71 results about "Subpixel edge detection" patented technology

The invention relates to an on-line detection method for diameter size of a shaft disc type part, and belongs to the technical field of image measurement, which can detect variation in the diameter size during processing of the shaft disc type part in real time and guarantee higher detection efficiency and measurement accuracy. The method comprises the following steps of: calibrating internal parameters of a charge coupled device (CCD) video camera and correcting distortion of an image caused by lens distortion based on a CCD video camera plane calibration method; calibrating external parameters of the CCD video camera, acquiring the image of the shaft disc type part with a known diameter by using the CCD video camera, extracting two edges of the part from the image by a subpixel edge detection algorithm, and calibrating to acquire external parameters of the part relative to the CCD video camera by using coordinates of the subpixel edges and the acquired internal parameters of the camera video; and finally, measuring, acquiring the image of the detected shaft disc type part by using the CCD video camera, extracting the subpixel edges of the part and calibrating to acquire the internal parameters and the external parameters so as to solve the diameter size of the detected part.

The invention discloses a rapid sub pixel edge detection and locating method based on machine vision. The method includes the following steps that firstly, a detection image is acquired; secondly, denoising pretreatment is conducted on the image; thirdly, the gradient Gx of each pixel point in the horizontal direction and the gradient Gy of each pixel point in the vertical direction are calculated; fourthly, the gradient magnitude G0 and the gradient direction Gtheta of each pixel points under polar coordinates are calculated; fifthly, neighborhood pixel points of each pixel point are determined; sixthly, pixel-level edge points are determined; seventhly, the distance between a sub pixel edge point of each pixel-level edge point in the eight-gradient direction and the pixel-level edge point is calculated; eighthly, the distance d between each sub pixel edge point in the actual gradient direction Gtheta and the corresponding pixel-level edge point is calculated; ninthly, a cosine lookup table method is adopted for calculating rectangular coordinates of each sub pixel edge point in the actual gradient direction Gtheta, so that the image edge points are detected and sub-pixel-level localization is conducted. The whole method is high in calculation accuracy and speed.

The invention particularly relates to a method for detecting the camber of plates based on machine vision, belonging to the automatic control field of steel rolling processes. The method comprises thefollowing steps: obtaining plate images through plane array CCD cameras installed right above a roller way outside of roller beds in front of and behind a rolling mill, and correcting the distortionof the images through camera calibration; performing Gaussian low-pass frequency domain filtering on the corrected images, and performing image sharpening differential operation on the filtered imagesto enhance the image boundary. Mathematical morphology transformation and threshold binarization are carried out on the images with enhanced boundaries to segment and completely extract the plate image. The pixel coordinates of the edge contour points are extracted by using the Canny edge detection operator after threshold segmentation and the sub-pixel coordinates of the plate edges are obtained by using the improved Zernike orthogonal moment sub-pixel edge detection algorithm, and the camber value is obtained by fitting.

The invention relates to a measuring device and a method based on an object geometry size interfused by a multiple source image, and the measuring device comprises a data acquisition device that performs the acquisition and the attached processing to the image data of the measured object, a data detecting and processing device that performs a fringe coarse positioning and the secondary image element fringe to the image data and adopts a data fitting method to perform a fitting to the secondary image element fringe, to obtain a fringe interval; a data fusion device that performs the interfusion to the measuring object data withdrawal by the data detecting and processing device. The invention expands a single CCD sensor measurement to a multiple CCD sensor measurement, the multiple sensor redundant information is fully utilized and the weighting fusion treatment is performed to the multiple sensor redundant information, thereby obtaining the estimation result of the multiple sensor fusion; and the precision of the measuring result is improved.

The invention discloses a vision-based high-precision rotation angle measurement method. The measurement method comprises the following steps that a characteristic mark with four circular envelope rectangles is adopted, wherein the characteristic mark is fastened to the rotating surface of a rotary table, and has the same rotation angle as the rotation surface; a region of interest ROI of the image is determined based on template matching, and the influence of the similar edge of the image background on the detection of the rectangular edge in the characteristic mark is eliminated; then high-precision extraction of the rectangular long edge is realized based on the sub-pixel edge detection method of Zernike moment to obtain sub-pixel coordinates of two long edges of the rectangles; fittingof the sub-pixel coordinates of the two long edges is carried out based on a least square fitting method to obtain corresponding fitted edge straight lines and the mean value of the slopes of the twostraight line is calculated; and the rotation angle of the rotation position relative to the reference position is calculated by using an arctangent trigonometric function of the mean value of the two slopes. The method solves the problems that the existing rotation angle measurement method is limited in measurement precision, tedious in process, complex in system and the like.

The invention relates to the field of technologies for detecting yarns, in particular to a device and a method for dynamically detecting multiple targets of yarns of spinning machines. The device comprises a CCD (charge coupled device) camera, an image processing module, a master control system, a sensing assembly, a compact spinning device and auxiliary devices. The CCD camera is provided with an illuminating system, can acquire images of the yarns and can transmit the acquired images to the image processing module, the image processing module comprises an image gray-scale transformation unit, an image filter unit, an image binarization processing unit, a sub-pixel edge detection unit and a size calibration unit, and measured yarn information is transmitted to the master control system by the image processing module and is analyzed and judged by the master control system, so that the quality of the yarns can be detected. The device and the method for dynamically detecting the multiple targets of the yarns of the spinning machines have the advantages that the quality of the yarns of the spinning machines can be automatically detected by the aid of the device and the method, and the accuracy of detection results can be improved.

The invention discloses a method for sub-pixel edge detection based on pseudo Zernike moments. The method is characterized by comprising the following steps: 1, performing noise removal on an input image; 2, performing pixel-level edge detection on the image after noise removal; 3, performing sub-pixel edge detection on a to-be-processed image by adopting a pseudo Zernike moment method; 4, performing error compensation of edge positions on the to-be-processed image; and 5, obtaining a corrected actual edge after the sub-pixel edge detection, and processing all pixels of the to-be-processed image according to the manner of the step 4 to finish the sub-pixel edge detection of the image. The method provided by the invention is insensitive to noise, so that the accuracy of a sub-pixel edge is improved, and the computational complexity required by edge detection is reduced.

The invention provides a sub-pixel-level PCB board defect reference point detection method. The method comprises the following steps: reading in a PCB grayscale image 1; preprocessing the PCB image, wherein self-adaptive median filtering denoising is used to remove pepper-and-salt-type noises to obtain a denoised image 2, a method of maximum inter-class variance is used to carry out binary segmentation of the image to obtain a binarized image 3, and a mathematical morphology method is used to process the image to further remove small interference points and holes to obtain an image 4; using acommunicate-domain algorithm to obtain a position and a scale of a preprocessed image feature region; and then enabling reference circle extraction and positioning efficiency and precision to be improved through a sub-pixel edge detection algorithm and fitting of a least square method. Requirements of high-precision reference point extraction and positioning are effectively satisfied, defects of low precision and low speed of previous reference point positioning are overcome, and the method has the characteristics of high applicability and high accuracy.

The invention provides a sub-pixel edge detection method based on improved morphology. The method comprises the steps that a digitized image of a product is obtained; morphology operators are applied for detecting the outline of the digitized image to obtain a pixel outline rough extraction region; Canny operators are adopted for detecting the whole pixel-level edge of the product from the pixel outline rough extraction region; by means of Gaussian edge functions obtained through ideal edge points and diffusion function convolution, the whole pixel-level edge is fitted into a sub-pixel-level edge of the product. According to the method, the edge detection operators of the morphology are improved, the edge of the image outline can be smoothed, edge details are kept better, anti-noise performance is improved, image edge information is kept, the smoothness and the continuity of the edge are kept, the image edge can be detected accurately, the connectivity of an original image is ensured, an image edge extraction region is reduced, and the processing speed is increased.

The invention discloses a low-frequency vibration calibration method based on monocular vision. The method comprises the following steps that: utilizing a nonlinear camera model with radial distortionto realize high-accuracy camera calibration; putting forward an effective image enhancement method to realize the enhancement of blurred images of different motion directions to guarantee the position accuracy of an enhanced edge; then, on the basis of a sub pixel edge detection method of a Zernike moment, realizing the high-accuracy characteristic edge extraction of an enhanced sequence image; on the basis of a vision measurement model, measuring the spatial motion displacement of the characteristic edge; and utilizing the output voltage signals, which are synchronously collected, of a low-frequency vibration sensor and a low-frequency vibration measuring instrument to realize low-frequency vibration calibration. By use of the method, on a premise that calibration accuracy is effectivelyguaranteed, low-frequency vibration calibration can be stably, reliably and quickly realized. By use of the method, the deficiencies that an existing low-frequency vibration calibration method is limited in calibration accuracy, complex in process and complex in system for the low-frequency vibration sensor and the measuring instrument and is not suitable for wide-frequency range calibration areeliminated.

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 Legendre moment-based image subpixel edge detection method. The method includes the steps of: S1. reading image information, graying an image and performing de-noising processing on the grayscale image; S2. adopting a Sobel operator to perform pixel-level edge positioning on the de-noised image: using a phenomenon that a weighted value of an adjacent point of a pixel points reaches a maximum value at an edge point to perform edge detection; and S3. adopting Legendre moment to perform subpixel edge detection on the image, and outputting an edge image. The Sobel operator has a smoothing effect on noise, thereby providing relatively accurate edge direction information, the Legendre moment is used to perform subpixel edge detection, the number of templates required by operations is reduced, the complexity of calculation is lowered, and better robustness is achieved in noise-proof performance at the same time.

The invention discloses a metal workpiece surface defect image detection method. The method comprises the steps of: firstly, collecting a surface image of a metal workpiece through professional imaging equipment, and then enabling the collected image to be subjected to image early-stage preprocessing including the steps of uneven illumination image gray scale correction, image filtering, image threshold segmentation and the like; further performing feature extraction and analysis on the preprocessed image, and introducing a sub-pixel edge detection algorithm to perform edge detection on the metal workpiece; and finally, carrying out template matching on the template image and the measurement image by adopting a gray-level co-occurrence matrix algorithm so as to carry out defect detection on the surface of the metal workpiece. According to the metal workpiece surface defect image detection method provided by the invention, the traditional manual visual inspection is replaced by an automatic detection technology based on machine vision, the production efficiency is improved, the labor cost is reduced, the metal workpiece defect detection precision is improved by adopting a sub-pixeledge detection algorithm, and the production quality is optimized.

The invention discloses a method for accurate focal length measurement of an optical imaging system based on moon imaging. The method for accurate focal length measurement of the optical imaging system based on moon imaging comprises the steps that a moon image is obtained through the optical imaging system, the edge of the moon image is extracted based on the sub pixel edge detection algorithm, primary fitting is conducted by means of extracted edge information to enable a primary rough fitting disc to be obtained, fitting is conducted on the moon disc with the least square method and progressive optimization is conducted based on distance constraint to enable a high-precision sub-pixel-level moon disc to be obtained, and the focal length of the optical imaging system is calculated by means of the relative position geometrical relationship. Compared with the prior art, the method for accurate focal length measurement of the optical imaging system based on moon imaging has the advantages that the problems that during ground development, the measurement cost of a large-focal-length optical imaging system is high, a measurement device is complicated, the difficulty of focal length measurement of the optical imaging system is high during in-orbit operation, and the precision is limited are solved, the moon is fully used for imaging, the diameter of the moon disc is obtained through sub pixel extraction with the least square method, both a 'full moon' and a 'waning moon' can be used for focal length measurement, and the method is an economical and reliable focal length measurement method for large-focal-length optical imaging systems.

The invention discloses a device and method for monitoring the abrasion of a pantograph slide plate based on sub-pixel edge detection. The device comprises a field control and data acquisition unit, aremote transmission unit, a data analysis unit, a data storage and release unit and a client access unit. The method comprises the steps that firstly an industrial camera collects a half-pantograph image of the pantograph slide plate; secondly, filtering processing is performed on the image; then the processed image is processed by adopting an adaptive Canny edge detection algorithm to obtain upper and lower edges of the pantograph slide plate with whole pixels, and sub-pixel edge extraction is performed on the upper and lower edges; coordinates of a sub-pixel slide plate edge image are transformed into world coordinates through camera calibration, and a residual abrasion curve of the half-pantograph slide plate is obtained; and finally, a complete residual abrasion curve of the slide plate is obtained through curve integration, the minimum residual thickness of the abrasion region is obtained so as to judge whether it is necessary to replace the slide plate or not. The device adoptsa non-contact online measurement method, and has the advantages of good stability and high detection accuracy.

The invention relates to an automatic detection device for size of a used position of a surgical instrument and a detection method thereof. A video camera consists of a CMOS image sensor and a CMOS camera; the above of a detecting platform is fixedly connected with a helical-toothed upright post guide rail through a guide rail pedestal; a helical-toothed groove of the upright post guide rail is connected and fixed with a guide rail sliding block; the front of the guide rail sliding block is fixedly connected with a camera frame; the above of the camera frame is fixed with the video camera; the below of the video camera is fixed with a calibration gauge block fixture, the front of the calibration gauge block fixture is provided with a width detecting fixture, and the left side of the calibration gauge block fixture is provided with a height/thickness detecting fixture; an illuminator is placed on a platform under the video camera, and the center of a camera lens, the center of the calibration gauge block fixture, and the centers of the calibration gauge block and a detection position are in the same perpendicular line; and the video camera is connected with a computer through a USB 2.0 universal serial bus interface. The detection method comprises the following steps: 1, image acquisition; 2, system calibration; and 3, image processing, namely automatic positioning of the used position and edge detection of sub-pixel.

The invention discloses a method for detecting burrs of a circular injection molded part. The method comprises the following steps: S1, collecting image information of a product and preprocessing the image information; S2, based on the image information preprocessed in the S1, performing threshold segmentation processing on the image information to obtain image information of a circular target area; S3, on the basis of the image information of the circular target area obtained in the S2, determining the shapes and positions of a plurality of rough edge contours on the peripheral side of the circular target area; and S4, extracting contour feature information of each burr, and outputting and storing the contour feature information. Firstly, collected product image information is preprocessed to enhance image quality, then, threshold segmentation is performed on a target feature region of the product image information, then, the shape and the position of a burr are determined through a region morphology detection method and a sub-pixel edge detection method, and finally, the burr is extracted, output and stored. The invention can quickly extra blurr. Therefore, a large amount of human resources can be saved, the detection efficiency is high, and the detection precision is high.

The invention discloses a symmetry-based mobile phone glass cover plate window area defect detection method, and relates to mobile phone glass cover plate defect detection, which comprises the following steps of: calculating a defect enhanced image by using an image sequence shot by black and white periodic structured light, and then preprocessing and denoising. The ROI is extracted, contour extraction is carried out on the window area twice through a sub-pixel edge detection method, the contour extraction is a real contour edge and a standard contour edge for straight line fitting, and the symmetry axis is calculated. And detecting edge breakage and burrs according to the difference between the two contours. And performing subtraction operation on pixel points at corresponding positions on two sides of the symmetry axis in a spatial domain, and extracting a binary image only having a defect area by using difference information so as to detect defects on the surface of the window area. The detection method provided by the invention overcomes the problems that traditional template matching needs a standard template and cannot adapt to detection of various models, and can realize rapid and high-recognition-rate automatic detection of the defects of the window area of the mobile phone glass cover plate.

The invention relates to a grinding wheel abrasion detection method and system based on machine vision. The method comprises the following steps of: calibrating the internal and external parameters of a camera; placing a workpiece in a view range of a camera, and shooting a workpiece image; converting the acquired image into a grayscale image; converting the image into a binary image by using an Otsu threshold segmentation method; removing noise in the image by using median filtering; carrying out sub-pixel edge detection by using a zernike moment; determining a region of interest; performing curve fitting on the detected edge; and determining the abrasion condition of the grinding wheel. According to the invention, the method can quickly and efficiently detect the abrasion condition of the grinding wheel, judges whether the grinding wheel needs to be trimmed or not, and improves the grinding machining quality.

The invention discloses a 417 bar code identification method based on sub-pixel edge detection, which aims at defects of the traditional 417 bar code identification method such as slow processing, poor anti-interference performance, large error an the like. The 417 bar code identification method based on sub-pixel edge detection can identify 417 bar codes fast, accurately and efficiently, and is very remarkable in identification effect especially when image resolution ratio of the 417 bar codes is low. When the 417 bar codes are blurring in images or poor in printed quality, the 417 bar code identification method also has a certain effect on improvement of 417 bar code recognition rate. The 417 bar code identification method is further suitable for conditions of intercepted 417 bar codes. The 417 bar code identification method based on sub-pixel edge detection has good instantaneity, strong fouling resistance and high identification efficiency, and can be widely applied to smart phones and other terminal embedded devices.

A method for compensating for modulation transfer function of high-resolution satellite images is provided, which relates to the design of MTFC (Modulation Transfer Function Compensation) processing method for domestic high resolution satellite images. The invention solves the problems that the prior image compensation method leads to the degradation of the image signal-to-noise ratio, the noise amplification, the low quality of the restored image, and the high computational complexity in the restoration process. The invention improves from two stages of the point spread function calculation and the image deconvolution restoration. Compared with the traditional pixel-level edge detection method, the precision of sub-pixel edge detection is higher. The high precision can ensure the accuracyof the subsequent edge spread function, line spread function and even the final point spread function. In the process of image deconvolution restoration, for the GPU and multithreaded optimization ofthe maximum likelihood iterative deconvolution algorithm, the windowing radius is calculated according to the point spread function radius, and the ultra-high resolution remote sensing image is windowed and divided into blocks, which effectively removes the influence of boundary ringing and further improves the computational efficiency.

The invention relates to an automatic method for detecting the dimension of the use part of a surgical instrument. The detection method provided by the invention comprises the following steps: image acquisition, system calibration and image processing, wherein the image processing includes automatic use part positioning and sub-pixel edge detection.

The invention provides a sub-pixel edge detection method, comprising the steps of: acquiring an original gray image, and acquiring a pixel-level thick edge of the original gray image; indexing edge points of the pixel-level thick edge, and extending the edge by interpolation along a gradient direction; and performing sub-pixel detection on the extended edge points to obtain sub-pixel positions ofthe edge. The interval between edges in complex edges is extended by interpolating edge point neighborhoods based on the gradient direction, so that the precision of sub-pixel edge detection is greatly improved, and the sub-pixel level positioning capability on the complex edges is also significantly enhanced.

The invention provides an oil well pipe sleeve damage image edge detection method based on grey correlation analysis and Zernike moment. For an intercepted oil well pipe sleeve damage image, at first, a grey correlation analysis algorithm is adopted for locating image edges coarsely, and then a Zernike moment operator is utilized for achieving sub-pixel level location of the image edges. By means of a fast edge detecting method combining the grey correlation analysis with the Zernike moment, the continuous effective image edges can be detected through the grey correlation analysis, a large number of image details can be retained by adjusting a threshold value, and the applicability range is wide. Due to Zernike, in the sub-pixel edge detecting process, the number of templates needed by sub-pixel edge detecting can be effectively reduced, orders of a function are reduced, the capacity of resisting disturbance is improved, and the edge locating precision is improved. According to the method, at first, coarse locating is carried out for sleeve damage image edge detection through the grey correlation analysis, then the Zernike moment operator is utilized, and therefore the method has the advantages of being high in locating precision and good in noise immunity, and the sleeve damage image edges can be located accurately to the sub-pixel level.