64 results about "Image geometry correction" patented technology

The present invention provides a Satellite-bone camera space image distortion correction method based on the stellar map matching, the stellar point on the image is used as referecen point, achieving the distortion correction, including following steps: (1)image segmentation, extracting the stellar facula in the image; (2) calculating areal coordinates of each facula by the centroid method, as a real coordinate; (3) processing pre-correction according to the deformation parameter obtained by the ground standardization of the camera; (4) processing stellar map matching by single frame or multiple frames Hausdorff distance stellar map matching method, calculating academic coordinate corresponding to the practical coordinate; (5) calculating deformation parameter of the image by the corresponding relation between the practical coordinate and the academic coordinate; (6) obtaining corrected image by the three steps fitting of a polynomial and interpolation method. In addition, the invention aiming at correctoon impact of the camera plain shaft point system error, provides a adaptive compensation method. The invention has small operation, evident image correcting effect, and effectively preventing the impact of the correction caused by the image noise and the input system error.

The invention discloses a geometric correction method for a spotlight-mode satellite SAR image. The method comprises the following steps: (1) reading related parameters and echo data, and constructing a positioning equation; (2) calculating the coordinates of the position vector and velocity vector of a satellite in a non-rotating geocentric coordinate system at a moment of radar illumination center according to the spatial geometric relationship between the satellite and the earth; (3) completing slope distance and Doppler frequency correction; (4) acquiring the coordinates as well as longitude and latitude of a certain pixel point (i, j) of the satellite SAR image in the non-rotating coordinate system; (5) determining whether all of the pixel points of the satellite SAR image are processed; and (6) re-dividing the latitude-longitude grid, and outputting the geometric correction result. The method provided by the invention can be used for carrying out row-by-row geometric correction on the satellite SAR image, and can process the satellite SAR image in blocks as required due to the completely independent treatment process between the rows of the satellite SAR image, thereby further improving the treatment efficiency.

The invention provides a projection image correction method and a projection device. The method comprises the steps of projecting a correction image of which four angles are equipped with correction identification points in a projection screen; moving the correction identification points in the correction image to the four angles of the projection screen; obtaining the coordinate values of the correction identification points in the correction image after movement; and projecting the coordinate values of the four vertex angles of the original projection image to the coordinates of the correction identification points according to an image geometric correction algorithm and carrying out image projection display according to the coordinate values after projection, thereby correcting the projection image automatically by the projection device. According to the method, the system and the projection device, the image geometric correction is realized by the low cost method without a camera or an accelerometer, the projection image and the projection screen can be matched precisely, the intelligence of the projection device is improved, and moreover, convenience is provided for a user.

The invention discloses an image characteristic registration based geometrical fine correction method for an aviation multispectral remote sensing image, comprising following steps of: 1) utilizing a Sobel edge detection operator to carry out edge extraction on both an aviation multispectral remote sensing image to be registered and a normal incidence high-resolution satellite image which is taken as a standard; 2) utilizing a Harris angular point detection method to detect angular points of the aviation multispectral remote sensing image to be registered and the normal incidence high-resolution satellite image which is taken as the standard; 3) carrying out rough correlation between the two images through a correlation method; 4) carrying out fine correlation between the angular points of the two images through calculating supporting strength; 5) carrying out back calculation to obtain a multinomial coefficient according to a multinomial model; and 6) carrying out gray level re-sampling by adopting a bilinear interpolation to obtain a registered image. According to the image characteristic registration based geometrical fine correction method for the aviation multispectral remote sensing image, disclosed by the invention, the geometrical correction problem of the aviation multispectral remote sensing image lacking of a ground reference point can be better solved and the geometrical correction precision of an aviation multispectral scanner is improved, so that an aviation remote sensing technology can be better applied to production livings of national economy.

An image geometric correction device 6 unifies the coordinate systems of images captured by image capturing devices 4-1 and 4-2 into a coordinate system, and makes a geometric correction to the images projected by image projection devices 1-1, 1-2, and 1-3 on the basis of an image projection region E in the coordinate system into which the coordinate systems of the captured images are unified. As a result, the image geometric correction device can make a geometric correction to the projected images without strictly placing the image capturing devices 4-1 and 4-2 at predetermined positions.

The invention discloses an examination paper marking method based on augmented reality technology. The method comprise the step processes of first-phase preparation work, core image processing work, score accounting period, score augmented display period and the like; the first-phase preparation work comprises design of answer sheets, entering of test paper standard answer, information acquisition of answer sheets and the like; the core image processing work comprises geometric correction of imagery, effective region positioning of the answer sheets, detection and extraction of characteristic elements of the effective regions of the answer sheets and result writing of xml files; the score accounting period mainly comprises accounting of scores of the answer sheets; the score display method mainly comprises displaying the scores by combining the augmented reality technology, recording for a user and the like. According to the method, the device is simple and convenient, the operation is easy, various conditions of correct choice, mistake choice, multiple choice, mis-choice and the like can be identified, the accuracy rate is higher, the detection identification speed is higher, the method is greatly convenient for the user, and the method has extremely good research values and practical values.

The invention relates to a method for geometrically correcting vertically mapped images of fisheye lenses in fisheye image mosaic. Fisheye images can severely distort radially, and a mosaic effect is affected. The method includes enabling the four vehicular fisheye lenses to respectively acquire near-distance scenes in four directions of a vehicle body; unifying images of the four fisheye lenses on a ground coordinate system by means of mapping images and estimating pitching angles; geometrically correcting the images to form third-angle vertical overlook images with consistent visual perception. By the method, the four fisheye lenses are divided into two groups according to adjacent relations, the pitching angles and mosaic angles of each two adjacent images in each group are respectively solved, the pitching angles and mosaic angles of two groups of images are solved according to adjacent relations between the two groups of images of the lenses, and finally results are converted into respective independent pitching angles of the four images. The method has the advantages that fisheye image mosaic can be effectively implemented, and a seamless high-definition image can be finally formed.

The invention discloses a missile-borne double-base forward vision SAR ship target positioning optimization method. The method comprises the following steps of determining a missile-borne double-base forward vision SAR platform, wherein the missile-borne double-base forward vision SAR platform detects whether there is a target in a range and takes a scene where the target is located as a target scene, the missile-borne double-base forward vision SAR platform includes a transmitter and receiver, the transmitter emits a signal to the target scene, and the receiver receives an echo signal reflected by the target scene, and then carries out distance direction and azimuth sampling so as to acquire sampled signals; successively carrying out matching filtering and imaging processing on the sampled signals and acquiring a double-base forward vision SAR image; determining a center point of the target scene, and calculating a distance between the receiver and the center point of the target scene; carrying out geometric image correction on the double-base forward vision SAR image and projecting to a ground distance plane and acquiring a ground distance SAR image; and according to the distance between the receiver and the center point of the target scene, acquiring missile-borne double-base forward vision SAR ship target positioning information.

The invention discloses a cooperated detecting method for ice of The Yellow River based on a multispectral image and an SAR image, wherein the cooperated detecting method is used for settling a technical problem of low precision in existing detecting methods for the ice of The Yellow River. The cooperated detecting method is characterized in that a remote sensing image is used; through performing preprocessing such as geometric correction, Bow-tie correction, cloud elimination and image cutting on a Yellow River remote sensing image, performing rough dividing on a river area, and integrally utilizing an NDSI detecting method and a clustering method, a divided area set m is obtained; then the area m is used as an SAR image detecting method input, a divided area set n is obtained through super-pixel dividing and calculating a salient map, and finally a solution for optimizing a Yellow River ice detecting model is searched through alternative iteration, thereby obtaining an optimal ice detecting area. Quick and effective testing for the ice of The Yellow River and high precision are realized.

The invention discloses a high-precision multi-leaf area measuring method, including the following steps: early photographing preparation; image acquisition; geometric correction of an image; image segmentation and binarization; sub-pixel edge extraction; calculation effective pixels of an objective table and leaves; removal of isolated noisy points; and leaf area calculation. The high-precision multi-leaf area measuring method provided by the invention adopts an image sub-pixel edge extraction technology, thereby improving the measurement precision of the leaf area; and utilizes an image processing technology to automatically calculate the leaf area, thereby improving the measurement efficiency.

The invention discloses an image watermarking method based on phase singular value correction. The method comprises steps of: firstly embedding a watermark into a low-frequency sub-band transformed by a host image PDTDFB; secondly, extracting the high-frequency sub-band phase matrix of a to-be-detected image subjected to quaternion two-dimensional wavelet transform, subjecting the phase matrix to singular value decomposition (SVD), selecting the first four singular values in the main singular value as the eigenvector of a training sample image, and subjecting the sample to the training learning of geometric transformation parameters such as scaling, translation and rotation so as to obtain an LS-SVR training model; then calculating a mathematical model of the geometric correction of the watermarked to-be-detected image, and determining the model parameters by using the obtained LS-SVR training model to perform geometric correction; and finally, extracting the watermark information from the corrected watermarked image.

The present invention discloses an airborne hyperspectral image geometric correction device based on a three-way array camera and an airborne hyperspectral image geometric correction method thereof. A low-cost airborne stable platform is employed to realize rough direction of the visual axis of a camera and allow the attitude angle of a hyperspectral camera to the ground to be stabilized in the range of +_0.05 degrees and +-0.1 degrees. Three high-speed area array CMOS cameras being in consolidation with the hyperspectral camera are employed to perform continuous imaging of the earth's surface, and the feature matching of the images of front and back area arrays is employed to measure the accurate attitude angle of the hyperspectral camera. The attitude angle measured by employing the high-speed area array CMOS cameras is employed to perform pixel-grade real-time geometric correction for the hyperspectral images. The problem of geometric correction of the hyperspectral images through adoption of the low-cost airborne stable platform is solved. The method is simple in structure, low in cost, high in precision and fast in calculation speed and can perform pixel-grade real-time geometric correction of the images.

The invention relates to a method for constructing the geometric correction of expandable multi-satellite multi-sensor remote sensing images, which comprises the following steps: 1) according to the processing flow of a geometric correction algorithm, dividing the geometric correction of the remote sensing images into the steps of image data input, the calculation of the range of output images, image correction model, the conversion of pixel positions, a resampling algorithm and the output of correction images, and then setting the processing logic of the steps in a general invoking method of an abstract class; and 2) dividing the process of the method for constructing the geometric correction into three layers further, namely a geometric correction processing abstract function interface layer, a specific realization layer and a basic arithmetic layer; and taking satellite names or sensor identification as input parameters by users, and selecting a specific realization algorithm to be used to realize the corresponding algorithm and complete the corresponding processing function during program running. The method has the advantage that the function of the geometric correction of the multi-satellite multi-sensor remote sensing images can be processed and the required function can be expanded under the condition of no need of modifying the existing programs.

This invention relates to infrared sequence image history trend analysis method and belongs to device fault diagnose field. This invention comprises the following steps: first to collect the infrared heat images caught in different periods of the device to be measured into one analysis file fold; second to input position points by human and use artificial conversion matrix, project conversion matrix or quadrics conversion matrix; then to use double linear insert and correct the image by geometry and aligns it according to reference image space position; third to use the characteristics of side scanning of thermal imaging device to scan images to get the interest points position and temperatures and give and display the change curve in different temperature at the interest points. íí

The invention discloses a robust image watermarking method based on joint statistical model correction. The method comprises steps of firstly embedding a watermark in a low-frequency subband of the PDTDFB transform of a host image; secondly, extracting the amplitude matrix and the relative phase matrix of a high-frequency subband coefficient of a to-be-detected image after PDTDFB conversion, usingthe Weibull distribution probability density function and Vonn distribution probability density function combined statistical model to model the amplitude matrix and the relative phase matrix of thecoefficient R, selecting the shape parameters and scale parameters obtained from the modeling as feature vectors of a training sample image, and then training and learning a sample to obtain the LS-SVR training model; then, calculating the mathematical model of the watermark image to be detected, and using the obtained LS-SVR training model to determine the model parameters for geometric correction; and finally, extracting the watermark from the corrected watermark image.

The invention provides a 4-path borehole wall video cylinder panoramic image generation method and device. The generation device comprises 4-path video input devices, a video preprocessing device, a cylindrical projection device, an image registration device and an image splicing and fusion device. The image preprocessing device comprises an image geometric correction device and an image color correction device. According to the 4-path borehole wall video cylinder panoramic image generation method and device, the panoramic image generation method is based on four camera systems, which are vertically arranged mutually, so that the technical problems above are solved fundamentally, and borehole imaging technology is allowed to be subjected to breakthrough and substantial development; through 4-path borehole wall videos, periphery of the borehole wall is imaged from four different directions simultaneously; video images in the adjacent two directions have a borehole wall same area imaging part, and video image frames in the same direction have a borehole wall same area imaging part; and through image features of video images, a panorama video is obtained after splicing processing.

The invention relates to an image registration method for a narrow overlapping dual field high spectral imaging instrument. According to the method, a ground coordinate of an image after geometric correction is utilized to roughly calculate a corresponding left and right field relation, mutual information in a certain relative position scope is calculated by taking the corresponding relation as reference, mutual information calculated when overlapped image elements are in a small quantity is removed, the largest mutual information is searched from all the left mutual information, a corresponding relation of a left and right field image corresponding to the largest mutual information is the best corresponding image matching relation. Through the method, the ground coordinate of the image after geometric correction and the mutual information are utilized to calculate the image matching relation, a splicing error problem caused by an inaccurate ground coordinate after geometric correction and an information registration matching error problem can be solved, and the registration relation of the image can be rapidly and accurately acquired.

The invention discloses a multi-view document image geometric correction method based on a mobile terminal. The multi-view document image geometric correction method comprises the following steps: shooting left and right view distorted document images of an unfolded bent document; compressing the shot left and right visual angle distorted document images; performing feature point matching on the compressed left and right visual angle distorted document images to obtain a feature point matching pair set; based on calibration internal parameters of the camera, solving a rotation matrix and a translation vector when left and right visual angle distorted document images are shot; calculating three-dimensional coordinate points of the unfolded and bent document page based on the solved rotationmatrix and translation vector of the camera when the left and right visual angle distorted document images are shot; constructing an expandable curved surface model for the unfolded curved document page by adopting the calculated three-dimensional coordinate points; and correcting the three-dimensional coordinate points to a two-dimensional plane based on the constructed developable curved surface model, and filling a blank area to obtain a final flat plane image. According to the scheme, the OCR recognition accuracy of the distorted document image can be improved.

The invention discloses an imaging detection system and method, and aims to solve the problems that an imaging area lacks matching characteristics and needs more than two times of imaging for effective coverage, and geometric parameters of flaws, defects and other diseases in an image need to be accurately measured. A background conformation device generates a laser background pattern and projectsthe background pattern to an imaging detection area. An imaging device is composed of a camera and a corrected pattern projection device. The imaging device acquires a target image having a correctedpattern and a background pattern. When one background conformation device cannot cover the imaging detection area, a plurality of background conformation devices are used for projecting background patterns, the background patterns projected by the plurality of background conformation devices are overlapped in the imaging area, and the space combination relation of the background patterns formed by the plurality of background conformation devices is not changed in the imaging process. Image geometric correction processing is completed according to the corrected pattern projected by the imagingdevice, and multi-image mosaic matching is completed based on the background pattern.

The invention provides a method for high-resolution image fusion. The method comprises: segmenting a to-be-displayed high-resolution image by responding to a display request of a user terminal to obtain a plurality of segmented images; carrying out overturning processing on each segmented image and segmenting each processed segmented image again to obtain a plurality of re-segmented images; zooming each re-segmented image to obtain a first image; cutting the first images according to a resolution of a fused bandwidth to obtain second images; carrying out overturning processing on the second images to obtain third images; and transmitting each third image to a corresponding ray machine to carry out projection and carrying out splicing and fusion to obtain a to-be-displayed image. Accordingto the invention, the steps of image geometric correction and fusion are simplified, so that the method, apparatus, and system for high-resolution image fusion are suitable for the whole industry; thehigh-resolution geometric correction and fusion costs are lowered; and the high-resolution geometric correction and fusion are realized by using the existing processing chip.

The invention discloses a method for measuring the ear length of multiple rice plants based on image processing. The method comprises the following steps: S1, photographing to-be-measured rice so as to obtain a rice image, and recording the image as jpg1; s2, performing image geometric correction processing on the image jpg1 to obtain an image jpg2; s3, performing illumination compensation processing on the image jpg2 obtained in the step S2, performing binaryzation on the image by using an Otsu method OTSU, segmenting a single rice plant into two parts from a binary image by using an image segmentation technology, and reserving the part containing the rice spike length to obtain an image jpg3; and S4, carrying out morphological expansion processing on the graph jpg3, then carrying out contour extraction, carrying out calculation by utilizing a Zhang skeleton quick parallel refining algorithm, and recording a result graph as a graph jpg4. The digital processing technology is applied toautomatic rice ear length measurement in the agricultural field, manual measurement and cumbersome instrument analysis are replaced, the labor cost and the equipment cost are saved, and the working efficiency is greatly improved.

The invention relates to the technical field of image processing, and particularly relates to an improved image registration algorithm based on SIFT features. The objective of the invention is to aimat feature point detection in the existing SIFT algorithm. The problems of large time consumption during the detection stage, low detection efficiency and the like in the prior art can be solved. An improved SIFT algorithm based on a pyramid structure and a feature point neighborhood is provided. The algorithm comprises the following specific steps of A, image acquisition; B, image geometric correction; C, image preprocessing; D, image registration; E, image fusion. The algorithm of the invention is used for the automatic detection of automobile parking claw parts.

The invention provides a method for measuring the ear length and the ear number of multiple wheat plants based on image processing, and belongs to the technical field of computer image processing. Themethod solves the problem that the number of spikelet is mainly observed and counted by human eyes in the existing manual visual inspection, and is characterized by comprising the following steps: S1, acquiring a wheat graph; s2, geometrically correcting the image; s3, carryng out single-plant wheat segmentation; s4, cutting out an effective area of the length of the wheat ears; s5, calculating the ear length; and S6, calculating the spikelet number of the wheat. The method has the advantages that the labor cost and the equipment cost are saved, the working efficiency is greatly improved, andthe like.

The invention discloses a method for geometrically correcting a woodcut Tibetan image. A large number of woodcut Tibetan document images are generated in the informatization era, geometric correctionof the woodcut Tibetan document images is the basis of document analysis and character recognition, but a traditional geometric correction method is difficult to implement. Therefore, according to thedeformation characteristics, the thought that the length is shortened and the complexity is reduced is adopted, a segmentation strategy for geometric correction of the woodcut Tibetan document imageis provided, the image is segmented into a plurality of sub-images in the long edge direction, and projection transformation is adopted for correcting all the sub-images. Method of the invention, theproblem that the existing projection method, Hough transformation method, Fourier transformation method, neighbor method and the like which take inclination angle detection as a main thought cannot besuitable due to irregular bending of the Tibetan image of the woodcut plate is solved; and the problem that the existing correction method taking mathematical modeling as a main idea is difficult toapply due to the complexity of deformation of each page and the difference between the pages is solved, and the geometric correction effect of the Tibetan document image of the woodcut board is improved.

The invention discloses an unmanned aerial vehicle route planning method suitable for agricultural investigation; the method comprises the steps: firstly, obtaining an overall image, obtaining the coordinate information of a ground control point, carrying out the image geometric correction of high-resolution satellite remote sensing data and the aerial photo of an unmanned aerial vehicle in a high-flight to-be-investigated region; then inputting the corrected unmanned aerial vehicle aerial photo into a pre-trained YOLOv3 field obstacle recognition model to obtain an obstacle coordinate of the area, and using a multi-scale combination aggregation segmentation algorithm to realize field boundary automatic extraction; obtaining the central points of the field pieces by using the polygonal triangulation skeleton diagram, eliminating redundant central points of the field pieces according to the coverage range of pictures shot by the unmanned aerial vehicle, and generating an initial route; then, generating a buffer area in a certain range by taking the obstacle as the center, judging whether the initial waypoint is in the buffer area or not by adopting a ray guide method; finally, eliminating the waypoint in the buffer area, and generating a final route.

An image geometric correction device 6 unifies the coordinate systems of images captured by image capturing devices 4-1 and 4-2 into a coordinate system, and makes a geometric correction to the images projected by image projection devices 1-1, 1-2, and 1-3 on the basis of an image projection region E in the coordinate system into which the coordinate systems of the captured images are unified. As a result, the image geometric correction device can make a geometric correction to the projected images without strictly placing the image capturing devices 4-1 and 4-2 at predetermined positions.

The invention provides an image geometric correction method for a space camera with a 45-degree rotary scanning mode. According to the image geometric correction method, a theory and an imaging process of the space camera with the 45-degree rotary scanning mode are analyzed, a reasonable positional correspondence model is established between a position in a remote sensing image and a corresponding ground position, and a coordinate conversion relationship between an original image and a corrected image is constructed; according to the coordinate conversion relationship, space coordinate projection is performed on the original remote sensing image; and a discrete point after projection is subjected to re-sampling by utilizing a Delaunay triangulation algorithm. For ground remote sensing devices sharing a main optical system, a plurality of wavebands can share a coordinate projection relationship and a Delaunay triangular mesh, thereby easily and conveniently correcting data of all the wavebands. The method is reasonable, easy to implement and obvious in effect of improving image distortion.

The invention discloses an SAR image real-time geometric correction method and a system based on RD plane pixel mapping. The SAR image real-time geometric correction method comprises two methods of mapping correspondence of an RD imaging plane and a ground distance plane and ground distance pixel scale determination. According to the method, an SAR radar imaging principle is utilized to calculatea distance sampling interval and an azimuth Doppler sampling interval in an original RD imaging plane, pixel coordinates of the original imaging result RD plane are calculated, and ground distance position coordinates of pixels of a real-time SAR imaging result are calculated through an imaging geometrical relationship. Meanwhile, in the pixel ground distance grid positioning process, ground distance pixel scale adjustment is achieved through ground distance position coordinates. The method can be applied to a real-time SAR imaging system, image geometric correction with a simple processing process and high real-time performance can be realized, and the problem of geometric correction of the input image of a real-time airborne / missile-borne SAR scene matching system is solved.

The invention discloses a projection method and projection system applied to a building facade, and belongs to the technical field of projection, and the method comprises the steps: obtaining an outercontour of a target building; according to the outer contour of the target building and a preset projection ratio, dividing the area where the outer vertical face of the target building is located into a plurality of rectangular projection areas; arranging a plurality of projectors outside the target building according to the plurality of rectangular projection areas; adjusting the projection angles of the plurality of projectors, and performing image geometric correction and image fusion on the plurality of projectors until the projection images of the plurality of projectors projected ontothe outer vertical surface of the target building form a complete image. According to the embodiment of the invention, the whole building facade is used as the projection screen, and the projection image is directly projected on the building facade through the plurality of projectors, so that the perfect combination of the projection image and the building facade is realized, and finally, the optimal visual effect is created.