CN118096742B - Suction nozzle detection method of chip mounter based on corner detection and three-dimensional modeling technology - Google Patents
Suction nozzle detection method of chip mounter based on corner detection and three-dimensional modeling technology Download PDFInfo
- Publication number
- CN118096742B CN118096742B CN202410489220.9A CN202410489220A CN118096742B CN 118096742 B CN118096742 B CN 118096742B CN 202410489220 A CN202410489220 A CN 202410489220A CN 118096742 B CN118096742 B CN 118096742B
- Authority
- CN
- China
- Prior art keywords
- suction nozzle
- dimensional
- corner
- camera
- coordinates
- 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.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 35
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 34
- 238000011897 real-time detection Methods 0.000 claims abstract description 8
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 239000011159 matrix material Substances 0.000 claims description 54
- 238000013519 translation Methods 0.000 claims description 18
- 238000000034 method Methods 0.000 claims description 16
- 238000005316 response function Methods 0.000 claims description 12
- 238000012897 Levenberg–Marquardt algorithm Methods 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000003708 edge detection Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to a chip mounter suction nozzle detection method based on corner detection and three-dimensional modeling technology, which comprises the following steps: adopting a high-resolution industrial camera to acquire images of a working area of the chip mounter, and preprocessing the acquired images; detecting the corner of the suction nozzle in the image of the suction nozzle of the chip mounter based on a corner detection algorithm Shi-Tomasi; extracting feature descriptors from corner points of the suction nozzle, matching the feature descriptors, and finding out corresponding three-dimensional points; estimating the pose of the camera by adopting a PnP algorithm to obtain the position and the pose of the suction nozzle under a camera coordinate system; performing three-dimensional reconstruction to obtain a three-dimensional model of the suction nozzle; and combining the suction nozzle model and the attitude information obtained by three-dimensional reconstruction, and carrying out real-time detection and tracking of the suction nozzle by projecting the suction nozzle model and the attitude information into an actual image. The invention can effectively realize the accurate detection and tracking of the suction nozzle of the chip mounter and improve the reliability and stability.
Description
Technical Field
The invention relates to a chip mounter suction nozzle detection method based on corner detection and three-dimensional modeling technology.
Background
In the SMT production line, the chip mounter is the most important and complex equipment, and is also the equipment most prone to failure, and in the chip mounter failure, the first failure influencing factor is the suction nozzle component, how to solve the chip mounter suction nozzle failure, and accurate detection and tracking problem have the important effect on the normal production of the SMT production line.
Disclosure of Invention
In order to solve the problems, the invention aims to provide a chip mounter suction nozzle detection method based on corner detection and three-dimensional modeling technology, which can effectively realize accurate detection and tracking of the chip mounter suction nozzle and improve reliability and stability.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
a chip mounter suction nozzle detection method based on corner detection and three-dimensional modeling technology comprises the following steps:
Adopting a high-resolution industrial camera to acquire images of a working area of the chip mounter, and preprocessing the acquired images;
detecting the corner of the suction nozzle in the image of the suction nozzle of the chip mounter based on a corner detection algorithm Shi-Tomasi;
Extracting feature descriptors from corner points of the suction nozzle, extracting corresponding feature descriptors from three-dimensional modeling data of a suction nozzle model, matching corner point features in the suction nozzle image with features in the three-dimensional model by using a feature matching algorithm FLANN, and finding corresponding three-dimensional points;
Estimating the posture of the camera by adopting a PnP algorithm according to the angular points and the corresponding three-dimensional points of the suction nozzle in the image to obtain the position and the posture of the suction nozzle under a camera coordinate system;
converting the three-dimensional coordinates of the suction nozzle into a world coordinate system by combining an internal reference matrix and an external reference matrix of the camera, and carrying out three-dimensional reconstruction by combining a multi-view geometric method to obtain a three-dimensional model of the suction nozzle;
and combining the suction nozzle model and the posture information obtained by three-dimensional reconstruction, carrying out real-time detection by projecting the suction nozzle model and the posture information into an actual image, and carrying out real-time detection and tracking of the suction nozzle by comparing the position of the suction nozzle in the actual image with the projection position of the reconstructed model to judge whether the position deviation and the posture error exist in the posture of the suction nozzle.
Preferably, the preprocessing includes denoising and edge detection, specifically as follows: the image acquired by the industrial camera is expressed as a two-dimensional gray image I (x, y), wherein (x, y) represents pixel coordinates;
denoising the two-dimensional gray-scale image I (x, y) by adopting a Gaussian filter:
；
Wherein, Representing the pixel value after denoising,Representing the values of the surrounding pixels,The weight of the Gaussian kernel is represented, a and b are indexes of the Gaussian kernel and are used for traversing each element in the Gaussian kernel;
Extracting edge information in an image using sobel operator, including horizontal gradient Vertical gradient：
;
。
Preferably, the corner detection algorithm Shi-Tomasi detects the corner of the suction nozzle in the image of the suction nozzle of the chip mounter, and the specific steps are as follows: a. according to horizontal gradientsVertical gradientCalculating a gradient product matrix M:
；
then, the eigenvalues of the gradient product matrix M are calculated AndThen, calculating a corner response function:
；
b. For each pixel (x, y), defining a domain window E, and comparing the corner response function value R (x, y) of each pixel with the corner response function values of other pixel values in the domain, if R (x, y) is a local maximum in the domain, namely:
;
Wherein, Representing the offset of the pixel point in the adjacent area relative to the current pixel point;
The pixel point is reserved as a corner point, otherwise, the pixel point is restrained;
c. b, processing all the pixel points in the step b to obtain corner response function images, and obtaining corner coordinates of the suction nozzle.
Preferably, the feature descriptor extraction is performed on the corner of the suction nozzle, specifically: detecting key points in angular point coordinates of the suction nozzle by using a rapid segmentation test algorithm, and calculating sub-pixel level positions of the angular points around the detected key points;
calculating the main direction of each key point, and determining the main direction of the key point by using the Haar wavelet response direction;
descriptors of key points are generated using the BRIEF algorithm, the descriptors being represented by an N-dimensional binary string.
Preferably, a feature matching algorithm FLANN is used for matching the corner features in the suction nozzle image with features in the three-dimensional model, and corresponding three-dimensional points are found, specifically as follows:
Extracting corresponding feature descriptors from three-dimensional modeling data of a suction nozzle model, matching the corner feature descriptors in the suction nozzle image with feature descriptors in the three-dimensional model by using a quick library approximate nearest neighbor algorithm, and finding out corresponding relations between the corner feature descriptors and the feature descriptors;
according to the corresponding relation obtained by feature matching, according to the pixel coordinates of the reference matrix and the matching points in the industrial camera, calculating corresponding three-dimensional point coordinates by using a triangulation method:
setting two corresponding points in the image AndThey correspond to two points in three-dimensional spaceAnd;
Converting pixel coordinates into normalized plane coordinates through an industrial camera internal reference matrix K:
；
；
according to the external parameter matrix of the industrial camera Converting the normalized plane coordinates into coordinates in a world coordinate system:
;
；
Through the above conversion, the corresponding three-dimensional point coordinates are calculated.
Preferably, the estimating the pose of the camera by adopting a PnP algorithm according to the angular point and the corresponding three-dimensional point of the suction nozzle in the image, so as to obtain the position and the pose of the suction nozzle under the camera coordinate system, specifically as follows:
(1) Randomly selecting N matching pairs to carry out PnP algorithm solution, optimizing the camera gesture by minimizing the reprojection error, and obtaining initial camera gesture estimation:
The camera projection equation is expressed as:
；
Where s is the scaling factor and, Is the coordinates of the i-th point in the image,Is a corresponding three-dimensional coordinate; r is a rotation matrix; t is a translation vector;
The re-projection error is defined as the distance between the actual observation point and the estimated point:
；
Wherein, A two-dimensional matrix of image points,Is a three-dimensional space point matrix;
The goal is to minimize the re-projection error for all feature point pairs:
；
(2) Iteratively selecting and optimizing a rotation matrix R and a translation vector t of the camera based on a Levenberg-Marquardt algorithm until a stopping condition is met, so as to obtain an optimal rotation matrix R and a translation vector t;
(3) Calculating the final camera pose according to the optimal rotation matrix R and the translation vector t;
(4) And calculating the position and the posture of the suction nozzle in the three-dimensional space according to the final camera posture and the three-dimensional coordinates of the matching points.
Preferably, the method converts the three-dimensional coordinate of the suction nozzle into a world coordinate system by combining the camera internal reference matrix and the camera external reference matrix, and performs three-dimensional reconstruction by combining a multi-view geometric method to obtain a three-dimensional model of the suction nozzle, which is specifically as follows:
According to the internal reference matrix K of the camera and the optimal rotation matrix R and translation vector t, the two-dimensional coordinates of the suction nozzle under each camera view are obtained Converted into three-dimensional coordinates in a camera coordinate systemAnd converted to coordinates in the world coordinate system:
；
；
;
Wherein, 、AndThe weight, the optimal translation vector and the rotation matrix of the kth camera are respectively;
and carrying out three-dimensional reconstruction by utilizing the three-dimensional coordinate information of the suction nozzle under the multi-view to obtain a three-dimensional model of the suction nozzle.
The invention has the following beneficial effects:
1. the invention can effectively realize the accurate detection and tracking of the suction nozzle of the chip mounter, and improve the reliability and stability;
2. the invention combines the suction nozzle model and the gesture information obtained by three-dimensional reconstruction to realize the real-time detection and tracking of the suction nozzle, thereby realizing the positioning and gesture estimation of the suction nozzle in real time;
3. According to the invention, the camera gesture is estimated through the PnP algorithm, and three-dimensional reconstruction is performed by combining a multi-view geometric method, so that the position and the gesture of the suction nozzle under different coordinate systems are comprehensively considered, and the detection accuracy and stability are improved.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
The invention is described in further detail below with reference to the attached drawings and specific examples:
referring to fig. 1, in the present invention, there is provided a chip mounter suction nozzle detection method based on corner detection and three-dimensional modeling technology, including the steps of:
Adopting a high-resolution industrial camera to acquire images of a working area of the chip mounter, and preprocessing the acquired images;
detecting the corner of the suction nozzle in the image of the suction nozzle of the chip mounter based on a corner detection algorithm Shi-Tomasi;
Extracting feature descriptors from corner points of the suction nozzle, extracting corresponding feature descriptors from three-dimensional modeling data of a suction nozzle model, matching corner point features in the suction nozzle image with features in the three-dimensional model by using a feature matching algorithm FLANN, and finding corresponding three-dimensional points;
Estimating the posture of the camera by adopting a PnP algorithm according to the angular points and the corresponding three-dimensional points of the suction nozzle in the image to obtain the position and the posture of the suction nozzle under a camera coordinate system;
converting the three-dimensional coordinates of the suction nozzle into a world coordinate system by combining an internal reference matrix and an external reference matrix of the camera, and carrying out three-dimensional reconstruction by combining a multi-view geometric method to obtain a three-dimensional model of the suction nozzle;
and combining the suction nozzle model and the posture information obtained by three-dimensional reconstruction, carrying out real-time detection by projecting the suction nozzle model and the posture information into an actual image, and carrying out real-time detection and tracking of the suction nozzle by comparing the position of the suction nozzle in the actual image with the projection position of the reconstructed model to judge whether the position deviation and the posture error exist in the posture of the suction nozzle.
In this embodiment, the preprocessing includes denoising and edge detection, specifically as follows: the image acquired by the industrial camera is represented as a two-dimensional gray scale image I (x, y), where (x, y) represents the pixel coordinates,
Denoising the two-dimensional gray-scale image I (x, y) by adopting a Gaussian filter:
；
Wherein, Representing the pixel value after denoising,Representing the values of the surrounding pixels,The weight of the Gaussian kernel is represented, a and b are indexes of the Gaussian kernel and are used for traversing each element in the Gaussian kernel;
Extracting edge information in an image using sobel operator, including horizontal gradient Vertical gradient：
;
。
In this embodiment, the corner detection algorithm Shi-Tomasi is based on to detect the corner of the suction nozzle in the image of the suction nozzle of the chip mounter, specifically as follows: a. according to horizontal gradientsVertical gradientCalculating a gradient product matrix M:
；
then, the eigenvalues of the gradient product matrix M are calculated AndThen, calculating a corner response function:
；
b. For each pixel (x, y), defining a domain window E, and comparing the corner response function value R (x, y) of each pixel with the corner response function values of other pixel values in the domain, if R (x, y) is a local maximum in the domain, namely:
;
Wherein, Representing the offset of the pixel point in the adjacent area relative to the current pixel point;
The pixel point is reserved as a corner point, otherwise, the pixel point is restrained;
c. b, processing all the pixel points in the step b to obtain corner response function images, and obtaining corner coordinates of the suction nozzle.
In this embodiment, feature descriptor extraction is performed on the corner of the suction nozzle, specifically: detecting key points in angular point coordinates of the suction nozzle by using a rapid segmentation test algorithm, and calculating sub-pixel level positions of the angular points around the detected key points;
calculating the main direction of each key point, and determining the main direction of the key point by using the Haar wavelet response direction;
descriptors of key points are generated using the BRIEF algorithm, the descriptors being represented by an N-dimensional binary string.
Preferably, a feature matching algorithm FLANN is used for matching the corner features in the suction nozzle image with features in the three-dimensional model, and corresponding three-dimensional points are found, specifically as follows:
Extracting corresponding feature descriptors from three-dimensional modeling data of a suction nozzle model, matching the corner feature descriptors in the suction nozzle image with feature descriptors in the three-dimensional model by using a quick library approximate nearest neighbor algorithm, and finding out corresponding relations between the corner feature descriptors and the feature descriptors;
according to the corresponding relation obtained by feature matching, according to the pixel coordinates of the reference matrix and the matching points in the industrial camera, calculating corresponding three-dimensional point coordinates by using a triangulation method:
setting two corresponding points in the image AndThey correspond to two points in three-dimensional spaceAnd;
Converting pixel coordinates into normalized plane coordinates through an industrial camera internal reference matrix K:
；
；
according to the external parameter matrix of the industrial camera Converting the normalized plane coordinates into coordinates in a world coordinate system:
;
；
Through the above conversion, the corresponding three-dimensional point coordinates are calculated.
In this embodiment, according to the angular point and the corresponding three-dimensional point of the suction nozzle in the image, estimating the pose of the camera by adopting a PnP algorithm to obtain the position and the pose of the suction nozzle under the camera coordinate system, specifically as follows:
(1) Randomly selecting N matching pairs to carry out PnP algorithm solution, optimizing the camera gesture by minimizing the reprojection error, and obtaining initial camera gesture estimation:
The camera projection equation is expressed as:
；
Where s is the scaling factor and, Is the coordinates of the i-th point in the image,Is a corresponding three-dimensional coordinate; r is a rotation matrix; t is a translation vector;
The re-projection error is defined as the distance between the actual observation point and the estimated point:
；
Wherein, A two-dimensional matrix of image points,Is a three-dimensional space point matrix;
The goal is to minimize the re-projection error for all feature point pairs:
；
(2) Iteratively selecting and optimizing a rotation matrix R and a translation vector t of the camera based on a Levenberg-Marquardt algorithm until a stopping condition is met, so as to obtain an optimal rotation matrix R and a translation vector t;
(3) Calculating the final camera pose according to the optimal rotation matrix R and the translation vector t;
(4) And calculating the position and the posture of the suction nozzle in the three-dimensional space according to the final camera posture and the three-dimensional coordinates of the matching points.
In this embodiment, the combination of the camera reference matrix and the camera reference matrix converts the three-dimensional coordinates of the suction nozzle into the world coordinate system, and performs three-dimensional reconstruction by combining the multi-view geometric method to obtain a three-dimensional model of the suction nozzle, which is specifically as follows:
According to the internal reference matrix K of the camera and the optimal rotation matrix R and translation vector t, the two-dimensional coordinates of the suction nozzle under each camera view are obtained Converted into three-dimensional coordinates in a camera coordinate systemAnd converted to coordinates in the world coordinate system:
；
；
;
Wherein, 、AndThe weight, the optimal translation vector and the rotation matrix of the kth camera are respectively;
and carrying out three-dimensional reconstruction by utilizing the three-dimensional coordinate information of the suction nozzle under the multi-view to obtain a three-dimensional model of the suction nozzle.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the invention in any way, and any person skilled in the art may make modifications or alterations to the disclosed technical content to the equivalent embodiments. However, any simple modification, equivalent variation and variation of the above embodiments according to the technical substance of the present invention still fall within the protection scope of the technical solution of the present invention.
Claims (4)
1. A chip mounter suction nozzle detection method based on corner detection and three-dimensional modeling technology is characterized by comprising the following steps:
Adopting a high-resolution industrial camera to acquire images of a working area of the chip mounter, and preprocessing the acquired images;
detecting the corner of the suction nozzle in the image of the suction nozzle of the chip mounter based on a corner detection algorithm Shi-Tomasi;
Extracting feature descriptors from corner points of the suction nozzle, extracting corresponding feature descriptors from three-dimensional modeling data of a suction nozzle model, matching corner point features in the suction nozzle image with features in the three-dimensional model by using a feature matching algorithm FLANN, and finding corresponding three-dimensional points;
The feature descriptor extraction is carried out on the corner points of the suction nozzle, specifically: detecting key points in angular point coordinates of the suction nozzle by using a rapid segmentation test algorithm, and calculating sub-pixel level positions of the angular points around the detected key points; calculating the main direction of each key point, and determining the main direction of the key point by using the Haar wavelet response direction;
Generating descriptors of the key points by using BRIEF algorithm, wherein the descriptors are represented by an N-dimensional binary string;
Estimating the posture of the camera by adopting a PnP algorithm according to the angular points and the corresponding three-dimensional points of the suction nozzle in the image to obtain the position and the posture of the suction nozzle under a camera coordinate system;
converting the three-dimensional coordinates of the suction nozzle into a world coordinate system by combining an internal reference matrix and an external reference matrix of the camera, and carrying out three-dimensional reconstruction by combining a multi-view geometric method to obtain a three-dimensional model of the suction nozzle; the method comprises the steps of combining a suction nozzle model and posture information obtained through three-dimensional reconstruction, carrying out real-time detection by projecting the suction nozzle model and posture information into an actual image, carrying out real-time detection and tracking of a suction nozzle by comparing the position of the suction nozzle in the actual image with the projection position of the reconstructed model, and judging whether the position and posture of the suction nozzle have position deviation and posture error;
the preprocessing comprises denoising and edge detection, and specifically comprises the following steps: the image acquired by the industrial camera is represented as a two-dimensional gray scale image I (x, y), where (x, y) represents the pixel coordinates,
Denoising the two-dimensional gray-scale image I (x, y) by adopting a Gaussian filter:
；
Wherein, Representing the pixel value after denoising,Representing the values of the surrounding pixels,The weight of the Gaussian kernel is represented, a and b are indexes of the Gaussian kernel and are used for traversing each element in the Gaussian kernel;
Extracting edge information in an image using sobel operator, including horizontal gradient Vertical gradient：
;
；
The feature matching algorithm FLANN is used for matching the corner features in the suction nozzle image with features in the three-dimensional model, and corresponding three-dimensional points are found out, specifically as follows:
Extracting corresponding feature descriptors from three-dimensional modeling data of a suction nozzle model, matching the corner feature descriptors in the suction nozzle image with feature descriptors in the three-dimensional model by using a quick library approximate nearest neighbor algorithm, and finding out corresponding relations between the corner feature descriptors and the feature descriptors;
according to the corresponding relation obtained by feature matching, according to the pixel coordinates of the reference matrix and the matching points in the industrial camera, calculating corresponding three-dimensional point coordinates by using a triangulation method:
setting two corresponding points in the image AndThey correspond to two points in three-dimensional spaceAnd;
Converting pixel coordinates into normalized plane coordinates through an industrial camera internal reference matrix K:
；
；
according to the external parameter matrix of the industrial camera Converting the normalized plane coordinates into coordinates in a world coordinate system:
;
；
Through the above conversion, the corresponding three-dimensional point coordinates are calculated.
2. The chip mounter suction nozzle detection method based on the corner detection and three-dimensional modeling technology according to claim 1, wherein the corner detection algorithm Shi-Tomasi based on the corner detection algorithm detects the corner of the suction nozzle in the image of the chip mounter suction nozzle, specifically comprising the following steps:
a. According to horizontal gradients Vertical gradientCalculating a gradient product matrix M:
；
then, the eigenvalues of the gradient product matrix M are calculated AndThen, calculating a corner response function:
；
b. For each pixel (x, y), defining a domain window E, and comparing the corner response function value R (x, y) of each pixel with the corner response function values of other pixel values in the domain, if R (x, y) is a local maximum in the domain, namely:
;
Wherein, Representing the offset of the pixel point in the adjacent area relative to the current pixel point;
c. b, processing all the pixel points in the step b to obtain corner response function images, and obtaining corner coordinates of the suction nozzle.
3. The suction nozzle detection method of the chip mounter based on the corner detection and the three-dimensional modeling technology according to claim 1, wherein the method is characterized in that according to the corner point of the suction nozzle in the image and the corresponding three-dimensional point, the pose of the camera is estimated by adopting a PnP algorithm, and the position and the pose of the suction nozzle under a camera coordinate system are obtained, specifically as follows:
(1) Randomly selecting N matching pairs to carry out PnP algorithm solution, optimizing the camera gesture by minimizing the reprojection error, and obtaining initial camera gesture estimation:
The camera projection equation is expressed as:
；
Where s is the scaling factor and, Is the coordinates of the i-th point in the image,Is a corresponding three-dimensional coordinate; r is a rotation matrix; t is a translation vector;
The re-projection error is defined as the distance between the actual observation point and the estimated point:
；
Wherein, A two-dimensional matrix of image points,Is a three-dimensional space point matrix;
The goal is to minimize the re-projection error for all feature point pairs:
；
(2) Iteratively selecting and optimizing a rotation matrix R and a translation vector t of the camera based on a Levenberg-Marquardt algorithm until a stopping condition is met, so as to obtain an optimal rotation matrix R and a translation vector t;
(3) Calculating the final camera pose according to the optimal rotation matrix R and the translation vector t;
(4) And calculating the position and the posture of the suction nozzle in the three-dimensional space according to the final camera posture and the three-dimensional coordinates of the matching points.
4. The chip mounter suction nozzle detection method based on the corner detection and three-dimensional modeling technology according to claim 1, wherein the method is characterized in that the combination of the camera internal reference matrix and the external reference matrix converts the three-dimensional coordinates of the suction nozzle into a world coordinate system, and the three-dimensional reconstruction is performed by combining a multi-view geometric method, so as to obtain a three-dimensional model of the suction nozzle, and the method is specifically as follows:
According to the internal reference matrix K of the camera and the optimal rotation matrix R and translation vector t, the two-dimensional coordinates of the suction nozzle under each camera view are obtained Converted into three-dimensional coordinates in a camera coordinate systemAnd converted to coordinates in the world coordinate system:
；
；
;
Wherein, 、AndThe weight, the optimal translation vector and the rotation matrix of the kth camera are respectively;
and carrying out three-dimensional reconstruction by utilizing the three-dimensional coordinate information of the suction nozzle under the multi-view to obtain a three-dimensional model of the suction nozzle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410489220.9A CN118096742B (en) | 2024-04-23 | 2024-04-23 | Suction nozzle detection method of chip mounter based on corner detection and three-dimensional modeling technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410489220.9A CN118096742B (en) | 2024-04-23 | 2024-04-23 | Suction nozzle detection method of chip mounter based on corner detection and three-dimensional modeling technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118096742A CN118096742A (en) | 2024-05-28 |
| CN118096742B true CN118096742B (en) | 2024-07-05 |
Family
ID=91165616
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410489220.9A Active CN118096742B (en) | 2024-04-23 | 2024-04-23 | Suction nozzle detection method of chip mounter based on corner detection and three-dimensional modeling technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118096742B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104359402A (en) * | 2014-11-17 | 2015-02-18 | 南京工业大学 | Detection method for rectangular pin element visual positioning |
| CN106537908A (en) * | 2014-09-05 | 2017-03-22 | 英特尔公司 | Camera calibration |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3765888B2 (en) * | 1996-10-09 | 2006-04-12 | Ｊｕｋｉ株式会社 | Part position detection method |
| WO2013029675A1 (en) * | 2011-08-31 | 2013-03-07 | Metaio Gmbh | Method for estimating a camera motion and for determining a three-dimensional model of a real environment |
| CN104981105B (en) * | 2015-07-09 | 2018-07-13 | 广东工业大学 | A kind of quickly accurate detection and method for correcting error for obtaining element central and deflection angle |
| CN108717712B (en) * | 2018-05-29 | 2021-09-03 | 东北大学 | Visual inertial navigation SLAM method based on ground plane hypothesis |
| CN110933926B (en) * | 2019-11-13 | 2021-04-06 | 浙江工业大学 | Automatic correction method for angle of suction nozzle element of chip mounter based on angular point detection |
| CN112066879B (en) * | 2020-09-11 | 2021-11-19 | 哈尔滨工业大学 | Air floatation motion simulator pose measuring device and method based on computer vision |
| CN116645418A (en) * | 2023-05-30 | 2023-08-25 | 佛山显扬科技有限公司 | Screen button detection method and device based on 2D and 3D cameras and relevant medium thereof |
| CN117237493A (en) * | 2023-10-07 | 2023-12-15 | 江苏尚匠文化传播有限公司 | Three-dimensional cartoon based on topology analysis and edge detection fusion technology |
-
2024
- 2024-04-23 CN CN202410489220.9A patent/CN118096742B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106537908A (en) * | 2014-09-05 | 2017-03-22 | 英特尔公司 | Camera calibration |
| CN104359402A (en) * | 2014-11-17 | 2015-02-18 | 南京工业大学 | Detection method for rectangular pin element visual positioning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118096742A (en) | 2024-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108764048B (en) | Face key point detection method and device | |
| US11727661B2 (en) | Method and system for determining at least one property related to at least part of a real environment | |
| CN108805904B (en) | Moving ship detection and tracking method based on satellite sequence image | |
| US9418480B2 (en) | Systems and methods for 3D pose estimation | |
| CN106981077B (en) | Infrared image and visible light image registration method based on DCE and LSS | |
| US20150262346A1 (en) | Image processing apparatus, image processing method, and image processing program | |
| CN106971406B (en) | Object pose detection method and device | |
| CN107516322B (en) | Image object size and rotation estimation calculation method based on log polar space | |
| WO2012023593A1 (en) | Position and orientation measurement apparatus, position and orientation measurement method, and storage medium | |
| CN110222661B (en) | Feature extraction method for moving target identification and tracking | |
| KR101460313B1 (en) | Apparatus and method for robot localization using visual feature and geometric constraints | |
| CN112613123A (en) | AR three-dimensional registration method and device for aircraft pipeline | |
| JP2010157093A (en) | Motion estimation device and program | |
| CN112652020A (en) | Visual SLAM method based on AdaLAM algorithm | |
| CN110956664B (en) | Real-time repositioning method for camera position of handheld three-dimensional scanning system | |
| CN118096742B (en) | Suction nozzle detection method of chip mounter based on corner detection and three-dimensional modeling technology | |
| CN116090094B (en) | Hull thermal model building method, device and equipment based on infrared thermal imaging | |
| CN108694348B (en) | Tracking registration method and device based on natural features | |
| CN116309882A (en) | Tray detection and positioning method and system for unmanned forklift application | |
| CN116091998A (en) | Image processing method, device, computer equipment and storage medium | |
| Islam et al. | A new feature-based image registration algorithm | |
| Lee et al. | Visual odometry for absolute position estimation using template matching on known environment | |
| CN115222912A (en) | Target pose estimation method and device, computing equipment and storage medium | |
| CN111915632A (en) | Poor texture target object truth value database construction method based on machine learning | |
| Viswanath et al. | A simplified error model for height estimation using a single camera |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |