CN111489389A - Light spot center detection method - Google Patents

Light spot center detection method Download PDF

Info

Publication number
CN111489389A
CN111489389A CN202010255471.2A CN202010255471A CN111489389A CN 111489389 A CN111489389 A CN 111489389A CN 202010255471 A CN202010255471 A CN 202010255471A CN 111489389 A CN111489389 A CN 111489389A
Authority
CN
China
Prior art keywords
light spot
edge
pixel
image
sub
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
CN202010255471.2A
Other languages
Chinese (zh)
Inventor
富容国
陈江南
周鸣
张红
杜振伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Science and Technology
Original Assignee
Nanjing University of Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing University of Science and Technology filed Critical Nanjing University of Science and Technology
Priority to CN202010255471.2A priority Critical patent/CN111489389A/en
Publication of CN111489389A publication Critical patent/CN111489389A/en
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Quality & Reliability (AREA)
  • Geometry (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a light spot center detection method, which comprises the following steps: collecting a light spot image, and carrying out denoising pretreatment on the image; roughly positioning the light spot by using an edge detection algorithm to obtain a pixel-level light spot edge; performing interpolation operation on the light spot edges of the pixel level to obtain sub-pixel level light spot edges; thinning the sub-pixel level light spot edge to obtain a thinned sub-pixel level light spot edge; and based on the thinned subpixel level light spot edge, calculating the light spot center by utilizing a Gaussian fitting principle. According to the method, the Canny edge detection operator and the interpolation method are combined, so that the Gaussian-fitted light spot edge is in a sub-pixel level, the edge information of the image can be retained to the maximum extent, the error positioning rate is reduced remarkably, the positioning precision is improved, a good positioning effect is obtained, the operation time is short, the real-time performance is high, and the requirement of industrial precision size measurement can be met.

Description

Light spot center detection method
Technical Field
The invention relates to the field of image processing, in particular to a light spot center detection method.
Background
In the visual detection system, laser measurement is applied to precise dimension measurement, non-contact measurement can be realized, the complex repeated calibration process of the measurement system in the traditional visual detection can be simplified, and the detection of the light spot center needs to be positioned to a sub-pixel level in order to further improve the precision of the measurement system.
The traditional light spot center detection method mainly comprises a gray scale gravity center method, a circle fitting algorithm and a Hough transformation method, wherein the gray scale gravity center method is a better method only under the condition that the light intensity of a light spot is uniform, otherwise, the light spot intensity has larger errors, the gravity center method is weak in anti-interference capacity, when interference light with certain intensity exists, the measurement result is often inaccurate, and therefore the measurement precision of the gravity center method is poor. However, due to the influence of edge noise, the circle fitting algorithm can cause the fitted curve to have a larger difference with the original light spot edge. The Hough transform method is not suitable for a real-time detection system and has low detection precision.
Disclosure of Invention
The invention aims to provide a high-precision light spot center detection method.
The technical solution for realizing the purpose of the invention is as follows: a method of spot center detection, the method comprising the steps of:
step 1, collecting a light spot image, and carrying out denoising pretreatment on the image;
step 2, roughly positioning the light spots by using an edge detection algorithm to obtain pixel-level light spot edges;
step 3, carrying out interpolation operation on the pixel-level light spot edge to obtain a sub-pixel-level light spot edge;
step 4, thinning the sub-pixel level light spot edge to obtain a thinned sub-pixel level light spot edge;
and 5, solving the center of the light spot by utilizing a Gaussian fitting principle based on the thinned subpixel level light spot edge.
Further, the denoising preprocessing in step 1 specifically adopts gaussian smoothing filtering.
Further, in step 2, the light spot is roughly positioned by using an edge detection algorithm to obtain a pixel-level light spot edge, and the implementation process is specifically realized by using an improved Canny edge detection algorithm and includes:
step 2-1, carrying out weighted average on the preprocessed light spot image by using four Sobel Gaussian convolution kernel templates to obtain the gradient amplitude and the direction of the light spot image, wherein the step comprises the following steps:
(1) calculating the X-direction partial derivative Px(x,y)：
Px(x,y)＝f(x,y)*Gx＝-f(x-1,y-1)+f(x+1,y-1)-2f(x-1,y)+2f(x+1,y)-f(x-1,y+1)+f(x+1,y+1)
Wherein f (x, y) is a pixel value at coordinate (x, y) in the spot image;
(2) calculating partial derivative P in Y directiony(x,y)：
Py(x,y)＝f(x,y)*Gy＝-f(x-1,y-1)-2f(x,y-1)-f(x+1,y-1)+f(x-1,y+1)+2f(x,y+1)+f(x+1,y+1)
(3) Calculating the 45 DEG directional derivative P45°(x,y)：
P45°(x,y)＝f(x,y)*G45°＝f(x,y-1)+2f(x+1,y-1)-f(x-1,y)+f(x+1,y)-2f(x-1,y+1)+f(x,y+1)
(4) Calculating the 135 DEG directional derivative P135°(x,y)：
P135°(x,y)＝f(x,y)*G135°＝-2f(x-1,y-1)-f(x,y-1)-f(x-1,y)+f(x+1,y)+f(x,y+1)+2f(x+1,y+1)
G abovex、Gy、G45°、G135°All are Sobel gaussian convolution kernel templates, which are respectively:
Figure BDA0002437138830000021
(5) calculating a horizontal direction difference P0(x,y)：
P0(x,y)＝Px+(P45°+P135°)/2
(6) Calculating the vertical difference P90(x,y)：
P90(x,y)＝Py+(P45°+P135°)/2
(7) Calculating the gradient amplitude P and the direction theta of the light spot image:
Figure BDA0002437138830000022
Figure BDA0002437138830000023
2-2, based on the gradient amplitude P and the direction theta of the light spot image, performing non-maximum suppression on the light spot image by using a 3 × 3 neighborhood template in 4 directions of horizontal, vertical, 45 degrees and 135 degrees;
step 2-3, setting two thresholds p1、p2And p is1＞p2The pixel value in the facula image after the non-maximum value is suppressed is higher than the threshold value p1And at two thresholds p1、p2And reserving the pixels in the middle, and removing the rest pixels to obtain the light spot edge of the pixel level.
Further, in step 3, a cubic spline interpolation method is specifically adopted for the interpolation operation of the pixel-level light spot edge.
Further, in the step 4, a morphological processing method is specifically adopted for performing the thinning processing on the sub-pixel level light spot edge.
Further, the step 5 of calculating the spot center by using the gaussian fitting principle includes:
the gray value distribution of the sub-pixel level light spot edge points is Gaussian distribution:
Figure BDA0002437138830000031
wherein P is the gradient amplitude obtained in step 2-1, σ is the variance of the Gaussian function, and (x)0,y0) Is the coordinate of the center of the light spot;
and 5-1, taking logarithm of the Gaussian distribution to obtain:
Figure BDA0002437138830000032
step 5-2, order
Figure BDA0002437138830000033
For pixels in the whole refined sub-pixel level light spot edge image, let:
Figure BDA0002437138830000034
in the formula, f (x)j,yj) As coordinates (x) in the sub-pixel level spot edge imagej,yj) Where j is 1,2, …, n, n is the total number of pixels in the sub-pixel level spot edge image;
then the system of equations is obtained according to the formula of step 5-1:
AB＝Y
the least squares solution to this system of equations is:
Figure BDA0002437138830000041
in the formula (I), the compound is shown in the specification,
Figure BDA0002437138830000042
represents the least squares solution of B, A+A generalized inverse matrix of matrix A;
step 5-3, solving the fitted spot center coordinates according to the least square solution of the step 5-2
Figure BDA0002437138830000043
Compared with the prior art, the invention has the following remarkable advantages: according to the method, the Canny edge detection operator and the interpolation method are combined, so that the Gaussian-fitted light spot edge is in a sub-pixel level, the edge information of the image can be retained to the maximum extent, the error positioning rate is reduced remarkably, the positioning precision is improved, a good positioning effect is obtained, the operation time is short, the real-time performance is high, and the requirement of industrial precision size measurement can be met.
The present invention is described in further detail below with reference to the attached drawing figures.
Drawings
Fig. 1 is a flowchart of a spot center detection method in one embodiment.
Fig. 2 is a schematic diagram of a spot image acquired in one embodiment.
Fig. 3 is a schematic diagram of a sub-pixel level spot edge after processing in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
In one embodiment, in conjunction with fig. 1, there is provided a spot center detection method, including the steps of:
step 1, collecting a light spot image, and carrying out denoising pretreatment on the image;
here, in order to reduce the influence of noise on the edge detection result as much as possible, it is necessary to filter out noise to prevent erroneous detection caused by the noise.
Step 2, roughly positioning the light spots by using an edge detection algorithm to obtain pixel-level light spot edges;
step 3, performing interpolation operation on the pixel-level light spot edge to obtain a sub-pixel-level light spot edge;
step 4, thinning the sub-pixel level light spot edge to obtain a thinned sub-pixel level light spot edge;
and 5, based on the thinned subpixel level light spot edge, solving the light spot center by utilizing a Gaussian fitting principle.
Further, in one embodiment, the denoising pre-processing in step 1 specifically employs gaussian smoothing filtering.
Here, the smoothed image is implemented by convolution with an image using a gaussian filter, and the gaussian smoothing function formula is expressed as follows:
Figure BDA0002437138830000051
after the original image f (x, y) is convolved with the gaussian filter, a smoothed image g (x, y) is obtained, and the process can be expressed as:
g(x,y)＝f(x,y)*H(x,y,σ)
where σ is the variance, the choice of gaussian convolution kernel size will affect the performance of the Canny detector, generally a gaussian convolution kernel of 5 × 5 is preferred:
Figure BDA0002437138830000052
further, in one embodiment, in step 2, the light spot is coarsely positioned by using an edge detection algorithm to obtain a pixel-level light spot edge, and the implementation is specifically implemented by using an improved Canny edge detection algorithm, and the implementation process includes:
step 2-1, carrying out weighted average on the preprocessed light spot image by using four Sobel Gaussian convolution kernel templates to obtain the gradient amplitude and the direction of the light spot image, wherein the step comprises the following steps:
(1) calculating the X-direction partial derivative Px(x,y)：
Px(x,y)＝f(x,y)*Gx＝-f(x-1,y-1)+f(x+1,y-1)-2f(x-1,y)+2f(x+1,y)-f(x-1,y+1)+f(x+1,y+1)
Wherein f (x, y) is a pixel value at coordinate (x, y) in the spot image;
(2) calculating partial derivative P in Y directiony(x,y)：
Py(x,y)＝f(x,y)*Gy＝-f(x-1,y-1)-2f(x,y-1)-f(x+1,y-1)+f(x-1,y+1)+2f(x,y+1)+f(x+1,y+1)
(3) Calculating the 45 DEG directional derivative P45°(x,y)：
P45°(x,y)＝f(x,y)*G45°＝f(x,y-1)+2f(x+1,y-1)-f(x-1,y)+f(x+1,y)-2f(x-1,y+1)+f(x,y+1)
(4) Calculate 135 degree squareDerivative P135°(x,y)：
P135°(x,y)＝f(x,y)*G135°＝-2f(x-1,y-1)-f(x,y-1)-f(x-1,y)+f(x+1,y)+f(x,y+1)+2f(x+1,y+1)
G abovex、Gy、G45°、G135°All are Sobel gaussian convolution kernel templates, which are respectively:
Figure BDA0002437138830000061
(5) calculating a horizontal direction difference P0(x,y)：
P0(x,y)＝Px+(P45°+P135°)/2
(6) Calculating the vertical difference P90(x,y)：
P90(x,y)＝Py+(P45°+P135°)/2
(7) Calculating the gradient amplitude P and the direction theta of the light spot image:
Figure BDA0002437138830000062
Figure BDA0002437138830000063
step 2-2, based on the gradient amplitude P and the direction theta of the light spot image, performing non-maximum suppression on the light spot image by using a 3 × 3 neighborhood template in 4 directions of horizontal, vertical, 45 degrees and 135 degrees, wherein specifically, on each point, a neighborhood center x is compared with two pixels in the corresponding gradient direction, if the center pixel is the maximum value, the non-maximum is retained, otherwise, the center is set to be 0, so that the non-maximum is suppressed, and the point with the maximum local gradient is retained to obtain a refined edge;
step 2-3, setting two thresholds p1、p2And p is1＞p2The pixel value in the facula image after the non-maximum value is suppressed is higher than the threshold value p1And at two thresholds p1、p2And reserving the pixels in the middle, and removing the rest pixels to obtain the pixel-level light spot edge.
Further, in one embodiment, the interpolation operation performed on the spot edge at the pixel level in step 3 specifically employs a cubic spline interpolation method.
Here, the quadratic function f (x, y) is continuous and differentiable over the interval [ a, b ], and the cubic spline interpolation function is defined as:
if the function S (x) satisfies:
(1) s (x) in each cell [ x ]i-1,xi]N) is a polynomial of not more than three degrees, where a is x0＜x1＜…＜xm＝b;
(2) S (x), S' (x), S "(x) are consecutive over the interval [ a, b ];
(3) if a function satisfies both of the conditions (1) and (2), the function is a cubic spline function of the sample points, and the interpolation condition S (x) is satisfiedj)＝yjWhere j is 0,1, …, n, s (x) may be called f (x, y) at node x0,x1,…,xnCubic spline interpolation function of (1).
Further, in one embodiment, the step 4 of refining the sub-pixel level light spot edge specifically adopts a morphological processing method.
Further, in one embodiment, the step 5 of finding the spot center by using the gaussian fitting principle includes:
the gray value distribution of the sub-pixel level light spot edge points is Gaussian distribution:
Figure BDA0002437138830000071
wherein P is the gradient amplitude obtained in step 2-1, σ is the variance of the Gaussian function, and (x)0,y0) Is the coordinate of the center of the light spot;
and 5-1, taking logarithm of the Gaussian distribution to obtain:
Figure BDA0002437138830000072
step 5-2, order
Figure BDA0002437138830000073
For pixels in the whole refined sub-pixel level light spot edge image, let:
Figure BDA0002437138830000074
in the formula, f (x)j,yj) As coordinates (x) in the sub-pixel level spot edge imagej,yj) Where j is 1,2, …, n, n is the total number of pixels in the sub-pixel level spot edge image;
then the system of equations is obtained according to the formula of step 5-1:
AB＝Y
the least squares solution to this system of equations is:
Figure BDA0002437138830000081
in the formula (I), the compound is shown in the specification,
Figure BDA0002437138830000082
represents the least squares solution of B, A+A generalized inverse matrix of matrix A;
step 5-3, solving the fitted spot center coordinates according to the least square solution of the step 5-2
Figure BDA0002437138830000083
As a specific example, the light spot image shown in fig. 2 is processed by the method of the present invention, and the result of the sub-pixel level light spot edge is shown in fig. 3, and the center coordinates of the light spot are obtained as (621.1215, 149.4846). The invention can reach the detection precision of sub-pixel level.
In conclusion, the Canny edge detection operator and the interpolation method are combined, so that the light spot edge of Gaussian fitting is in a sub-pixel level, the edge information of the image can be retained to the maximum extent, the error positioning rate is obviously reduced, the positioning precision is improved, a better positioning effect is obtained, the operation time is short, the real-time performance is high, and the requirement of industrial precision size measurement can be met.
The foregoing illustrates and describes the principles, general features, and advantages of the present invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.

Claims (6)

1. A method of detecting the center of a spot, the method comprising the steps of:
step 1, collecting a light spot image, and carrying out denoising pretreatment on the image;
step 2, roughly positioning the light spots by using an edge detection algorithm to obtain pixel-level light spot edges;
step 3, carrying out interpolation operation on the pixel-level light spot edge to obtain a sub-pixel-level light spot edge;
step 4, thinning the sub-pixel level light spot edge to obtain a thinned sub-pixel level light spot edge;
and 5, solving the center of the light spot by utilizing a Gaussian fitting principle based on the thinned subpixel level light spot edge.
2. The method for detecting the center of a light spot according to claim 1, wherein the denoising preprocessing in step 1 specifically employs gaussian smoothing filtering.
3. The method according to claim 1, wherein in step 2, the light spot is coarsely positioned by using an edge detection algorithm to obtain a pixel-level light spot edge, and the method is specifically implemented by using an improved Canny edge detection algorithm, and the implementation process includes:
step 2-1, carrying out weighted average on the preprocessed light spot image by using four Sobel Gaussian convolution kernel templates to obtain the gradient amplitude and the direction of the light spot image, wherein the step comprises the following steps:
(1) calculating the X-direction partial derivative Px(x,y)：
Px(x,y)＝f(x,y)*Gx＝-f(x-1,y-1)+f(x+1,y-1)-2f(x-1,y)+2f(x+1,y)-f(x-1,y+1)+f(x+1,y+1)
Wherein f (x, y) is a pixel value at coordinate (x, y) in the spot image;
(2) calculating partial derivative P in Y directiony(x,y)：
Py(x,y)＝f(x,y)*Gy＝-f(x-1,y-1)-2f(x,y-1)-f(x+1,y-1)+f(x-1,y+1)+2f(x,y+1)+f(x+1,y+1)
(3) Calculating the 45 DEG directional derivative P45°(x,y)：
P45°(x,y)＝f(x,y)*G45°＝f(x,y-1)+2f(x+1,y-1)-f(x-1,y)+f(x+1,y)-2f(x-1,y+1)+f(x,y+1)
(4) Calculating the 135 DEG directional derivative P135°(x,y)：
P135°(x,y)＝f(x,y)*G135°＝-2f(x-1,y-1)-f(x,y-1)-f(x-1,y)+f(x+1,y)+f(x,y+1)+2f(x+1,y+1)
G abovex、Gy、G45°、G135°All are Sobel gaussian convolution kernel templates, which are respectively:
Figure FDA0002437138820000021
(5) calculating a horizontal direction difference P0(x,y)：
P0(x,y)＝Px+(P45°+P135°)/2
(6) Calculating the vertical difference P90(x,y)：
P90(x,y)＝Py+(P45°+P135°)/2
(7) Calculating the gradient amplitude P and the direction theta of the light spot image:
Figure FDA0002437138820000022
Figure FDA0002437138820000023
2-2, based on the gradient amplitude P and the direction theta of the light spot image, performing non-maximum suppression on the light spot image by using a 3 × 3 neighborhood template in 4 directions of horizontal, vertical, 45 degrees and 135 degrees;
step 2-3, setting two thresholds p1、p2And p is1＞p2The pixel value in the facula image after the non-maximum value is suppressed is higher than the threshold value p1And at two thresholds p1、p2And reserving the pixels in the middle, and removing the rest pixels to obtain the pixel-level light spot edge.
4. The method according to claim 1, wherein the interpolation operation for the pixel-level spot edge in step 3 is specifically a cubic spline interpolation method.
5. The method for detecting the center of a light spot according to claim 1, wherein the step 4 of refining the edges of the light spot at the sub-pixel level specifically adopts a morphological processing method.
6. The method according to claim 1 or 3, wherein the step 5 of finding the spot center by using the Gaussian fitting principle comprises:
the gray value distribution of the sub-pixel level light spot edge points is Gaussian distribution:
Figure FDA0002437138820000024
wherein P is the gradient amplitude obtained in step 2-1, σ is the variance of the Gaussian function, and (x)0,y0) Is the coordinate of the center of the light spot;
and 5-1, taking logarithm of the Gaussian distribution to obtain:
Figure FDA0002437138820000031
step 5-2, order
Figure FDA0002437138820000032
For pixels in the whole refined sub-pixel level light spot edge image, let:
Figure FDA0002437138820000033
in the formula, f (x)j,yj) As coordinates (x) in the sub-pixel level spot edge imagej,yj) Where j is 1,2, …, n, n is the total number of pixels in the sub-pixel level spot edge image;
then the system of equations is obtained according to the formula of step 5-1:
AB＝Y
the least squares solution to this system of equations is:
Figure FDA0002437138820000034
in the formula (I), the compound is shown in the specification,
Figure FDA0002437138820000035
represents the least squares solution of B, A+A generalized inverse matrix of matrix A;
step 5-3, solving the fitted spot center coordinates according to the least square solution of the step 5-2
Figure FDA0002437138820000036
CN202010255471.2A 2020-04-02 2020-04-02 Light spot center detection method Withdrawn CN111489389A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010255471.2A CN111489389A (en) 2020-04-02 2020-04-02 Light spot center detection method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010255471.2A CN111489389A (en) 2020-04-02 2020-04-02 Light spot center detection method

Publications (1)

Publication Number Publication Date
CN111489389A true CN111489389A (en) 2020-08-04

Family

ID=71794550

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010255471.2A Withdrawn CN111489389A (en) 2020-04-02 2020-04-02 Light spot center detection method

Country Status (1)

Country Link
CN (1) CN111489389A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112686842A (en) * 2020-12-21 2021-04-20 苏州炫感信息科技有限公司 Light spot detection method and device, electronic equipment and readable storage medium
CN112767439A (en) * 2021-01-06 2021-05-07 中国人民解放军63863部队 Automatic detection and positioning method for light spot target
CN113192120A (en) * 2021-04-25 2021-07-30 无锡信捷电气股份有限公司 Circle positioning algorithm based on two-dimensional edge measurement and least square principle
CN116309510A (en) * 2023-03-29 2023-06-23 清华大学 Numerical control machining surface defect positioning method and device

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112686842A (en) * 2020-12-21 2021-04-20 苏州炫感信息科技有限公司 Light spot detection method and device, electronic equipment and readable storage medium
CN112686842B (en) * 2020-12-21 2021-08-24 苏州炫感信息科技有限公司 Light spot detection method and device, electronic equipment and readable storage medium
CN112767439A (en) * 2021-01-06 2021-05-07 中国人民解放军63863部队 Automatic detection and positioning method for light spot target
CN112767439B (en) * 2021-01-06 2022-11-18 中国人民解放军63863部队 Automatic detection and positioning method for light spot target
CN113192120A (en) * 2021-04-25 2021-07-30 无锡信捷电气股份有限公司 Circle positioning algorithm based on two-dimensional edge measurement and least square principle
CN116309510A (en) * 2023-03-29 2023-06-23 清华大学 Numerical control machining surface defect positioning method and device
CN116309510B (en) * 2023-03-29 2024-03-22 清华大学 Numerical control machining surface defect positioning method and device

Similar Documents

Publication Publication Date Title
CN111489389A (en) Light spot center detection method
CN108921176B (en) Pointer instrument positioning and identifying method based on machine vision
CN102999886B (en) Image Edge Detector and scale grating grid precision detection system
CN109784250B (en) Positioning method and device of automatic guide trolley
CN102156996A (en) Image edge detection method
CN110969656A (en) Airborne equipment-based laser beam spot size detection method
CN116468641A (en) Infrared blood vessel image enhancement processing method
CN115457063A (en) Method, device and equipment for extracting edge of circular hole of PCB (printed Circuit Board) and storage medium
CN112308872A (en) Image edge detection method based on multi-scale Gabor first-order derivative
CN116071272A (en) Image correction method and device, electronic equipment and storage medium thereof
CN112991374A (en) Canny algorithm-based edge enhancement method, device, equipment and storage medium
CN115527049A (en) High-precision measurement method for lead frame pin spacing
CN108416358B (en) Space target feature extraction method and device
CN110751156A (en) Method, system, device and medium for table line bulk interference removal
CN117853510A (en) Canny edge detection method based on bilateral filtering and self-adaptive threshold
CN117351063A (en) Method, device, equipment and storage medium for measuring micro-nano device size
CN116758423A (en) Power transmission line foreign matter detection method based on white point rate method
CN116563298A (en) Cross line center sub-pixel detection method based on Gaussian fitting
CN111553927A (en) Checkerboard corner detection method, checkerboard corner detection system, computer device and storage medium
CN110599510A (en) Picture feature extraction method
CN110298799B (en) PCB image positioning correction method
CN110490889B (en) Radar target extraction method based on edge detection
CN114025089A (en) Video image acquisition jitter processing method and system
CN112489065A (en) Checkerboard calibration fixed point sub-pixel extraction method
CN112651936A (en) Steel plate surface defect image segmentation method and system based on image local entropy

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WW01 Invention patent application withdrawn after publication
WW01 Invention patent application withdrawn after publication

Application publication date: 20200804