CN113674174B - Line scanning cylinder geometric correction method and device based on significant line matching - Google Patents
Line scanning cylinder geometric correction method and device based on significant line matching Download PDFInfo
- Publication number
- CN113674174B CN113674174B CN202110965505.1A CN202110965505A CN113674174B CN 113674174 B CN113674174 B CN 113674174B CN 202110965505 A CN202110965505 A CN 202110965505A CN 113674174 B CN113674174 B CN 113674174B
- Authority
- CN
- China
- Prior art keywords
- line
- image
- matching
- corrected
- significant
- 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
- 238000012937 correction Methods 0.000 title claims abstract description 52
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000005070 sampling Methods 0.000 claims abstract description 66
- 230000004044 response Effects 0.000 claims description 25
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 5
- 230000001131 transforming effect Effects 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 238000003384 imaging method Methods 0.000 description 10
- 238000000605 extraction Methods 0.000 description 7
- 230000007547 defect Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 239000000463 material Substances 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a line scanning cylinder geometric correction method and device based on significant line matching, wherein the method comprises the following steps: step S1, performing line scanning acquisition and unfolding on a cylindrical workpiece to be corrected by using a line scanning camera to serve as an image to be corrected, and acquiring a reference image; step S2, carrying out gradient projection on the reference image and the image to be corrected; step S3, selecting a salient line according to the gradient projection value; step S4, for each significant line, searching a candidate matching line corresponding to the significant line on the image to be corrected; step S5, sampling the matching line pairs to obtain q matching line pairs, and determining the optimal sampling matching line pairs through multiple iterations; step S6, taking the optimal sampling matching line pair as a reference, and distributing a matching line on the image to be corrected for each significant line on the reference image; and S7, converting the image to be corrected by using the matching line pair to obtain an image aligned with the reference image, thereby realizing geometric correction.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a line scanning cylinder geometric correction method and device based on significant line matching.
Background
In the industry, in order to detect surface defects in some cylindrical workpieces, imaging of their surface is required. Common applications are defect detection of surfaces such as lipsticks, cylindrical pencils, etc.
The main methods for imaging cylinders are currently: the method comprises the steps of placing a workpiece on a mechanism capable of enabling the workpiece to rotate, fixing a line scanning camera above a cylindrical workpiece, rotating the cylindrical workpiece at a constant speed through the mechanism, and sending signals to the line scanning camera to collect images. However, cylinder imaging acquired in this manner has two problems:
a. the starting position is different each time the cylinder is unfolded.
The reason is that: the position facing the line scanning camera corresponds to the unfolding starting position when the workpiece is placed, and in fact, the cylindrical workpiece is difficult to place at the same position every time.
b. There is a distortion of the cylindrical image in the direction of the expansion.
The reason is that: since the rotation angular velocity of the mechanism is not an absolute constant velocity, the image obtained by the line scan camera is deformed. In this way the cylinder deployment image is stretched or shortened to a different extent and the stretching is non-linear.
Based on these two problems, it is difficult to align cylindrical workpieces using conventional image matching techniques like planar workpiece imaging, and large-scale inspection using machine vision is not possible.
In order to solve the above problems, there is a technology for reducing image distortion by using ultra-high precision mechanism equipment to make its angular velocity more uniform, but the cost is too high to be popular.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention aims to provide a line scanning cylinder geometric correction method and device based on significant line matching, which realize the aim of aligning a cylinder image obtained by line scanning on a reference image based on significant line matching, so that the same batch of cylinder materials can be reasonably aligned, and the cylinder materials can be subjected to large-scale visual defect detection like plane materials.
To achieve the above and other objects, the present invention provides a line scan cylinder geometry correction method based on significant line matching, comprising the steps of:
step S1, acquiring an unfolded image of a perfect cylindrical workpiece by using a line scanning camera, preprocessing to obtain an ideal image as a reference image, and performing line scanning acquisition on the unfolded image of the cylindrical workpiece to be corrected by using the line scanning camera as an image to be corrected;
step S2, respectively carrying out gradient projection on the reference image and the image to be corrected;
s3, performing significant line selection on the reference image and the image to be corrected according to the gradient projection value;
step S4, for each significant line of the reference image, searching a candidate matching line corresponding to the significant line on the image to be corrected;
step S5, sampling the matched line pairs obtained in the step S4 to obtain q matched line pairs, correcting the ordinate of each significant line on the image to be corrected according to the significant line of the reference image of each matched line pair, searching whether significant lines exist in the corresponding range on the reference image according to the correction result, updating the confidence coefficient of the current group of sampled matched line pairs according to the search result, and determining the optimal sampled matched line pairs through multiple iterations and according to the confidence coefficient;
step S6, taking the optimal sampling matching line pair as a reference, and distributing a matching line on the image to be corrected for each significant line on the reference image;
and S7, converting the image to be corrected by using the obtained matching line pair to obtain an image aligned with the reference image, thereby realizing geometric correction.
Preferably, in step S1, if the side of the cylinder changes along the Y direction of the image, the width of the reference image is the same as the width of the image to be corrected, and the height is different; if the side surface of the cylinder is changed along the X direction of the image, the height of the reference image is the same as that of the image to be corrected, and the widths of the reference image and the image to be corrected are different.
Preferably, in step S1, the cylinder is expanded by default in the Y direction, and if the cylinder is expanded in the X direction, the expanded image is rotated by 90 degrees.
Preferably, in step S2, gradients are calculated along the Y direction of the image for the reference image and the image to be corrected, and horizontal projection is performed to obtain a column vector with a length equal to the height of the image.
Preferably, in step S3, a line with a gradient projection value greater than a preset threshold is taken as a significant line.
Preferably, in step S4, the i-th salient line S in the reference image is set to R (i) And the kth significant line S in the image to be corrected W (k) The matching response M (i, k) is calculated as:
according toThe matching response value of each salient line in the image to be corrected is used for selecting p salient lines with the smallest matching response as candidate matching lines, wherein R (S R (i) J) represents the ith salient row S of the reference image R R (i) The gray value of the upper j point, W (S W (k) J) represents the kth significant line S of the image W to be corrected W (k) The gray value of the upper j point.
Preferably, step S5 further comprises the steps of:
step S500, randomly extracting q different salient lines from the salient lines in the reference image, and sorting the salient lines from small to large according to the ordinate;
step S501, randomly extracting matching lines according to probability from p candidate matching lines of each extracted significant line;
step S502, for q salient lines in the reference image, if the ordinate of the ith salient line is smaller than the ordinate of the jth salient line (i, j e {1,2, …, q } and i+.j), the matching line drawn on the image to be corrected must also satisfy the condition, otherwise, the sampling is discarded and step S500 is re-executed;
step S503, correcting the ordinate of each significant line on the image to be corrected according to q matching line pairs obtained by sampling by taking the sampling line of the reference image as a reference, searching whether significant lines exist in the corresponding range on the reference image according to the correction result, and updating the confidence coefficient of the current group of sampling matching pairs according to the search result;
step S504, returning to step S500 for iteration to the set iteration times, and storing the matching line pair with the highest confidence as the optimal sampling matching line pair.
Preferably, in step S501, the following operations are performed:
step1: randomly generating 1 integer between 1 and p, wherein the integer corresponds to the selected candidate matching row;
step2, randomly generating 1 value with a value range of [0,1], and sampling one matching line from the current candidate matching line when the value is smaller than the probability of the candidate matching line; otherwise, step1 is re-executed until a matching line is selected that meets the condition.
Preferably, step S503 further includes:
after each group of sampling matching pairs is obtained, the confidence coefficient of the matching pair is initially 0;
correcting the ordinate of each significant line on the image to be corrected by taking the sampling behavior of the reference image as a benchmark;
after the corrected ordinate is obtained, searching whether a significant line exists in a corresponding range on the reference image, and if so, increasing the current confidence by 1.
Preferably, in step S6, each significant line position of the reference image is matched again to find the best corresponding line in the vicinity of the corresponding position of the image to be corrected according to the transformation.
In order to achieve the above object, the present invention further provides a line scan cylinder geometry correction device based on significant line matching, comprising:
the image acquisition processing unit is used for acquiring an unfolded image of a perfect cylindrical workpiece by using the line scanning camera, preprocessing the unfolded image to obtain an ideal image as a reference image, and performing line scanning acquisition on the unfolded image of the cylindrical workpiece to be corrected by using the line scanning camera to obtain an image to be corrected;
the gradient projection unit is used for respectively carrying out gradient projection on the reference image and the image to be corrected;
the salient line selection unit is used for selecting salient lines of the reference image and the image to be corrected according to the gradient projection values;
the rough matching unit is used for searching candidate matching lines corresponding to each significant line of the reference image on the image to be corrected;
the sampling correction optimizing unit is used for sampling the matching line pairs obtained through the coarse matching unit to obtain q matching line pairs, correcting the ordinate of each significant line on the image to be corrected according to the significant line of the reference image of each matching line pair, searching whether significant lines exist in the corresponding range on the reference image according to the correction result, updating the confidence coefficient of the current group of sampling matching line pairs according to the search result, and determining the optimal sampling matching line pairs according to the confidence coefficient through multiple iterations;
the fine matching unit is used for taking the optimal sampling matching line pair as a reference, and distributing a matching line on the image to be corrected for each significant line on the reference image;
and the geometric correction unit is used for transforming the image to be corrected by using the obtained matching line pair to obtain an image aligned with the reference image, thereby realizing geometric correction.
Compared with the prior art, the line scanning cylinder geometric correction method and device based on significant line matching are characterized in that a line scanning camera is utilized to obtain a reference image and an image to be corrected, gradient projection is carried out on the reference image and the image to be corrected respectively, significant line selection is carried out on the reference image and the image to be corrected according to gradient projection values, each significant line of the reference image is searched for a candidate matching line corresponding to the reference image to be corrected, the obtained matching line pairs are sampled to obtain q matching line pairs, the significant line reference of the reference image of each significant line pair is used for correcting the ordinate of each significant line on the image to be corrected, whether significant lines exist in a corresponding range on the reference image is searched for according to a correction result, the confidence level of the current group of sampling matching line pairs is updated according to the search result, the optimal sampling matching line pairs are determined according to the confidence level, one matching line pair is allocated on the image to be corrected for each significant line on the reference image, and finally the obtained matching line pairs are used for carrying out conversion on the image to be corrected, and the image to be matched with the same line on the reference image to obtain a cylinder image to be matched with the same visual scale, and the cylinder can be aligned with the reference image pair on the basis of the reference image to obtain a large-scale visual defect, and the cylinder can be aligned with the cylinder image to be reasonably detected.
Drawings
FIG. 1 is a flow chart of the steps of a method for geometric correction of a line scan cylinder based on significant line matching in accordance with the present invention;
FIG. 2 is a flow chart of a line scan cylinder geometry correction based on salient line matching in an illustrative embodiment of the present invention;
fig. 3 is a schematic structural diagram of a linear scanning cylinder geometry correction device based on significant line matching in the present invention.
Detailed Description
Other advantages and effects of the present invention will become readily apparent to those skilled in the art from the following disclosure, when considered in light of the accompanying drawings, by describing embodiments of the present invention with specific embodiments thereof. The invention may be practiced or carried out in other embodiments and details within the scope and range of equivalents of the various features and advantages of the invention.
Fig. 1 is a flow chart of steps of a line scan cylinder geometry correction method based on significant line matching according to the present invention, and fig. 2 is a flow chart of line scan cylinder geometry correction based on significant line matching according to an embodiment of the present invention. As shown in fig. 1 and 2, the line scan cylinder geometry correction method based on significant line matching of the present invention includes the following steps:
step S1, acquiring an unfolded image of a perfect cylindrical workpiece by using a line scanning camera, preprocessing to obtain an ideal image as a reference image, and performing line scanning acquisition on the unfolded image of the cylindrical workpiece to be corrected by using the line scanning camera as an image to be corrected.
In the invention, the line scanning camera is utilized to conduct line scanning acquisition on the cylindrical workpiece to obtain an unfolded image, the acquired image is the image to be corrected and is marked as W, and in the specific embodiment of the invention, in order to ensure that all information of the cylinder is acquired and excessive repetition is avoided, the cylindrical workpiece is acquired with data of more than 1.5 circles and less than 2 circles.
Likewise, a line scan camera is used to acquire an unfolded image of a perfectly cylindrical workpiece, and the finally desired ideal image is obtained as a reference image through manual translation, cutting and other operations and is marked as R. The reference image contains exactly the unfolded content of the cylindrical workpiece, and the image to be corrected contains the repeated cylindrical side content. The two are only different along the unfolding direction, but the imaging content is not changed. The invention aims to correct an image to be corrected by taking a reference image as a benchmark.
In the embodiment of the invention, if the side surface of the cylinder is changed along the Y direction of the image, the width of the reference image is the same as that of the image to be corrected, and the heights of the reference image and the image to be corrected are different; if the side surface of the cylinder is changed along the X direction of the image, the height of the reference image is the same as that of the image to be corrected, and the widths of the reference image and the image to be corrected are different.
For convenience of description, it is assumed in the present invention that the side of the cylinder is unfolded and then changed along the Y direction of the image, and heights of the reference image and the image to be corrected are respectively denoted as h R And h W The widths of the reference image and the image to be corrected are the same and are collectively denoted as w. The operation is similar when the cylinder side is unfolded and then changed in the X-direction of the image.
And S2, respectively carrying out gradient projection on the reference image and the image to be corrected.
Specifically, the reference image and the image to be corrected are respectively subjected to gradient calculation along the Y direction of the image, and horizontal projection is carried out to obtain column vectors with the length of image height.
Taking reference image R as an example, the i-th row gradient projection P R (i) The method comprises the following steps:
the gradient projection calculation method for the image to be corrected is the same and will not be described here.
In the invention, the cylinder is unfolded in the Y direction by default, if the cylinder is unfolded in the X direction, the unfolded image needs to be rotated by 90 degrees, so that no difference exists between the reference image and the image to be corrected in the X direction, and only the expansion difference exists in the Y direction.
And S3, performing significant line selection on the reference image and the image to be corrected according to the gradient projection value.
Because of the huge calculation amount of the progressive matching, the method can be used forThe presence of a large number of insignificant rows results in numerous mismatches. Thus, the present invention screens out relatively significant rows. In a specific embodiment of the present invention, the threshold h is manually set t The gradient projection value is larger than the threshold value h t As significant rows.
After the reference image and the image to be corrected have performed significant line selection, a group of display line numbers are respectively obtained and marked as S R And S is W The number of the elements is m and n respectively. The ordinate value for the ith salient line of the reference image is noted as y R (S R (i) The ordinate value of the j-th significant line of the image to be corrected is denoted as y W (S W (j) A kind of electronic device. The storage of salient lines needs to be ordered from small to large on the ordinate.
And S4, for each significant line of the reference image, searching a candidate matching line corresponding to the significant line on the image to be corrected.
The rough matching is performed in the step, namely, each salient line of the reference image is matched with each salient line of the image to be corrected, and p salient lines with the smallest matching response are searched on the image to be corrected and serve as candidate matching lines. Specifically, the ith salient row S in the reference image R (i) And the kth significant line S in the image to be corrected W (k) The matching response M (i, k) is calculated as:
R(S R (i) J) represents the ith salient row S of the reference image R R (i) The gray value of the upper j point, W (S W (k) J represents the kth significant line S of the image W to be corrected W (k) The gray value of the upper j point,
and selecting p significant lines with the smallest matching response as candidate matching lines according to the matching response values of the significant lines of the reference image and the significant lines in the image to be corrected, wherein a parameter p is set by a user, generally taking 2-4, normalizing the p matching responses (dividing by the total sum) to make the sum 1, and subtracting the p normalized values from 1 to obtain the matching probability of each candidate matching line. In the present invention, the 1 reduction is used to ensure that the smaller the matching response, the greater the probability. The following are illustrated: assuming that the i-th significant line on the reference image, the 2 (p=2) smallest matching response behaviors a and b on the image to be corrected have specific response values of M (i, a) =100, and M (i, b) =300, then after normalization, the two values are 100/(100+300) = 0.25,300/(100+300) =0.75, so the matching probability is 1-0.25= 0.75,1-0.75=0.25, so the matching probability of the matching response line a is 0.75, and the matching probability of b is 0.25, which means that the matching probability of the line a on the image to be corrected and the i-line of the reference image is very high, and the probability is 0.75.
And S5, sampling the matched line pairs obtained in the step S4 to obtain q matched line pairs, and correcting the ordinate of the corresponding significant line on the image to be corrected according to the significant line standard of the reference image of each matched line pair.
After the rough matching in step S4, each significant line of the reference image and the corresponding candidate matching line in the image to be corrected form a matching line pair, and the matching line pair obtained by the rough matching has different degrees of incorrect matching, but a large number of correct matching line pairs also exist, so that the reliability of each matching line pair needs to be further determined. In the specific embodiment of the invention, the matching line pairs are sampled in a random sampling mode, and the reliability of the matching line pairs is judged. One sampling needs to be performed for a plurality of times of randomization until a proper matching row pair is extracted, and the specific steps of step S5 are as follows:
step S500, randomly extracting q different salient lines from the salient lines in the reference image as sampling lines, and sorting from small to large according to the ordinate. The parameter q is set by the user, the value of which is related to the significant line actually extracted and the imaging quality. When the number of the significant lines is large and the imaging quality is good, the number of the significant lines can be appropriately set to be large, otherwise, the number of the significant lines needs to be small, and the range of the significant lines is usually 5-30.
Step S501, for each extracted sampling line, randomly extracting a matching line from p candidate matching lines according to probability, which specifically includes:
step1: randomly generating 1 integer, wherein the value range is 1,2, … and p, and the value corresponds to the selected candidate matching row;
step2, randomly generating 1 value, wherein the value range is [0,1], and when the value is smaller than the matching probability of the candidate matching row, sampling one matching row from the current candidate matching row; otherwise, step1 is re-executed until a matching line is selected that meets the condition.
Step S502, verifying sequence consistency. For q significant lines (sampling lines) in the reference image, if the ordinate of the ith significant line is smaller than the ordinate of the jth significant line (i, j e {1,2, …, q } and i+.j), then the matching line drawn on the image to be corrected must also satisfy this condition, otherwise the sampling is discarded and step S500 is re-executed.
After the sample matching operation is performed, q matching line pairs are obtained.
Step S503, correcting the ordinate of each significant line on the image to be corrected by taking the sampling line of the reference image as a standard according to the q matching line pairs obtained by sampling, searching whether the significant line exists in the corresponding range on the reference image according to the correction result, and updating the confidence coefficient of the current group of sampling matching pair according to the search result.
Specifically, after each set of sample matching pairs (q matching line pairs are one), the confidence of the set of sample matching pairs is initially 0. Assume q matched line pairs obtained by sampling, wherein the squat marks of q salient lines on a reference image areThe longitudinal sitting mark of the matching line corresponding to the image W to be corrected is +.>The ordinate of each salient line on the image to be corrected is corrected (i.e. all salient lines are used to verify whether the currently selected set of matching line pairs is reasonable) based on the sampled line (salient line) of the reference image. In the j-th significant line S W (j) For example, its original ordinate is y W (S W (j) The corrected ordinate is +.>
a. When (when)In the time-course of which the first and second contact surfaces,
b. when (when)When k=1, 2, …, q-1
c. When (when)Time of day
After the corrected ordinate is obtainedAfter that, the corresponding range on the reference image +.> If so, the current confidence is increased by 1, where r t For the parameters set by the user, usually 2-5,r are taken t The smaller the value, the more strict the determination, r t The larger the value, the more relaxed the decision.
Step S504, returning to step S500, iterating to a set iteration number (e.g. 100), and storing a set of matching line pairs with the greatest confidence as the optimal sampling matching line pairs, i.e. selecting a set of best matching groups from all matching groups, each group being q matching line pairs.
And S6, taking the optimal sampling matching line pair as a reference, and distributing a matching line on the image to be corrected for each significant line on the reference image.
Assuming that the optimal sampling matching line pair is obtained through iterative optimization, wherein the longitudinal sitting marks of q optimal matching lines on the reference image are as followsThe squat marks of the corresponding optimal matching row on the image W to be corrected are as followsAnd (3) matching each significant line position of the reference image again to find the best corresponding line nearby the corresponding position of the image to be corrected according to the position of the corresponding position of the image to be corrected obtained through transformation. With the ith salient row S on the reference image R (i) For example, its original ordinate is y R (S R (i) On the image to be corrected, the ordinate corresponding thereto is +.>(note that here, instead of operating on the original reference image, the coordinates on the reference image are transformed to the coordinates on the image to be corrected, so its subscript is set to the reference image R, and the superscript is set to the image to be corrected W):
a. when (when)In the time-course of which the first and second contact surfaces,
b. when (when)When k=1, 2, …, q-1
c. When (when)Time of day
For salient lines S on a reference image R (i) Obtaining the corresponding position of the image on the image to be correctedAfter that, in the range->Internal re-matching search and S R (i) The corresponding optimal matching line, let k beIntegers in the range, matching response value is +.> The line with the smallest response value is used as the salient line S of the reference image R (i) In the corresponding line of the image to be corrected, the description is denoted as S 'for convenience' W (i) With an ordinate y W (S′ W (i))。
And S7, converting the image to be corrected by using the obtained matching line pair to obtain an image aligned with the reference image, thereby realizing geometric correction.
In the present invention, the position u' of the u-th line of the aligned image C on the image W to be corrected is:
a. when u is<y R (S R (1) At the time of the transfer of the sample),
the row u' corresponding to the 1 st and 2 nd significant rows on the image W to be corrected is presumed by using the proportional relation of the ordinate positions of the rows:
wherein u' -y W (S′ W (1) I) represents the u' th row to y on the image W W (S′ W (1) Distance of row; u-y R (S R (1) A) represents the ith row through the (y) th row on the image R R (S R (1) Distance of row; y is W (S′ W (1))-y W (S′ W (2) Y) represents the y-th on the image W W (S′ W (1) Line (y) W (S′ W (2) Distance of row, y R (S R (1))-y R (S R (2) Y) represents the y-th on the image R R (S R (1) Line (y) R (S R (2) The present invention obtains the above equation in proportion to the distance between adjacent lines.
b. When y is R (S R (i))≤u<y R (S R (i+1)), i=1, 2, … q-1
The row u' corresponding to the ith and (i+1) th significant rows on the image W to be corrected is presumed by using the proportional relation of the ordinate positions of the rows:
correspondingly, u' -y W (S′ W (i) I) represents the u' th row to y on the image W W (S′ W (i) Distance of row; u-y R (S R (i) A) represents the ith row through the (y) th row on the image R R (S R (i) Distance of row; y is W (S′ W (i))-y W (S′ W (i+1)) represents the y-th on the image W W (S′ W (i) Line (y) W (S′ W Distance of (i+1)) row, y R (S R (i))-y R (S R (i+1)) represents the y-th on the image R R (S R (i) Line (y) R (S R (i+1)) row distance.
c. When u is greater than or equal to y R (S R (q)) in the case of (a) the process,
the row u' corresponding to the q-1 and the q-th significant row on the image W to be corrected is presumed by using the proportional relation of the ordinate positions of the q-1 and the q-th significant rows:
correspondingly, u' -y W (S′ W (q)) represents the (u' th row to y) on the image W W (S′ W (q)) distance of rows; u-y R (S R (q)) represents the (u) th line to the (y) th line on the image R R (S R (q)) distance of rows; y is W (S′ W (q))-y W (S′ W (q-1)) represents the y-th on the image W W (S′ W (q)) line to y W (S′ W (q-1)) distance of row, y R (S R (q))-y R (S R (q-1)) represents the y-th on the image R R (S R (q)) line to y R (S R (q-1)) row distance.
When the correspondence is obtained, the line u' may be a decimal value, and the gray values of all the data of the u th line in the aligned image C need to be obtained through linear interpolation:
wherein, the liquid crystal display device comprises a liquid crystal display device,representing +.>V is the gray value of the horizontal coordinate and is 1,2, …, w and w are the widths of the reference image and the image to be corrected.
Fig. 3 is a schematic structural diagram of a linear scanning cylinder geometry correction device based on significant line matching in the present invention. As shown in fig. 3, the linear scan cylinder geometry correction device based on significant line matching of the present invention comprises:
the image acquisition processing unit 201 is configured to acquire an expanded image of a perfectly-intact cylindrical workpiece by using a line scan camera, perform preprocessing to obtain an ideal image as a reference image, and perform line scan acquisition of the expanded image of the cylindrical workpiece to be corrected by using the line scan camera as an image to be corrected.
In the present invention, the image acquisition processing unit 201 performs line scanning to the cylindrical workpiece by using the line scanning camera to acquire an expanded image, where the acquired image is an image to be corrected, denoted by W, and in the specific embodiment of the present invention, in order to ensure that all information of the cylinder is acquired and avoid excessive repetition, data of more than 1.5 circles and less than 2 circles is acquired for the cylindrical workpiece.
Likewise, the image acquisition processing unit 201 acquires an unfolded image of a perfectly cylindrical workpiece by using a line scan camera, and obtains a final desired ideal image as a reference image, denoted as R, through operations such as manual translation and clipping. The reference image contains exactly the unfolded content of the cylindrical workpiece, and the image to be corrected contains the repeated cylindrical side content. The two are only different along the unfolding direction, but the imaging content is not changed. The invention aims to correct an image to be corrected by taking a reference image as a benchmark.
In the embodiment of the invention, if the side surface of the cylinder is changed along the Y direction of the image, the width of the reference image is the same as that of the image to be corrected, and the heights of the reference image and the image to be corrected are different; if the side surface of the cylinder is changed along the X direction of the image, the height of the reference image is the same as that of the image to be corrected, and the widths of the reference image and the image to be corrected are different.
For convenience of description, it is assumed in the present invention that the side of the cylinder is unfolded and then changed along the Y direction of the image, and heights of the reference image and the image to be corrected are respectively denoted as h R And h W The widths of the reference image and the image to be corrected are the same and are collectively denoted as w. The operation is similar when the cylinder is unfolded laterally and then changed along the X-direction of the image
And the gradient projection unit 202 is used for respectively carrying out gradient projection on the reference image and the image to be corrected.
Specifically, the gradient projection unit 202 calculates gradients of the reference image and the image to be corrected along the Y direction of the image, and performs horizontal projection to obtain a column vector with a length of image height.
Taking reference image R as an example, the i-th row gradient projection P R (i) The method comprises the following steps:
the gradient projection calculation method for the image to be corrected is the same and will not be described here.
In the invention, the cylinder is unfolded in the Y direction by default, if the cylinder is unfolded in the X direction, the unfolded image needs to be rotated by 90 degrees, so that no difference exists between the reference image and the image to be corrected in the X direction, and only the expansion difference exists in the Y direction.
A salient line selection unit 203, configured to perform salient line selection on the reference image and the image to be corrected according to gradient projection values.
Since not only is the line-by-line matching computationally intensive, but also numerous mismatches can occur due to the large number of insignificant lines present. Thus, the present invention screens out relatively significant rows. In a specific embodiment of the present invention, the threshold h is manually set t The gradient projection value is larger than the threshold value h t As significant rows.
After the reference image and the image to be corrected have performed significant line selection, a group of display line numbers are respectively obtained and marked as S R And S is W The number of the elements is m and n respectively. The ordinate value for the ith salient line of the reference image is noted as y R (S R (i) The ordinate value of the j-th significant line of the image to be corrected is denoted as y W (S W (j) A kind of electronic device. The storage of salient lines needs to be ordered from small to large on the ordinate.
A coarse matching unit 204, configured to, for each significant line of the reference image, find a candidate matching line corresponding to the significant line on the image to be corrected.
Specifically, the rough matching unit 204 matches each salient line of the reference image with each salient line of the image to be corrected, and searches for p salient lines with the smallest matching response on the image to be corrected as candidate matching lines. Specifically, the ith salient row S in the reference image R (i) And the kth significant line S in the image to be corrected W (k)The matching response M (i, k) is calculated as:
thus, for each salient line of the reference image, selecting p salient lines with smallest matching responses as candidate matching lines according to the matching response values of the salient lines and each salient line in the image to be corrected, wherein a parameter p is set by a user, generally taking 2-4, normalizing the p matching responses (dividing by the total sum) to make the sum 1, and subtracting the p normalized values from 1 to obtain the matching probability of each candidate matching line. In the present invention, the 1 reduction is used to ensure that the smaller the matching response, the greater the probability.
The sampling correction optimizing unit 205 is configured to sample the matching line pairs obtained by the coarse matching unit 204 to obtain q matching line pairs, correct the ordinate of each significant line on the image to be corrected according to the significant line of the reference image of each matching line pair, find whether there is a significant line in the corresponding range on the reference image according to the correction result, update the confidence coefficient of the current set of sampling matching line pairs according to the search result, and determine the optimal sampling matching line pair according to the confidence coefficient through multiple iterations.
When the rough matching unit 204 performs rough matching, each significant line of the reference image and the corresponding candidate matching line in the image to be corrected form a matching line pair, and the matching line pair obtained by the rough matching has different degrees of error matching, but also has a large number of correct matching line pairs, so that the reliability of each matching line pair needs to be further determined. In the specific embodiment of the invention, the matching line pairs are sampled in a random sampling mode, and the reliability of the matching line pairs is judged. One sampling requires performing a plurality of randomizations until a suitable matching pair of rows is extracted, the sampling correction optimizing unit 205 includes:
the reference image salient line extraction module is used for randomly extracting q different salient lines from the salient lines in the reference image and sequencing the q different salient lines from small to large according to the ordinate. The parameter q is set by the user, the value of which is related to the significant line actually extracted and the imaging quality. When the number of the significant lines is large and the imaging quality is good, the number of the significant lines can be appropriately set to be large, otherwise, the number of the significant lines needs to be small, and the range of the significant lines is usually 5-30.
The candidate matching line extraction module is used for extracting matching lines from p candidate matching lines according to probability for each extracted significant line, and the candidate matching line extraction module specifically operates as follows:
step1: randomly generating 1 integer, wherein the value range is 1,2, … and p, and the value corresponds to the selected candidate matching row;
step2, randomly generating 1 value, wherein the value range is [0,1], and when the value is smaller than the matching probability of the candidate matching row, sampling one matching row from the current candidate matching row; otherwise, step1 is re-executed until a matching line is selected that meets the condition.
The sequential consistency verification module is configured to perform sequential consistency verification on the extraction result of the candidate matching line extraction module, specifically, for q significant lines in the reference image, if the ordinate of the ith significant line is smaller than the ordinate of the jth significant line (i, j∈ {1,2, …, q } and i+.j), the matching line extracted on the image to be corrected must also satisfy the condition, otherwise, the current sample is discarded and returned to the reference image significant line extraction module.
After the sample matching operation is performed, q matching line pairs are obtained.
The significant line correction module is used for correcting the ordinate of each significant line on the image to be corrected by taking the sampling line of the reference image as a standard according to q matching line pairs obtained by sampling, searching whether the significant line exists in the corresponding range on the reference image according to the correction result, and updating the confidence coefficient of the current group of sampling matching pairs according to the search result.
Specifically, after each set of sample matching pairs is obtained, the confidence of the set of sample matching pairs is initially 0. Assume q matched line pairs obtained by sampling, wherein the squat marks of q salient lines on a reference image areOn the image W to be correctedThe longitudinal sitting mark of the corresponding matching row is +.>The ordinate of each significant line on the image to be corrected is corrected with reference to the sampling line (significant line) of the reference image. In the j-th significant line S W (j) For example, its original ordinate is y W (S W (j) The corrected ordinate is +.>
a. When (when)In the time-course of which the first and second contact surfaces,
b. when (when)When k=1, 2, …, q-1
c. When (when)Time->
After the corrected ordinate is obtainedAfter that, the corresponding range on the reference image +.> Internal search for presence of salient linesIf so, the current confidence level is increased by 1, where r t For the parameters set by the user, usually 2-5,r are taken t The smaller the value, the more strict the determination, r t The larger the value, the more relaxed the decision.
And the iteration execution module returns the reference image salient line extraction module to iterate to a set iteration number (for example, 100), and stores a group of matching line pairs with the highest confidence as optimal sampling matching line pairs, namely, a group of best matching groups is selected from all matching groups, and each group is q matching line pairs.
A fine matching unit 206, configured to assign a matching line on the image to be corrected for each significant line on the reference image with the optimal sampling matching line pair as a reference.
The sampling correction optimizing unit is assumed to obtain an optimal sampling matching line pair through iterative optimization, wherein the longitudinal sitting marks of q optimal matching lines on a reference image are as followsThe ordinate mark of the optimally matched line corresponding to the image W to be corrected is +.>And (3) matching each significant line position of the reference image again to find the best corresponding line nearby the corresponding position of the image to be corrected according to the position of the corresponding position of the image to be corrected obtained through transformation. With the ith salient row S on the reference image R (i) For example, its original ordinate is y R (S R (i) On the image to be corrected, the ordinate corresponding thereto is +.>
a. When (when)In the time-course of which the first and second contact surfaces,
b. when (when)When k=1, 2, …, q-1
c. When (when)Time of day
For salient lines S on a reference image R (i) Obtaining the corresponding position of the image on the image to be correctedAfter that, in the range->Internal re-matching search and S R (i) The corresponding optimal matching line, let k beIntegers in the range, matching response value is +.> The line with the smallest response value is used as the salient line S of the reference image R (i) In the corresponding line of the image to be corrected, the description is denoted as S 'for convenience' W (i) With an ordinate y W (S′ W (i))。
A geometric correction unit 207 for transforming the image to be corrected with the obtained matching line pair to obtain an image aligned with the reference image, thereby realizing geometric correction.
In the present invention, the position u' of the u-th line of the aligned image C on the image W to be corrected is:
a. when u is<y R (S R (1) At the time of the transfer of the sample),
the row u' corresponding to the 1 st and 2 nd significant rows on the image W to be corrected is presumed by using the proportional relation of the ordinate positions of the rows:
wherein u' -y W (S′ W (1) I) represents the u' th row to y on the image W W (S′ W (1) Distance of row; u-y R (S R (1) A) represents the ith row through the (y) th row on the image R R (S R (1) Distance of row; y is W (S′ W (1))-y W (S′ W (2) Y) represents the y-th on the image W W (S′ W (1) Line (y) W (S′ W (2) Distance of row, y R (S R (1))-y R (S R (2) Y) represents the y-th on the image R R (S R (1) Line (y) R (S R (2) The present invention obtains the above equation in proportion to the distance between adjacent lines.
b. When y is R (S R (i))≤u<y R (S R (i+1)), i=1, 2, … q-1
The row u' corresponding to the ith and (i+1) th significant rows on the image W to be corrected is presumed by using the proportional relation of the ordinate positions of the rows:
correspondingly, u' -y W (S′ W (i) I) represents the u' th row to y on the image W W (S′ W (i) Distance of row; u-y R (S R (i) A) represents the ith row through the (y) th row on the image R R (S R (i) Distance of row; y is W (S′ W (i))-y W (S′ W (i+1)) represents the y-th on the image W W (S′ W (i) Line (y) W (S′ W Distance of (i+1)) row, y R (S R (i))-y R (S R (i+1)) represents the y-th on the image R R (S R (i) Line (y) R (S R (i+1)) row distance.
c. When u is greater than or equal to y R (S R (q)) in the case of (a) the process,
the row u' corresponding to the q-1 and the q-th significant row on the image W to be corrected is presumed by using the proportional relation of the ordinate positions of the q-1 and the q-th significant rows:
wherein u' -y W (S′ W (q)) represents the (u' th row to y) on the image W W (S′ W (q)) distance of rows; u-y R (S R (q)) represents the (u) th line to the (y) th line on the image R R (S R (q)) distance of rows; y is W (S′ W (q))-y W (S′ W (q-1)) represents the y-th on the image W W (S′ W (q)) line to y W (S′ W (q-1)) distance of row, y R (S R (q))-y R (S R (q-1)) represents the y-th on the image R R (S R (q)) line to y R (S R (q-1)) row distance.
When the correspondence is obtained, the line u' may be a decimal value, and the gray values of all the data of the u th line in the aligned image C need to be obtained through linear interpolation:
wherein v is the value of 1,2, … and w on the abscissa.
The above embodiments are merely illustrative of the principles of the present invention and its effectiveness, and are not intended to limit the invention. Modifications and variations may be made to the above-described embodiments by those skilled in the art without departing from the spirit and scope of the invention. Accordingly, the scope of the invention is to be indicated by the appended claims.
Claims (10)
1. A line scanning cylinder geometric correction method based on significant line matching comprises the following steps:
step S1, acquiring an unfolded image of a perfect cylindrical workpiece by using a line scanning camera, preprocessing to obtain an ideal image as a reference image, and performing line scanning acquisition on the unfolded image of the cylindrical workpiece to be corrected by using the line scanning camera as an image to be corrected;
step S2, respectively carrying out gradient projection on the reference image and the image to be corrected;
s3, performing significant line selection on the reference image and the image to be corrected according to the gradient projection value;
step S4, for each significant line of the reference image, searching a candidate matching line corresponding to the significant line on the image to be corrected;
step S5, sampling the matched line pairs obtained in the step S4 to obtain q matched line pairs, correcting the ordinate of each significant line on the image to be corrected according to the significant line of the reference image of each matched line pair, searching whether significant lines exist in the corresponding range on the reference image according to the correction result, updating the confidence coefficient of the current group of sampled matched line pairs according to the search result, and determining the optimal sampled matched line pairs through multiple iterations and according to the confidence coefficient;
step S6, taking the optimal sampling matching line pair as a reference, and distributing a matching line on the image to be corrected for each significant line on the reference image;
and S7, converting the image to be corrected by using the obtained matching line pair to obtain an image aligned with the reference image, thereby realizing geometric correction.
2. A method of geometry correction of a line scan cylinder based on significant line matching as defined in claim 1, wherein: in step S1, if the side of the cylinder changes along the Y direction of the image, the width of the reference image is the same as the width of the image to be corrected, and the heights of the reference image and the image to be corrected are different; if the side surface of the cylinder is changed along the X direction of the image, the height of the reference image is the same as that of the image to be corrected, and the widths of the reference image and the image to be corrected are different.
3. A method of geometry correction of a line scan cylinder based on significant line matching as defined in claim 2, wherein: in step S1, the cylinder is expanded by default in the Y direction, and if the cylinder is expanded in the X direction, the expanded image is rotated by 90 degrees.
4. A method of geometry correction of a line scan cylinder based on significant line matching as defined in claim 1, wherein: in step S3, a line with a gradient projection value greater than a preset threshold is taken as a significant line.
5. A method of geometry correction of a line scan cylinder based on significant line matching as defined in claim 1, wherein: in step S4, the i-th salient line S in the reference image is processed R (i) And the kth significant line S in the image to be corrected W (k) The matching response M (i, k) is calculated as:
selecting p significant lines with the smallest matching response as candidate matching lines according to the matching response values of the candidate matching lines and each significant line in the image to be corrected, wherein R (S R (i) J) represents the ith salient row S of the reference image R R (i) The gray value of the upper j point, W (S W (k) J represents the kth significant line S of the image W to be corrected W (k) The gray value of the upper j point.
6. The method of geometric correction of a line scan cylinder based on salient line matching of claim 1, wherein step S5 further comprises the steps of:
s500, randomly extracting q different salient lines from salient lines in a reference image, and sequencing the salient lines from small to large according to an ordinate;
step S501, randomly extracting matching lines according to probability from p candidate matching lines of each extracted significant line;
step S502, for q salient lines in the reference image, if the ordinate of the ith salient line is smaller than the ordinate of the jth salient line (i, j e {1,2, …, q } and i+.j), the matching line drawn on the image to be corrected must also satisfy the condition, otherwise, the sampling is discarded and step S500 is re-executed;
step S503, correcting the ordinate of each significant line on the image to be corrected according to q matching line pairs obtained by sampling by taking the sampling line of the reference image as a reference, searching whether significant lines exist in the corresponding range on the reference image according to the correction result, and updating the confidence coefficient of the current group of sampling matching pairs according to the search result;
step S504, returning to step S500 for iteration to the set iteration times, and storing the matching line pair with the highest confidence as the optimal sampling matching line pair.
7. The method of line scan cylinder geometry correction based on salient line matching of claim 6, wherein in step S501, the following operations are performed:
step1: randomly generating 1 integer between 1 and p, wherein the integer corresponds to the selected candidate matching row;
step2, randomly generating 1 value with a value range of [0,1], and sampling one matching line from the current candidate matching line when the value is smaller than the probability of the selected candidate matching line; otherwise, step1 is re-executed until a matching line is selected that meets the condition.
8. The method of line scan cylinder geometry correction based on salient line matching of claim 7, wherein step S503 further comprises:
after each group of sampling matching pairs is obtained, the confidence coefficient of the matching pair is initially 0;
correcting the ordinate of each significant line on the image to be corrected by taking the sampling behavior of the reference image as a benchmark;
after the corrected ordinate is obtained, searching whether a significant line exists in a corresponding range on the reference image, and if so, increasing the current confidence by 1.
9. The method for geometric correction of a line scan cylinder based on salient line matching as claimed in claim 8, wherein in step S6, each salient line position of the reference image is transformed to obtain a corresponding position of the image to be corrected, and a re-matching is performed in the vicinity thereof to find the best corresponding line.
10. A line scan cylinder geometry correction device based on salient line matching, comprising:
the image acquisition processing unit is used for acquiring an unfolded image of a perfect cylindrical workpiece by using the line scanning camera, preprocessing the unfolded image to obtain an ideal image as a reference image, and performing line scanning acquisition on the unfolded image of the cylindrical workpiece to be corrected by using the line scanning camera to obtain an image to be corrected;
the gradient projection unit is used for respectively carrying out gradient projection on the reference image and the image to be corrected;
the salient line selection unit is used for selecting salient lines of the reference image and the image to be corrected according to the gradient projection values;
the rough matching unit is used for searching candidate matching lines corresponding to each significant line of the reference image on the image to be corrected;
the sampling correction optimizing unit is used for sampling the matching line pairs obtained through the coarse matching unit to obtain q matching line pairs, correcting the ordinate of each significant line on the image to be corrected according to the significant line of the reference image of each matching line pair, searching whether significant lines exist in the corresponding range on the reference image according to the correction result, updating the confidence coefficient of the current group of sampling matching line pairs according to the search result, and determining the optimal sampling matching line pairs according to the confidence coefficient through multiple iterations;
the fine matching unit is used for taking the optimal sampling matching line pair as a reference, and distributing a matching line on the image to be corrected for each significant line on the reference image;
and the geometric correction unit is used for transforming the image to be corrected by using the obtained matching line pair to obtain an image aligned with the reference image, thereby realizing geometric correction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110965505.1A CN113674174B (en) | 2021-08-23 | 2021-08-23 | Line scanning cylinder geometric correction method and device based on significant line matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110965505.1A CN113674174B (en) | 2021-08-23 | 2021-08-23 | Line scanning cylinder geometric correction method and device based on significant line matching |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113674174A CN113674174A (en) | 2021-11-19 |
| CN113674174B true CN113674174B (en) | 2023-10-20 |
Family
ID=78544877
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110965505.1A Active CN113674174B (en) | 2021-08-23 | 2021-08-23 | Line scanning cylinder geometric correction method and device based on significant line matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113674174B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114782276B (en) * | 2022-04-29 | 2023-04-11 | 电子科技大学 | Resistivity imaging dislocation correction method based on adaptive gradient projection |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102236798A (en) * | 2011-08-01 | 2011-11-09 | 清华大学 | Image matching method and device |
| CN103136760A (en) * | 2013-03-26 | 2013-06-05 | 华北电力大学（保定） | Multi sensor image matching method based on fast and daisy |
| CN103456022A (en) * | 2013-09-24 | 2013-12-18 | 中国科学院自动化研究所 | High-resolution remote sensing image feature matching method |
| CN103489191A (en) * | 2013-09-24 | 2014-01-01 | 中国科学院自动化研究所 | Method for detecting changes of remarkable target of remote sensing image |
| CN104077782A (en) * | 2014-07-11 | 2014-10-01 | 中国科学院自动化研究所 | Satellite-borne remote sense image matching method |
| CN107918927A (en) * | 2017-11-30 | 2018-04-17 | 武汉理工大学 | A kind of matching strategy fusion and the fast image splicing method of low error |
-
2021
- 2021-08-23 CN CN202110965505.1A patent/CN113674174B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102236798A (en) * | 2011-08-01 | 2011-11-09 | 清华大学 | Image matching method and device |
| CN103136760A (en) * | 2013-03-26 | 2013-06-05 | 华北电力大学（保定） | Multi sensor image matching method based on fast and daisy |
| CN103456022A (en) * | 2013-09-24 | 2013-12-18 | 中国科学院自动化研究所 | High-resolution remote sensing image feature matching method |
| CN103489191A (en) * | 2013-09-24 | 2014-01-01 | 中国科学院自动化研究所 | Method for detecting changes of remarkable target of remote sensing image |
| CN104077782A (en) * | 2014-07-11 | 2014-10-01 | 中国科学院自动化研究所 | Satellite-borne remote sense image matching method |
| CN107918927A (en) * | 2017-11-30 | 2018-04-17 | 武汉理工大学 | A kind of matching strategy fusion and the fast image splicing method of low error |
Non-Patent Citations (1)
| Title |
|---|
| 基于匹配策略融合的低误差快速图像拼接算法;杨明东;《计算机应用研究》;第36卷(第4期);1222-1227 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113674174A (en) | 2021-11-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107543828B (en) | Workpiece surface defect detection method and system | |
| JP2002516440A (en) | Image recognition and correlation system | |
| US10430938B2 (en) | Method of detecting defects in an object | |
| CN109726746B (en) | Template matching method and device | |
| CN108520514B (en) | Consistency detection method for electronic elements of printed circuit board based on computer vision | |
| CN108010082B (en) | Geometric matching method | |
| CN107240130B (en) | Remote sensing image registration method, device and system | |
| CN113674174B (en) | Line scanning cylinder geometric correction method and device based on significant line matching | |
| TWI291543B (en) | A method and a system for creating a reference image using unknown quality patterns | |
| WO2015035462A1 (en) | Point feature based 2d-3d registration | |
| CN114331995A (en) | Multi-template matching real-time positioning method based on improved 2D-ICP | |
| CN108763575B (en) | Image control point automatic selection method based on image control point database | |
| US20130108176A1 (en) | Image processing apparatus and control method therefor | |
| CN113095187A (en) | Examination paper correction method based on image feature matching alignment | |
| US7113652B2 (en) | System and method for using normalized gray scale pattern find | |
| CN114863129A (en) | Instrument numerical analysis method, device, equipment and storage medium | |
| CN111462246A (en) | Equipment calibration method of structured light measurement system | |
| CN113658080B (en) | Linear scanning cylinder geometric correction method and device based on characteristic point matching | |
| CN108992033B (en) | Grading device, equipment and storage medium for vision test | |
| CN107169498A (en) | It is a kind of to merge local and global sparse image significance detection method | |
| CN114998216A (en) | Method and device for rapidly detecting surface defects of transparent part | |
| CN112215240B (en) | Optimization method for improving 2D complex edge detection precision | |
| CN114897797A (en) | Method, device and equipment for detecting defects of printed circuit board and storage medium | |
| CN111368823B (en) | Pointer type instrument reading identification method and device | |
| CN113160284B (en) | Guidance space-consistent photovoltaic image registration method based on local similar structure constraint |
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 |