world coordinates and the pixel coordinates.

Real-time image stabilization using computationally efficient corner detection and correspondence is provided. The real-time image stabilization performs a scene learning process on a first frame of an input video to obtain reference features and a detection threshold value. The presence of jitter is determined in a current frame of the input video by comparing features of the current frame against the reference features using the detection threshold value. If the current frame is found to be unstable, corner points are obtained from the current frame. The obtained corner points are matched against reference corner points of the reference features. If the number of matched corner points is not less than a match point threshold value, the current frame is modeled using random sample consensus. The current frame is corrected to compensate for the jitter based on the results of the modeling.

A capacitance-type detecting device according to an embodiment of the invention includes a plurality of detecting electrodes that are provided on a sensor substrate such that capacitance is formed between adjacent electrodes. Capacitance formed in a corner detection region in which the detection sensitivity of the sensor substrate is relatively low is more than that formed in a central detection region in which the detection sensitivity is relatively high.

The invention discloses a two-dimensional image sequence based three-dimensional reconstruction method of a target, and relates to the three-dimensional reconstruction method of the target, which solves the problem that in the traditional image-based three-dimensional reconstruction method, the reconstruction precision is low due to more points needing to be reconstructed and large calculation quantity. The three-dimensional reconstruction method comprises the following steps of: using a camera to obtain a two-dimensional image sequence of the target, calculating and matching each image through a scale invariant feature transform (SIFT) algorithm, and calculating the geometric relationship between images; carrying out the corner detection of each image in a Gaussian scale pyramid generated in the realizing process of the SIFT algorithm, and obtaining the multi-scale corner features of the images; taking the obtained SIFT matching point as a center, searching a corner corresponding to each image in a limited range of a restrained distance, and matching the corners obtained by each image to obtain the matched corner; and realizing the three-dimensional reconstruction of the target by carrying out the three-dimensional reconstruction of the matched corner according to a projection matrix of a camera. The two-dimensional image sequence based three-dimensional reconstruction method is applied to the three-dimensional reconstruction of the target.

The invention discloses an information acquisition and transfer method for assistant vision systems. The method comprises the following steps: (1) extracting two original digital images of an object in different angles by using two cameras at the same time; (2) extracting characteristic points of the two original digital images by means of the Harris corner detection; (3) extracting three-dimensional geometrical information of the characteristic points by using the two cameras; (4) making a rectangular region where each characteristic point serves as the center, finding out the position of the next frame characteristic point and calculating motion vectors of the characteristic point; (5) dividing the road surface information of the original digital image by using a color histogram, according to the chromatic information and calculating the road information; (6) coding the motion information of the characteristic point of the original image, the three-dimensional geometrical information of the characteristic point and the road information respectively; and (7) transferring the coded information to a person with vision disorders via the information transfer array unit in the assistant vision system. The information acquisition and transfer method is advantageous in the accurate extraction of three-dimensional geometrical information of the object, and helps the patients with vision disorders to walk directionally and safely.

With an object of solving problems existing in the prior art, the invention provides a fully automatic calibration method for a high performance camera under a complicated background. The method is combined with a Robust's checkerboard corner detection method, and two groups of checkerboards corners are adopted to serve as filters on the basis of the characteristics of the corners to filter marker images. Eight filters of two types are reduced to four filters of two types, which means that the wave processing amount halves and therefore, the calibration speed increases. Further, the method utilizes a Zhang's camera calibration method to mark the camera calibration parameters. The method utilizes the marked camera calibration parameters to standardize images to be corrected, and a normalized correlation of the standardized images is calculated, and then sub-pixel precision checkerboard corners are obtained. The method then re-projects the coordinate of the checkerboard corners onto image space to obtain a precise coordinate of the corner image. The newly obtained coordinate of the corners is then substituted into the Zhang's calibration method to obtain new computed camera parameters. By repeating the above steps, a performer can obtain highly precise camera parameters. According to the embodiments of the invention, it is possible for a performer to complete automatic corner detection and camera calibration without having to resort to human-machine interactive operations.

The invention discloses an analysis method of a video-based crowd density and abnormal behavior detection system, and relates to the fields such as intelligent video surveillance and target detection. The analysis method comprises the steps of performing density grading by using an improved method based on pixel statistics and texture statistics; performing estimation on the number of people by using a method combining pixel statistics and foreground corner detection; introducing a local optical flow, proposing and implementing a crowd abnormal behavior detection algorithm based on the average kinetic energy change rate. The analysis method disclosed by the invention not only can be applicable to general surveillance video, but also can be more applicable to crowd gathering video in prominent large public places. The analysis method is less in system time consumption, has good effectiveness and practicability and meets actual requirements.

The invention puts forward an all-weather target real-time tracking method based on visible light and infrared images. The method comprises the following steps: preprocessing a visible light image and an infrared image to remove image noise and blind pixels; extracting corners from the visible light image and the infrared image by use of a Harris corner detection method, tracking the detected corners by use of an optical flow method, and registering the visible light image and the infrared image by use of an RANSAC algorithm; performing target detection on the registered images by use of an MHI method, and morphologically processing a detected target by use of a corrosion expansion method; and using a Kalman filter to track and outline the detected target, performing track correlation on the target based on the characteristics of the target, so as to monitor the target in real time. The detection probability of a target in a visible light image and an infrared image can be improved, and a moving target can be detected and identified with high efficiency under multiple spectrums and all weather.

The invention relates to the field of image processing, and discloses a method of extracting a rectangular building from a remote sensing image. The method comprises the steps of obtaining a plurality of superpixel region blocks by superpixel segmentation of a remote sensing image; determining two seed points on a target building in the remote sensing image; merging the plurality of superpixel region blocks based on the determined seed points; carrying out corner detection for the remote sensing image; calculating to generate a corner distance saliency map related to each pixel point in the remote sensing image based on the corner detection result; carrying out binary segmentation for the corner distance saliency map; determining priori information based on the merging result of the plurality of superpixel region blocks and the distance saliency map after binary segmentation; obtaining a building segmentation result by segmenting the remote sensing image based on the priori information; carrying out morphological image processing for the building segmentation result; and obtaining a rectangular target building by rectangular fitting of the building segmentation result after the morphological image processing. The above the method can accurately extract the rectangular building from the remote sensing image.

A checker corner detection method uses a Harris corner detection algorithm to detect a checker image, thus obtaining candidate corners; a coordinate of the candidate corner is very precise and at a sub pixel level; taking each candidate corner as a center so as to obtain a square symmetrical template; using the square symmetrical template to process the candidate corners, and removing false corners, thus obtaining check corners. The checker corner detection method analyzes gray value distributed regularity in checker image corner neighbourhood, uses the square symmetrical template taking the corner as the center, and employs the square symmetrical template to process the candidate corners and removes the false corners so as to obtain the check corners; the method is short in calculation time, and high in detection precision.

The invention provides a multi-feature fusion video news abstract extraction method, and the method obtains a video news basic processing unit through the lens dividing and lens calibrating of original video news. The method comprises the steps: carrying out the theme caption detection of the video news basic processing unit in the static video abstract through corner detection, color filtering and morphological analysis, and obtaining news main figure information through face identification and clustering; carrying out the silencing analysis of the video news basic processing unit in a dynamic video abstract, and obtaining silencing segments; judging whether each silencing segment contains theme caption or not: adding the theme caption to a final abstract sequence if each silencing segment contains theme caption, and vice versa; and thereby obtaining meaningful high semantic dynamic and static video abstracts. The method can obtain the static abstract, and can flexibly browse and organize the video contents. The obtained dynamic abstract comprises abundant multimedia information, can clearly represent the contents of an original video, and is high in entertainment and ornamental values.

The invention provides a full-automatic camera parameter calibration method based on a colored calibration board. The method includes the following steps that firstly, the colored chess calibration board is drawn, wherein four neighborhoods of each corner are a combination of different colors; secondly, the movement of the colored chess calibration board is controlled by a mechanical arm to set presetting bits and capture calibration images; thirdly, corner detection is carried out according to the calibrated images obtained in the second step; fourthly, fake corners are removed according to whether 'color codes' of the corners are in a 'color code' library of the calibration board or not; fifthly, all parameters for camera calibration are obtained based on space coordinates of the true corners obtained in the fourth step and pixel coordinates of the true corners in the calibration image; sixthly, the camera parameters are acquired to complete camera calibration.

The invention relates to an electronic image stabilizing method for digital videos, belonging to the technical field of digital image processing. The method comprises the following steps of: firstly, extracting characteristic points in two adjacent images in a video sequence by utilizing an SUSAN corner detection operator, matching the characteristic points by utilizing a relevant matching method and rejecting false matching point pairs; then selecting three pairs from the correct matching point pairs and solving for an affine transformation parameter between two images by utilizing a corresponding coordinate relation of the matching point pairs on a two-dimensional image; and finally, carrying out an affine inverse transformation on a next image according to an affine formula to obtain an image registered with a previous image in space so as to realize the electronic image stabilization of the digital videos. The electronic image stabilizing method of the digital videos can automatically eliminate jitters of the video images in real time, wherein the jitters comprise translations and rotating movements among images.

The invention provides a retinal blood vessel morphology quantization method based on a connected region. The method obtains a retinal blood vessel segmentation image after the fundus image is preprocessed, and then performs post-processing on the blood vessel segmentation image. On this basis, the vascular network is thinned and boundary treated, and the vascular centerline network and vascular boundary map are obtained. Corner detection is then performed and removed from the vascular centerline network so that the vascular segments of the vascular network form separate communication areas. Traversing is performed on the blood vessel segment, approximate the centerline of the blood vessel segments, and the blood vessel segment is changed into a broken line to calculate the direction of the blood vessel. At last, that initial diameter value is calculated, the center of the circle is selected by sliding on the centerline of the blood vessel segment, a semicircle window is created according to the direction of the circle cardiovascular and the diameter value measured in the early stage, and the distance between the window and the two intersection points of the blood vessel boundary is taken as a new diameter value. From this iteration, a set of vessel diameter values are measured, and the median value is the vessel diameter of the vessel segment. The invention is applicable to the quantification of large-scale retinal blood vessel morphology, and has high reliability.

The invention provides an FPGA-based real-time panoramic image mosaic method. The method is characterized by, to begin with, carrying out gray conversion on collected video data; then, extracting feature points of each frame image in a video stream through a Harris corner detection method; after obtaining corner coordinates of the front-back two frame images, carrying out calculation on regions with corners as centers respectively through an NCC-SSDA fusion algorithm to find out the correspondence relation between the feature points of the two frame images; removing unmatched point pairs through an RANSAC algorithm, estimating affine transform model parameters, finding an optimum model and taking all of local inner points according with the model as the final matching result; and finally, carrying out fusion on image edges through a weighted smoothing algorithm, and seamlessly displaying images after mosaic. The method can quickly finish single-camera 360-degree panoramic mosaic, and has the advantages of high registration accuracy, high real-time performance and good stability.

The invention relates to the technical field of image processing, and provides a corner detection-based distorted image correction method for the problem of image distortion caused by non-flat shooting of a rectangular image, which comprises the following steps of: extracting the edge of a target area of an image to be corrected to obtain an edge image; extracting candidate angular points according to the edge image to obtain a candidate angular point set; Filtering candidate angular points in the candidate angular point set to obtain four target angular points; according to the method, the to-be-corrected image is converted according to the target corner point to obtain the corrected image, so that the problem of image distortion caused by the non-flat-shot image is solved, the accuracy of subsequent OCR recognition can be further improved, the input of manpower and material resources is reduced for downstream OCR application, and the method is suitable for correcting the image distortion of the rectangular area of the target area.

The invention discloses a tracking and matching parallel computing method for a wearable device so as to achieve augmented reality tracking and matching. According to the tracking and matching parallel computing method, the SCAAT-EKF feature tracking technology is adopted, complementary fusion data acquisition is conducted on multiple sensors in the wearable device, and data collision can be effectively avoided; the operation strategy based on the double kernal CPU+GPU group kernel multi-channel is utilized, corner detection and extraction based on the Harris algorithm are conducted in the GPU, the double kernal CPU is used for conducting P-KLT tracking and matching calculation, and therefore algorithm fast parallel processing is achieved. The tracking and matching parallel computing method mainly comprises the steps of hybrid tracking and feature extraction for the wearable device, accurate extraction of feature points of target natural features without marks, Harris corner point detection achieved based on a GPU parallel processing mechanism, the CPU-based P-KLT parallel feature tracking algorithm and the secondary matching optimization algorithm. The tracking and matching parallel computing method for the wearable device achieves combination of the sensors of the wearable device and visual tracking and matching, and has the wide prospect in the augmented reality three-dimensional registration aspect.

The invention discloses an independent-following vehicle course control device and control method, and belongs to the field of automatic control and agricultural work vehicle application. The course control device comprises a detection system and a steering control system; the detection system comprises a front-wheel corner detection mechanism and an opposite course deflection-angle detection mechanism; the front-wheel corner detection mechanism is composed of a front wheel guide shaft, an absolute rotary encoder and an encoder mounting base, and the opposite course deflection-angle detection mechanism is composed of reflection type infrared sensors and a reflection board; the left side and the right side of the front end of a following vehicle are each provided with four reflection type infrared sensors, and the reflection board is arranged at the rear end of a guide vehicle; and according to reflection induction information of the infrared sensors, an opposite course deflection angle between the following vehicle and the guide vehicle is obtained. A steering control mechanism is composed of a stepping motor and a chain wheel transmission pair. A course control system modeling method and a control algorithm can stably and reliably achieve independent following of the following vehicle.

The invention relates to a three-dimensional scene reconstruction method based on a statistical model, which comprises the following steps: using a Harris corner detection algorithm to extract the corners in each image and generating a three-dimensional point set X and a camera parameter set M; using a Markov Chain Monte Carlo(MCMC) method to estimate the match probability among the image corners and the three-dimensional points and subjecting the image corners to weighted mean by using the match probability between the image corners and the three-dimensional points to obtain a virtual measuring point matrix V; subjecting the virtual measuring points to projective reconstruction by using a projective factorization algorithm capable of processing occlusion, adding a deterministic annealing algorithm to iteratively solve a global optimal protective reconstruction result, and using a camera self-calibration algorithm based on an absolute dual quadric surface to promote the projective reconstruction to metric reconstruction. The original process of one-time calculation is converted into a process of iterative refinement. Even though a matching relationship is unknown or a primary matching result is bad, a three-dimensional reconstruction result is finally obtained precisely through an iterative feedback method.

The invention relates to a urine test paper physiological index quantification identification method based on a cloud platform, and the urine test paper comprises three parts including a two-dimensional code, a standard color card and a test paper. After using the test paper for a urine test, a user takes a photo of the test paper through a mobile phone, a test paper image is automatically acquired and uploaded to the cloud platform, the cloud form integrates an image identification method, the image is processed, and a result is returned to the mobile phone. The identification method comprises the steps: Harris corner detection and calibration are performed on the image; the image space is converted from RGB to CIELAB, the chromatic aberration between the color of the test paper and the color lump of a standard colourimetric card is calculated, a KNN (k-Nearest Neighbour) is adopted to sort the color of the test paper to a corresponding kind in the colourimetric card, and the analysis for test paper measurement index is automatically completed. According to the invention, the method is rapid and simple, problems of ambient light interference and low human eye identification accuracy are avoided, the cost is low, in addition, the method can be used for screening patients with early chronic kidney disease and patients with diabetes, and the risk of diseases is reduced.

The invention relates to the field of machine vision, and discloses a pointer type instrument reading automatic identification method based on Faster R-CNN and U-Net, and the method comprises: S1), making a Faster R-CNN data set; S2) establishing a Faster R-CNN network model, and respectively training and testing the Faster R-CNN network model; S3) constructing a U-Net network model, establishinga loss function L, and respectively training and testing the U-Net network model; S4) fitting a scale line contour by utilizing a scale line segmentation result; S5) calibrating the dial image by using perspective transformation; S6) detecting a pointer area by utilizing a Faster R-CNN network model; and S7) obtaining a pointer inclination angle of the dial calibration image and a final result. According to the invention, the Faster R-CNN model is adopted to replace ORB and other traditional corner detection algorithms, so that the accuracy of detecting the area where the instrument dial and the instrument pointer are located is improved; and a Hough transform algorithm is replaced by an image segmentation and contour fitting method, a U-Net model and a corresponding loss function are redesigned for the characteristics of the electric power instrument, and the automatic identification accuracy is high.

The invention discloses a rapid and high-precision matching method of a depth image and a color image. The method comprises the following steps: (1) Carrying out interframe smoothing and histogram equalization on the depth image shot by a depth camera; carrying out median filtering on the color image; (2) Carrying out multi-scale spatial Harris-Laplace corner detection on the depth image and the color image so as to obtain characteristic points; (3) Describing the detected characteristic points; (4) Using a KNN algorithm to match the characteristic points of the depth image and the color image so as to obtain candidate characteristic point matching pairs; (5) Using a RANSAC algorithm to eliminate the mismatched candidate characteristic point matching pairs so as to obtain the characteristic point matching pairs. The prior art is difficult to be used in the matching field of the depth image and an optical image. By using the method of the invention, the above problem can be solved. And robustness and accuracy can be increased.