CN111354038B - Anchor detection method and device, electronic equipment and storage medium - Google Patents
Anchor detection method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN111354038B CN111354038B CN201811575222.0A CN201811575222A CN111354038B CN 111354038 B CN111354038 B CN 111354038B CN 201811575222 A CN201811575222 A CN 201811575222A CN 111354038 B CN111354038 B CN 111354038B
- Authority
- CN
- China
- Prior art keywords
- straight line
- line
- straight
- anchor
- determining
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention discloses an anchor detection method and device, electronic equipment and a storage medium. The anchor detection method comprises the following steps: acquiring an image comprising an anchor image; detecting a straight line in the image; determining a similarity of the detection object to the anchor based on the characteristics of the detected straight line; based on the similarity, a position of the anchor in the image is located.
Description
Technical Field
The present invention relates to the field of information technologies, and in particular, to a method and apparatus for detecting an anchor, an electronic device, and a storage medium.
Background
As technology advances, anchors are often detected using various visual techniques. In the related art, an image including an anchor is acquired by a camera, and then a process is performed based on multiple images such as a deep learning model, so as to track the position of the anchor in the image. However, in the conventional anchor tracking method such as the deep learning model, when the object to be tracked is blocked or the color of the object is similar to that of the surrounding environment, anchor detection failure is likely to occur.
Disclosure of Invention
In view of this, embodiments of the present invention desirably provide an anchor detection method and apparatus, an electronic device, and a storage medium.
The technical scheme of the invention is realized as follows:
an anchor detection method comprising:
acquiring an image comprising an anchor image;
detecting a straight line in the image;
determining a similarity of the detection object to the anchor based on the characteristics of the detected straight line;
based on the similarity, a position of the anchor in the image is located.
Based on the above scheme, the detecting the straight line in the image includes:
detecting straight lines in the image and obtaining at least one first straight line;
performing straight line classification according to the equation of the first straight line;
and forming a second straight line based on the first straight line belonging to the same class according to the classification result of the first straight line.
Based on the above scheme, the determining the similarity between the detection object and the anchor based on the characteristic of the detected straight line comprises:
and determining the similarity of the detection object and the anchoring object according to the intersection point, the included angle and/or the length of the second straight line of the detection object.
Based on the above scheme, the method further comprises:
vertically projecting a first point corresponding to an ith second straight line to the ith second straight line to obtain a first projection coordinate, wherein the first point is a point with the largest included angle between a connecting line between the head and tail points of the first straight line corresponding to the second straight line and an origin and a preset coordinate axis;
projecting a second point corresponding to the ith second straight line to obtain a second projection coordinate, wherein the second point is a point with the minimum included angle between a connecting line between the head and tail points of the first straight line corresponding to the second straight line and an origin and a preset coordinate axis;
determining a distance between an intersection point between an ith second straight line and a jth second straight line and the first projection coordinate and the second projection coordinate;
if the distance is smaller than a first distance threshold, determining that the ith second straight line and the jth second straight line intersect;
and determining the length of the ith second straight line based on an intersection point coordinate, wherein the intersection point coordinate is the coordinate of the intersection point of the ith second straight line and at least two other second straight lines.
Based on the above scheme, the method further comprises: and if the ith second straight line is intersected with the jth second straight line, determining an included angle between the ith straight line and the jth second straight line.
Based on the above scheme, the classifying the straight line according to the equation of the first straight line includes:
determining a rectangular coordinate equation of the first straight line according to coordinates of the head point and the tail point of the first straight line;
determining the slope and intercept of the first straight line according to the rectangular coordinate equation;
determining a paradigm distance between the two first lines based on the slope and the intercept;
and classifying the two first straight lines with the normal form distance smaller than a second distance threshold value.
Based on the above scheme, the determining the similarity between the detection object and the anchor based on the characteristic of the detected straight line comprises:
matching an included angle between an xth second line in the second line set of the detected object and other second lines with an included angle between a y-th line in the template of the anchor and other lines;
based on the matching result of the included angle, determining whether the x second straight line and the y straight line meet an included angle matching condition;
determining whether the x second straight line and the y straight line meet a length matching condition;
the second line and the y line simultaneously meet the included angle matching condition and the length matching condition, and an affine matrix for affine transformation of the second line to the y line is determined according to the coordinates of the head and tail points of the second line and the coordinates of the head and tail points of the y line;
Transforming the second straight lines except the x second straight line according to the affine matrix to obtain a second straight line set after affine transformation of the detected object;
determining the similarity of the second straight line set and the straight lines in the template;
and determining whether the detection object is the anchor according to the similarity.
Based on the above scheme, the determining whether the x-th second straight line and the y-th straight line meet a length matching condition includes: determining whether a length difference between the xth second line and the yth line is less than a length threshold; and if the length difference between the x second straight line and the y straight line is smaller than the length threshold value, determining that the x second straight line and the y straight line meet the length matching condition. Based on the above scheme, the method further comprises:
the x second straight line of the detection object and a plurality of straight lines in the template simultaneously meet the included angle matching condition and the length matching condition to generate a combined set, wherein the combined set comprises a plurality of combinations of the x second straight line and at least two straight lines;
the determining similarity of the detection object and the anchoring object based on the characteristics of the detected straight line comprises:
Traversing each of the set of combinations, determining a similarity of the assay to the anchor based on each of the combinations.
Based on the above scheme, the method further comprises:
determining an included angle range according to included angles of all straight lines in the template;
and if the included angle between the second straight line of the detected object and other second straight lines is out of the included angle range, removing the second straight lines of the detected object, of which the included angles with other second straight lines are out of the included angle range, from the second straight line set.
Based on the above scheme, the method further comprises:
acquiring the image and obtaining an image comprising candidate contour regions imaged by the anchor;
selecting a contour region with the largest area from the candidate contour regions as a target contour region;
the detecting a straight line in the image includes:
detecting a straight line in the target contour area.
An anchor detection device comprising:
an image module for acquiring an image containing an anchor for imaging;
a straight line module for detecting a straight line in the image;
a similarity module for determining a similarity of the detection object to the anchor based on the characteristics of the detected straight line;
A positioning module for positioning a position of the anchor in the image based on the similarity.
An electronic device includes:
a memory;
and the processor is connected with the memory and is used for realizing the anchor detection method provided by any of the technical schemes by executing the computer executable instructions on the memory.
A computer storage medium having stored thereon computer executable instructions; after the computer executable instructions are executed, the method for detecting the anchoring object provided by any of the previous technical schemes is realized.
According to the technical scheme provided by the embodiment of the invention, through straight line detection in the image, the similarity of the detected object and the anchoring object is determined based on the characteristics of the detected straight line; based on the similarity, determining whether the currently detected object is an anchor, and if the similarity between the detected object and the anchor is high, considering the area where the currently detected object is located as the area where the anchor is located, thereby realizing positioning of the anchor in the image; even if the detection object is partially shielded, the detection object is considered to be the required detection object as long as the similarity with the detection object is high enough, so that the successful detection of the detection object is realized, and the detection device has the characteristics of high detection efficiency and accurate detection degree.
Drawings
FIG. 1 is a flow chart of a first method for detecting an anchor according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a system for acquiring images according to an embodiment of the present invention;
fig. 3 is a schematic flow chart of a straight line detection according to an embodiment of the present invention;
FIG. 4 is a schematic flow chart of determining similarity between a detection object and an anchor based on straight line characteristics according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of an anchor detection device according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a template creation process according to an embodiment of the present invention;
FIG. 7 is a flow chart of an anchor detection method according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical scheme of the invention is further elaborated below by referring to the drawings in the specification and the specific embodiments.
As shown in fig. 1, the present embodiment provides an anchor detection method, including:
step S110: acquiring an image comprising an anchor image;
step S120: detecting a straight line in the image;
step S130: determining a similarity of the detection object to the anchor based on the characteristics of the detected straight line;
Step S140: based on the similarity, a position of the anchor in the image is located.
The image in this embodiment may be an image acquired of the anchor, and as such, the imaging of the anchor is necessarily contained in the image, but it may be desirable to locate the region in which the anchor is located in the entire image.
In this embodiment, the anchor may be an object of a known shape, the characteristics of the line to which the anchor corresponds being known. Therefore, in this embodiment, by directly detecting a straight line from an image, and then obtaining the feature of the detected straight line, the similarity between the detected object and the anchor is determined based on the feature, and if the similarity is sufficiently high, for example, the similarity is greater than a similarity threshold, it is determined that the current object is the anchor, and then the image area where the current object is located is the imaging area of the anchor.
The step S110 may include: the image is received from an acquisition device or acquired by itself.
Fig. 2 shows a schematic diagram of an industrially acquired image. In fig. 2, there are signal lamp, camera, pulley production line; the accessory box may be one of the aforementioned anchors, the accessory box moving with the operation of the pulley line. When the accessory box moves to the lower part of the camera and triggers the photoelectric switch, the camera collects pictures and sends the pictures to the processing unit for identification, and the identification result is output and displayed. Thus, the processing unit of the device acquires the image acquired by the camera. An image of the accessory cassette is thus obtained.
In step S120, the image is detected directly from the image, instead of generating the image based on the points in the image. There are several ways to detect straight lines in the image, several alternatives are provided below:
detecting a Hough transformation straight line;
straight line segmentation detection (Line Segment Detector, LSD) straight line detection.
The hough transform line detection can be implemented as follows:
for any point a (X0, Y0) in the rectangular coordinate system, the straight line passing through the point a satisfies y0=kx0+b (k is the slope, b is the intercept); then the linear cluster at X-Y plane passing point a (X0, Y0) can be represented by y0=kx0+b, but cannot be represented if the slope of the line perpendicular to the X axis is infinite. Therefore, the special situation can be solved by converting the rectangular coordinate system into the polar coordinate system.
The straight line segmentation detection (Line Segment Detector, LSD) straight line detection may be as follows: the origin O of the polar coordinate system is selected as the focus (w/2,h/2) of the image. Establishing a Cartesian coordinate system; in which is the image coordinate system. The conversion relation between the polar coordinate system and the Cartesian coordinate system is as follows. Therefore, when the head and tail points of a line segment are known, the corresponding points can be solved. The specific angle is calculated by referring to the polar coordinate representation of the straight line detection.
The specific manner of detecting the straight line in the image is not particularly limited to any of the above.
The characteristics of the straight line detected in step S130 include, but are not limited to, at least one of:
the length of the straight line;
an angle of the straight line;
an included angle between the straight lines;
the relative positional relationship between straight lines includes, for example: intersecting and/or parallel.
The characteristics of the lines are combined to directly reflect the characteristics of the detected object, so that in step S130, the similarity between the detected object and the anchor may be determined based on the characteristics of the lines, and then whether the current detected object is the anchor may be determined based on the similarity.
In summary, in this embodiment, the anchor with the known straight line characteristic can be detected by directly using the straight line detection method, which has the characteristics of simpler implementation, higher accuracy and lower resource cost compared with the deep learning model, and meanwhile, when the contour to be detected is blocked by an unknown object, both the traditional algorithm and the deep learning algorithm will fail, and the method is still applicable.
In some embodiments, as shown in fig. 3, the step S110 may include:
step S111: detecting straight lines in the image and obtaining at least one first straight line;
Step S112: performing straight line classification according to the equation of the first straight line;
step S113: and forming a second straight line based on the first straight line belonging to the same class according to the classification result of the first straight line.
Because of illumination of an image or the like or errors of the line detection itself, one line is sometimes detected as a plurality of lines, so that a line directly detected from an image is referred to as a first line in this embodiment, where the first line may be a line with a length, rather than a line with a wireless length defined mathematically. The first line has endpoints, or such endpoints are also referred to as endpoints. After these first lines are detected, an equation of the first line may be obtained, which may be referred to as a line equation. The straight line equation reflects the characteristics of a straight line, such as the slope and intercept of the straight line.
In this embodiment, the straight lines are classified based on the equation of the first straight line, for example, two straight lines which are sufficiently similar are regarded as being classified into the same class, and then the second straight line is obtained based on the first straight line of the same class.
In some embodiments, the step S112 may include at least one of:
Comparing the slopes and the intercepts of different first straight lines, and classifying a plurality of first straight lines with the same slopes and the same intercepts into one type;
based on the slope and intercept of different first straight lines, calculating the normal distance between any two first straight lines, and classifying the two first straight lines into one type if the normal distance is smaller than a second distance threshold. Specifically, the step S112 may include: determining a rectangular coordinate equation of the first straight line according to coordinates of the head point and the tail point of the first straight line; determining the slope and intercept of the first straight line according to the rectangular coordinate equation; determining a paradigm distance between the two first lines based on the slope and the intercept; and classifying the two first straight lines with the normal form distance smaller than a second distance threshold value. In this embodiment, the end-to-end points may be two end points of a straight line.
The paradigm distance, known as the Manhattan distance, may be determined by the following equation:
taking the coordinates (X1, Y1) and the coordinates (X2, Y2) as examples:
a first absolute value of the difference between X1 and X2 is taken,
a second absolute value of the difference between Y1 and Y2 is taken,
and calculating the sum of the first absolute value and the second absolute value to obtain the normal form distance of the coordinates (X1, Y1) and the coordinates (X2, Y2).
In short, there are various ways of straight line classification based on the equation, and the method is not limited to any of the above.
In step S113, a second straight line connecting the first straight lines of the same class in series is generated based on the first straight lines of the same class. The implementation manner of the step S113 is various, and two alternative manners are provided in this embodiment:
based on the coordinates of the head and tail points of each first straight line in the same class, soft fitting of the second straight line is carried out, so that even if certain deviation exists in the first straight line, the second straight line can be successfully and accurately fitted;
and performing hard fitting of the second straight line based on the coordinates of the head and tail points of the first straight lines in the same class.
When soft fitting of the second straight line is carried out, after coordinates of head and tail points of a plurality of second straight lines in the same class are obtained, the head and tail points of all first straight lines in the same class are located on the second straight line or around the second straight line through soft fitting.
The hard fit in making the second line is: the coordinates of the head and tail points of the first straight lines can be directly found, and then the first points and the second points can be obtained. For example, the predetermined coordinate axis may be an x-axis or a y-axis, the x-axis being perpendicular to the y-axis, the x-axis and the y-axis forming a two-dimensional coordinate system. The first point is a point corresponding to the maximum included angle, and the second point is a point corresponding to the minimum included angle. The end-to-end point may also be referred to as an endpoint, including: two end points at two ends of the first straight line.
By forming the second line, the phenomenon of a large number of repeated lines caused by directly detecting the lines from the image can be reduced, and the subsequent accurate positioning of the anchor in the image is facilitated.
Further, the step S130 may include:
and determining the similarity between the detection object and the anchor according to the intersection point and the included angle of at least two second straight lines of the detection object and/or the length of the second straight lines.
In this embodiment, a picture to be detected may detect a plurality of contours to be detected, each contour to be detected may be fitted with a plurality of second lines, and the features of the detecting lines may include: the length of the second straight lines, the included angle between the second straight lines, the intersection point and the like.
In this embodiment, the similarity between the detection object and the detection object may be determined by matching one or more of the above-described features of the second straight line with the features of the anchor-known straight line.
Further, the method further comprises: a characteristic of the second line is detected.
The detecting the characteristic of the second line may include: included angle and length. The second straight line included angle can be obtained as follows:
vertically projecting a first point corresponding to an ith second straight line to the ith second straight line to obtain a first projection coordinate, wherein the first point is a point with the largest included angle between a connecting line between the head and tail points of the first straight line corresponding to the second straight line and an origin and a preset coordinate axis
Projecting a second point corresponding to the ith second straight line to obtain a second projection coordinate, wherein the second point is a point with the minimum included angle between a connecting line between the head and tail points of the first straight line corresponding to the second straight line and an origin and a preset coordinate axis;
determining a distance between an intersection point between an ith second straight line and a jth second straight line and the first projection coordinate and the second projection coordinate;
if the distance is smaller than a first distance threshold, determining that the ith second straight line and the jth second straight line intersect;
and determining the length of the ith straight line based on the intersection point coordinates of the ith second straight line and the other two second straight lines intersected with the ith second straight line.
For example, if the i-th second line and the second lines a and B intersect respectively, and the intersection point of the i-th second line and the second line a is close to the first projection coordinate, and the intersection point of the i-th second line and the second line B is close to the second projection coordinate, the length of the second line may be calculated based on the intersection point of the i-th second line and the second lines a and B, respectively.
In some embodiments, the ith second straight line may intersect more than two second straight lines, and then the following method may be used in calculating the length of the ith second straight line:
Selecting an m1 th second straight line with the minimum distance from the first projection coordinate on the i th second straight line;
selecting an m2 th second straight line with the smallest distance from the first projection coordinate on the i th second straight line;
calculating the length of the ith second straight line based on the first intersection point of the ith straight line and the m1 th second straight line and the second intersection point of the ith second straight line and the m2 nd second straight line; to improve the accuracy of the length.
In some embodiments, the method comprises: and if the ith second straight line is intersected with the jth second straight line, determining an included angle between the ith straight line and the jth second straight line.
The included angle can be directly solved by the linear equation of the two second lines, or can be solved by other modes. For example, the first point Pfi = (x) fi ，y fi ) Projection point ptfi= (xt) on straight line i fi ，yt fi ) And a second point Pni = (x ni ，y ni ) Projection point ptni= (xt) on straight line i ni ，yt ni ) The point Pfi (or Pni) j not equal to i on the straight line i Pfi, which is closest to the head and tail points of other straight lines, is calculated, and the intersection point coordinate prfi= (xr) of the straight line i and j is calculated fi ，yr fi ) At this time, the included angle between i and j is:
the preset coordinate axis may be an x-axis or a y-axis. If the end point coordinate distance of the two second straight lines in the anchoring profile is smaller than the set threshold value, the two second straight lines are considered to be intersected, and the straight line included angle can be obtained through a straight line equation.
The steps are repeatedly executed, so that the included angle between any two second straight lines of the outline and the length of the second straight lines can be anchored.
And detecting a plurality of second lines in the anchoring profile, wherein each second line has respective head and tail points, then finding other second lines closest to and farthest from the head and tail points of each second line, and further judging whether the two second lines intersect according to the second lines. In some embodiments, each second straight line finds two second intersecting straight lines, and the length of the second straight line can be confirmed through a cross-section line.
In some embodiments, as shown in fig. 4, the step S130 may include:
step S131: matching an included angle between an xth second line in the second line set of the detected object and other second lines with an included angle between a y-th line in the template of the anchor and other lines;
step S132: based on the matching result of the included angle, determining whether the x second straight line and the y straight line in the template meet an included angle matching condition or not;
step S133: determining whether the x second straight line and the y straight line meet a length matching condition;
step S134: the second line and the y line simultaneously meet the included angle matching condition and the length matching condition, and an affine matrix for affine transformation of the second line to the y line is determined according to the coordinates of the head and tail points of the second line and the coordinates of the head and tail points of the y line;
Step S135: transforming the second straight lines except the x second straight line according to the affine matrix to obtain a second straight line set after affine transformation of the detected object;
step S136: determining the similarity of the second straight line set and the straight lines in the template;
step S137: and determining whether the detection object is the anchor according to the similarity.
Because the angle of the anchor relative to the camera may change, the imaging of the anchor in the image may change, such as translation. In this embodiment, for accurate matching, matching of angles and lengths between straight lines is performed to determine the similarity of the detected straight line to the straight line in the anchor template.
And x and y are positive integers, wherein x is smaller than or equal to the total number of the second straight lines, and y is smaller than or equal to the total number of the straight lines in the template.
In some embodiments, the templates may be pre-generated for the electronic device and stored in the electronic device. In some embodiments, the template may also be pre-received by the electronic device from other devices.
And if the included angle between each second straight line and any straight line in the templates meets the included angle similarity condition and the length similarity condition, the detected second straight line is similar to the corresponding straight line in the templates, and then the coordinates of the two similar or matched straight lines are inconsistent due to displacement and the like, so that the affine matrix can be solved through the change of the coordinates of the head and tail points of the two similar or matched straight lines. The affine matrices are used for affine change of other detected second straight lines, so that the second straight lines after affine change can be obtained quickly, then the second straight lines are matched with the straight lines in the template, and a first ratio between the number of matched bars and the total number of the straight lines in the template or a second ratio between the number of matched bars and the number of unmatched bars can be obtained. The first ratio and/or the second ratio may be used herein directly as a parameter for evaluating the degree of similarity between the detector and the anchor. If the first ratio and/or the second ratio is greater than the ratio threshold, the detection object and the anchor are considered to be the same, and the detection object is the imaging area of the anchor in the image area.
The determining whether the x-th second straight line and the y-th straight line meet a length matching condition includes: determining whether a length difference between the xth second line and the yth line is less than a length threshold;
and if the length difference between the x second straight line and the y straight line is smaller than the length threshold value, determining that the x second straight line and the y straight line meet the length matching condition.
In some embodiments, the method further comprises: the picture to be detected can detect a plurality of outlines, and the outlines need to be traversed; thereby finding an image from the plurality of images with the anchor imaged and obtaining the position of the anchor imaged in the image.
In some embodiments, the method may further comprise: the x second straight line of the detection object and a plurality of straight lines in the template simultaneously meet the included angle matching condition and the length matching condition to generate a combined set, wherein the combined set comprises a plurality of combinations of the x second straight line and at least two straight lines; the step S130 may include: traversing each of the set of combinations, determining a similarity of the assay to the anchor based on each of the combinations.
For example, in the step S130, it may include: traversing the combination set to determine an affine matrix corresponding to each combination; and on the basis of the solved second straight lines detected by each affine matrix transformation, matching the transformed second straight lines with the straight lines in the templates, and determining whether the currently detected detection object is the anchor according to the similarity degree or the matching degree of the transformed second straight lines and the straight lines.
If the anchor is a rectangle, trapezoid, etc., there may be multiple parallel sides, then an x second line may be similar to or match multiple lines in the template, in this embodiment, a set of two matched lines is considered as a combination, forming a combination set. And traversing each combination in the combination set, solving an affine matrix of each combination, and determining the similarity between the detection object and the anchor based on each combination so as to determine whether the detection object is similar to the anchor. In some embodiments, the method further comprises:
determining an included angle range according to included angles of all straight lines in the template;
and if the included angle between the second straight line of the detected object and other second straight lines is out of the included angle range, removing the second straight lines of the detected object, of which the included angles with other second straight lines are out of the included angle range, from the second straight line set.
Once the angles between each line in the template and the other lines are determined, the range of angles between the lines in the template is determined. In this embodiment, in order to reduce the subsequent calculation amount, the range of the included angle may be determined in advance based on the included angle between any two straight lines in the template. And then using the range of angles to extract an abnormal second line from the second set of lines.
In this embodiment, the method further includes:
acquiring the image;
obtaining an image comprising candidate contour regions imaged by the anchor;
selecting a contour region with the largest area from the candidate contour regions as a target contour region;
the step S120 may include: detecting a straight line in the target contour area.
In this embodiment, the region in which the object is imaged and the background region may be segmented by pixel-level segmentation, so as to obtain one or more candidate contour regions.
And currently, only obtaining one candidate contour region, and determining the candidate contour region as the target contour region.
A plurality of candidate contour regions are currently obtained, from which one or more target contour regions may be selected. For example, a candidate contour region having the largest current area is selected as the target contour region, and a straight line detection is performed in step S110.
Detection of a detection object that is similar or matches the anchor within the target contour region may be considered successful in anchor positioning and detection may be stopped.
If no anchor is detected in the current target contour region, the detection is continued with the candidate contour region having the next largest area as the target contour region, until the anchor is detected, or until all the candidate contour regions are regarded as the target contour region to complete the detection.
In some embodiments, the imaging size of the anchor may be determined directly according to the acquisition parameters of the camera and the size of the anchor, and one or more candidate contour regions are selected as the target contour regions based on the imaging size, so that each target contour region is subjected to linear detection in parallel, and the positioning rate of the anchor is accelerated. The acquisition parameters may include, but are not limited to, parameters related to image acquisition such as focal length and/or depth of field.
As shown in fig. 5, this embodiment further provides an anchor detecting device, including:
an image module 110 for acquiring an image containing an anchor for imaging;
a straight line module 120 for detecting a straight line in the image;
a similarity module 130 for determining a similarity of the detection object to the anchor based on the characteristics of the detected straight line;
A positioning module 140 for positioning the position of the anchor in the image based on the similarity.
In some embodiments, the image module 110, the line module 120, the similarity module 130, and the positioning module 140 may all be program modules that, when executed by a processor, enable the acquisition of images, the detection of lines, the determination of similarity, and the positioning of anchors in images.
In some embodiments, the straight line module 120 is configured to detect a straight line in the image and obtain at least one first straight line; performing straight line classification according to the equation of the first straight line; and forming a second straight line based on the first straight line belonging to the same class according to the classification result of the first straight line.
In some embodiments, the similarity-based module 130 is configured to determine the similarity between the detection object and the anchor according to an intersection point, an included angle, and/or a length of the second straight line of the at least two second straight lines of the detection object.
In some embodiments, the apparatus further comprises:
the first projection module is used for vertically projecting a first point corresponding to an ith second straight line to the ith second straight line to obtain a first projection coordinate, wherein the first point is a point with the largest included angle between a connecting line between the head and tail points of the first straight line corresponding to the second straight line and an origin and a preset coordinate axis;
The second projection module is used for projecting a second point corresponding to the ith second straight line to obtain a second projection coordinate, wherein the second point is a point with the minimum included angle between a connecting line between the head and tail points of the first straight line corresponding to the second straight line and the origin and a preset coordinate axis;
a length module, configured to determine a distance between an intersection point between an ith second straight line and a jth second straight line and the first projection coordinate and the second projection coordinate; if the distance is smaller than a first distance threshold, determining that the ith second straight line and the jth second straight line intersect; and determining the length of the ith second straight line based on an intersection point coordinate, wherein the intersection point coordinate is the coordinate of the intersection point of the ith second straight line and at least two other second straight lines.
In some embodiments, the apparatus further comprises:
and the first included angle module is used for determining the included angle between the ith straight line and the jth second straight line when the ith second straight line is intersected with the jth second straight line.
In some embodiments, the straight line module 120 is specifically configured to determine a rectangular coordinate equation of the first straight line according to coordinates of the head and tail points of the first straight line; determining the slope and intercept of the first straight line according to the rectangular coordinate equation; determining a paradigm distance between the two first lines based on the slope and the intercept; and classifying the two first straight lines with the normal form distance smaller than a second distance threshold value.
In some embodiments, the similarity module 130 is specifically configured to match an included angle between an xth second line and other second lines in the second line set of the detected object with an included angle between an yth line and other lines in the template of the anchor; based on the matching result of the included angle, determining whether the x second straight line and the y straight line in the template meet an included angle matching condition or not; determining whether the x second straight line and the y straight line meet a length matching condition; if the x second straight line and the y straight line simultaneously meet the included angle matching condition and the length matching condition, determining an affine matrix for affine transformation of the x second straight line to the y straight line according to the coordinates of the head and tail points of the x second straight line and the coordinates of the head and tail points of the y straight line; transforming the second straight lines except the x second straight line according to the affine matrix to obtain a second straight line set after affine transformation of the detected object; determining the similarity of the second straight line set and the straight lines in the template; and determining whether the detection object is the anchor according to the similarity.
In some embodiments, the apparatus further comprises:
the combination module is used for simultaneously meeting the included angle matching condition and the length matching condition with the x second straight line of the detected object and a plurality of straight lines in the template to generate a combination set, wherein the combination set comprises a plurality of combinations of the x second straight line and at least two straight lines;
the similarity module 130 is specifically configured to traverse each of the combinations in the set of combinations and determine a similarity of the detection object to the anchor based on each of the combinations.
In some embodiments, the apparatus further comprises:
the included angle range module is used for determining an included angle range according to included angles of all straight lines in the template;
and the rejecting module is used for rejecting the second straight lines with included angles with other second straight lines in the detected object, which are located outside the included angle range, from the second straight line set.
In some embodiments, the apparatus further comprises:
a candidate contour region module for acquiring the image and obtaining a candidate contour region of the image comprising the anchor imaging;
The target contour region module is used for selecting a contour region with the largest area from the candidate contour regions as a target contour region;
the straight line module 120 is specifically configured to detect a straight line located in the target contour area.
A specific example is provided below in connection with any of the embodiments described above:
example 1:
a hardware part and a system algorithm part.
As shown in fig. 2, the hardware part includes: the belt production line, photoelectric sensor to and be located the camera directly over the belt production line, after the accessory box removes to the camera below, and trigger photoelectric switch, the camera gathers the picture, and sends the processing unit and discerns, and the most output shows the recognition result.
In the first part, the template is established, and the picture serving as the template needs to have the following standards: (1) The picture background is pure (2) the outline of the anchor to be detected is clear and no shielding exists (2) the outline of the anchor to be detected is maximum and only one template anchor is contained.
The specific template building method can refer to fig. 6, which comprises the following steps:
the outline extraction template picture is selected, and the specific steps include: pixel segmentation, setting a pixel threshold value, and extracting candidate contour areas in an image through the pixel segmentation; specifically, for example; comparing the area of each contour candidate region, and screening out all other regions in the region with the largest reserved area;
Binarizing the image;
extracting a contour from the binarized picture;
further extracting the outline with the largest area;
the straight line detection is performed by using a straight line detection algorithm, for example, the straight line is extracted from the contour of the region by using a straight line detection algorithm, and the algorithm used in the specific embodiment is Hough straight line detection or LSD straight line detection algorithm, so that all the straight lines in the contour are detected;
calculating rectangular coordinate equations of all straight lines, obtaining start and end coordinates of each branching section through straight line detection, and setting the coordinates of the first point of the ith straight line as psi= (x) si ，y si ) The coordinates of the tail point are pei= (x) ei ，y ei ) Then there is a straight line i equation:
at this time, the slope a of the straight line i is obtained i With intercept of b i After all slopes and intercepts are obtained, the slopes and the intercepts are clustered, and the number of clustering centers is the number of outline lines of the anchor to be detected.
After clustering, performing straight line fitting on all head and tail points of a straight line i belonging to the same class, calculating an included angle between an original detection straight line from all points to an image origin and an x axis, and solving a first point Pfi = (x) of a maximum included angle fi ，y fi ) Projection point ptfi= (xt) on straight line i fi ，yt fi ) And a second point Pni = (x) corresponding to the minimum angle ni ，y ni ) Projection point ptni= (xt) on straight line i ni ，yt ni ) Taking the furthest example of the projection calculation formula, if the straight line i passes through the origin, pfi is the point furthest from the origin, and Pni is the point closest to the origin:
calculating the point Pfi (or Pni) j not equal to i on the straight line i Pfi with the closest distance from the head and tail points of other straight lines, and obtaining the intersection point coordinate PRfi= (xr) of the straight line i and j fi ，yr fi ) At this time, the included angle between i and j is
The same method finds the point Pni (or Pfi) Pfi where Pni is closest to the other straight line and finds the intersection point coordinate prni= (xr) of straight lines i and j ni ，yr ni ) At this time, the length calculation formula of i is:
record length L of each straight line i Two included angles theta fi θ ni Final detection of Anchor content C in template line And (5) straight lines.
Record the maximum value of all included angles as theta min Minimum value of theta max 。
After the template is built, the test can be performed, when the accessory box blocks the photoelectric switch, and after the camera collects the picture to be detected, referring to fig. 7, the detection can comprise the following operations:
binarizing the image, extracting the outline, and selecting the outline, wherein the method specifically comprises the following steps: dividing pixels, and extracting candidate contours through pixel threshold values; preliminary screening of the profile by area threshold, filtering out an anchor profile having an area less than the anchor, to obtain N area Candidate contours; for the first contour (N area =1) all straight lines in the profile are detected using a straight line detection algorithm.
Using a straight line detection algorithm, for example, a rectangular coordinate equation of all straight lines is calculated, the start and end coordinates of each branching section can be obtained through straight line detection, and the initial coordinate of the ith straight line is set as psi= (x) si ，y si ) Tail coordinates are pei= (x) ei ，y ei ) The slope a of the straight line i is obtained by using the formula (1) i With intercept of b i After all slopes and intercepts are obtained, the normal form distance of each two groups of straight lines is calculated, and the formula is as follows:
L d ＝a i -a j +b i -b j (j≠i) (5)
at this time, a clustering second distance threshold value needs to be setTd, when L d At less than or equal to Td, the anchors are linearly classified into one type and finally to N line A similar straight line, wherein the slope threshold T is set due to the fact that the slope of the straight line tends to infinity by utilizing the point slope equation (namely, when the straight line is perpendicular to the x axis) a For straight line i, if a i ≥T a Or a is set when the straight line is perpendicular to the x axis i ＝T a 。
Projection coordinates ptfi= (xt) were calculated using the same method as in template creation (5) fi ，yt fi ) And Ptni= (xt) ni ，yt ni )。
Calculating the included angle between the straight line i and all other straight lines to obtain N D Initial selection is carried out on the included angles, and all theta is reserved min ≤θ d ≤θ max (d∈[1,N D ]) Obtaining N' D And an included angle. Determining the second straight line M1, so that a certain projection coordinate distance between a certain projection point of the second straight line M1 and the second straight line i is minimum; determining a second intersection point coordinate of an intersection point of the ith second straight line and the Mth 1 second straight line, and determining an included angle between the ith second straight line and the Mth 1 second straight line; determining the second straight line of the M2 th line, so that a certain projection coordinate distance between a certain projection point of the second straight line of the M1 st line and the second straight line of the i th line is minimum; determining a third intersection point coordinate of an intersection point of the ith second straight line and the M2 th second straight line, and determining an included angle between the ith second straight line and the M2 th second straight line; and determining the length of the ith second straight line on the anchor based on the second intersection point coordinate and the third intersection point coordinate.
Calculating the similarity of a straight line i and a template straight line by using the filtered included angle, wherein the similarity calculation formula of the straight line i and the similarity calculation formula of the straight line j are as follows:
at this time, the following is calculated: beta i ＝min{β ij }(j∈[1,C line ]) (7)
When j=v (v e 1, c line ]) Has beta value i Taking the minimum value, the straight line i and the template are representedIs highest.
Setting a similarity threshold T β When beta is i ≤T β When the included angle between the straight line i and the straight line i is consistent with the matching of the template, the two straight lines are j n And j is equal to f When the ladder polygon of parallel lines is selected as an anchor, a plurality of matched straight lines can be obtained through angle screening, v is not unique, and N is arranged in a template linei The' straight line is matched with the i straight line, and j is the same at the moment n And j is equal to f Not only, there are multiple groups of combinations, there are N in total ci A combination.
Select combination n ci ＝1(n ci ∈[1,N ci ]) At this time, there are straight lines i and j n 1 And j is equal to f 1 Intersection, calculating i and j n 1 The intersection point of (a) is ptrni= (xtr) ni ，ytr ni ) And i and j f 1 The intersection point of (a) is ptrfi= (xtr) fi ，ytr fi ) Calculate length LP i Setting a threshold Tl, when |LP i -L v And when the I is less than or equal to Tl, carrying out subsequent matching.
Let n be ci ＝1(n ci ∈[1,N ci ]) Meets the length screening and assumes that the line corresponds to the line v in the template m (m∈[1,N′ linei ]) Let m=1, at which time v m With prfv= (xr) fv ，yr fv )，PRnv＝(xr nv ，yr nv ) At this time, it is assumed that PRni corresponds to PRnv, and in practical industrial field applications, the camera height will be fixed, so that only translation and rotation transformation occurs between the anchor in the test image and the anchor in the template image.
Firstly, translating the straight line i so that the PRni and PRnv pairs coincide, and translating the affine matrix can be as follows:
wherein the method comprises the steps of
Let PTRni' =ptrnv= (xtr) after PTRni is affine transformed ni '，ytr ni '), PTRfi passes through PTRfi' = (xtr) after affine transformation fi '，ytr fi ') and then rotate θ around PTRni, the affine matrix is
Wherein the method comprises the steps of
Finally, the global affine transformation matrix is as follows
After the affine matrix is calculated, the affine matrix is used for carrying out the calculation on the residual N obtained in the step (4) area -1 straight line for affine transformation. Affine transformation is carried out on the head coordinates and the tail coordinates of all the straight lines to obtain
Obtaining a point-inclined straight line equation of Ps 'and Pf' to obtain a corresponding straight line slope a 'and an intercept b', and obtaining the residual N in the step (5) area -1 angle matching result of straight line, if the straight line is matched with the straight line j in the template, calculating the similarity of the straight lines
L d ＝a′-a j +b′-b j (14)
When L d Judging that the straight lines accord with each other when Td is less than or equal to Td, judging the similarity of all the straight lines by the same method, and finally obtaining C' line The bars fit the straight line of the match, at which point there is anchor match similarity:
Setting a matching degree threshold Tnl whenIf so, judging that the detection is successful, and adding M p→r As an affine matrix for the whole picture. All points on the test picture can be matched on the template picture through an affine matrix.
All combinations are traversed but still unmatched, let m=m+1, v m (m∈[1,N′ linei ]) Traversing all N' linei And repeating steps (7) to (9).
If the matching fails, let n ci ＝n ci +1(n ci ∈[1,N ci ]) Traversing n ci And (3) repeating steps (6) to (9) in other combinations.
If the straight line i fails to match, traversing the rest N line Repeating steps (4) to (10) in a quasi-linear manner
If profile N area Let N, if 1 fails to match area ＝N area +1, repeating (3) through (10) until a matching profile is found.
In a word, all second straight lines on the outline to be detected are obtained, the included angles and the truncated lengths of the ith second straight line and all other straight lines on the outline to be detected are obtained, an included angle list and a truncated length list are generated, the included angles and the truncated lengths of the ith second straight line and the generated second straight line of the anchoring outline are matched, and the matching condition of the ith second straight line and the second line of the anchoring outline is predicted preliminarily; it can be quickly determined whether the respective candidate contour region is currently the imaging region of the anchor. The candidate contours are typically ranked and the contours with the largest area are screened first, whereby detection time can be saved.
The detection speed is high, the precision is far higher than that of a pure deep learning algorithm, and pixel-level point positioning can be realized.
The anti-shielding, even if the anchor is shielded by objects with similar colors, has extremely strong robustness.
The universality is strong, and the method is applicable to all polygonal anchors; and the speed is extremely high, and the affine matrix can be positioned at one time and efficiently aiming at the non-step symmetrical anchor.
As shown in fig. 8, the present embodiment provides an electronic device, including:
a memory;
and a processor, coupled to the memory, for implementing the method for detecting a method anchor provided in any one of the foregoing aspects by executing computer-executable instructions located on the memory, for example, one or more of the method for detecting a method anchor shown in fig. 1, 3, 4, 6, and 7.
The memory may be various types of memory, such as random access memory, read only memory, flash memory, etc. The memory may be used for information storage, for example, storing computer-executable instructions, etc. The computer-executable instructions may be various program instructions, such as target program instructions and/or source program instructions, etc.
The processor may be various types of processors such as a central processing unit, a microprocessor, a digital signal processor, a programmable array, a digital signal processor, an application specific integrated circuit, or an image processor, among others.
The processor may be connected to the memory via a bus. The bus may be an integrated circuit bus or the like.
In some embodiments, the electronic device may further include: a communication interface, the communication interface may include: network interfaces, e.g., local area network interfaces, transceiver antennas, etc. The communication interface is also connected with the processor and can be used for information receiving and transmitting.
In some embodiments, the electronic device further comprises a human-machine interaction interface, which may comprise various input-output devices, such as a keyboard, touch screen, etc., for example.
The present embodiment provides a computer storage medium storing computer-executable instructions; after the computer executable instructions are executed, the method for detecting a method anchor provided in any one of the foregoing embodiments is implemented, for example, one or more of the methods for detecting a method anchor shown in fig. 1, 3 and 4, 6 and 7.
The computer storage medium may be various storage media including a recording medium having a recording function, such as a CD, a floppy disk, a hard disk, a magnetic tape, an optical disk, a U-disk, or a removable hard disk. The computer storage medium may optionally be a non-transitory storage medium, where the computer storage medium is readable by a processor, so that after the computer executable instructions stored on the computer storage mechanism are acquired and executed by the processor, the anchor detection method provided in any one of the foregoing solutions is implemented, for example, to perform an anchor detection method applied in a terminal device or an anchor detection method applied in an application server.
In the several embodiments provided by the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above described device embodiments are only illustrative, e.g. the division of the units is only one logical function division, and there may be other divisions in practice, such as: multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. In addition, the various components shown or discussed may be coupled or directly coupled or communicatively coupled to each other via some interface, whether indirectly coupled or communicatively coupled to devices or units, whether electrically, mechanically, or otherwise.
The units described as separate units may or may not be physically separate, and units displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units; some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present application may be integrated in one processing module, or each unit may be separately used as one unit, or two or more units may be integrated in one unit; the integrated units may be implemented in hardware or in hardware plus software functional units.
Those of ordinary skill in the art will appreciate that: all or part of the steps for implementing the above method embodiments may be implemented by hardware associated with program instructions, where the foregoing program may be stored in a computer readable storage medium, and when executed, the program performs steps including the above method embodiments; and the aforementioned storage medium includes: a mobile storage device, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk or an optical disk, or the like, which can store program codes.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (12)
1. An anchor detection method, comprising:
acquiring an image comprising an anchor image;
detecting a straight line in the image;
determining a similarity of the detection object and the anchor object based on the characteristics of the detected straight line;
Locating a position of the anchor in the image based on the similarity;
wherein the detecting a straight line in the image includes:
detecting straight lines in the image and obtaining at least one first straight line;
performing straight line classification according to the equation of the first straight line;
forming a second straight line based on the first straight line belonging to the same class according to the classification result of the first straight line;
the determining similarity of the detection object and the anchoring object based on the characteristics of the detected straight line comprises:
and determining the similarity between the detection object and the anchoring object according to the intersection point and the included angle of at least two second straight lines of the detection object and/or the length of the second straight lines.
2. The method according to claim 1, wherein the method further comprises:
vertically projecting a first point corresponding to an ith second straight line to the ith second straight line to obtain a first projection coordinate, wherein the first point is a point with the largest included angle between a connecting line between the head and tail points of the first straight line corresponding to the second straight line and an origin and a preset coordinate axis;
projecting a second point corresponding to the second straight line of the ith to obtain a second projection coordinate, wherein the second point is a point with the minimum included angle between a connecting line between the head and tail points of the first straight line corresponding to the second straight line and an origin and a preset coordinate axis;
Determining a distance between an intersection point between an ith second straight line and a jth second straight line and the first projection coordinate and the second projection coordinate;
if the distance is smaller than a first distance threshold, determining that the ith second straight line and the jth second straight line intersect;
and determining the length of the ith second straight line based on an intersection point coordinate, wherein the intersection point coordinate is the coordinate of the intersection point of the ith second straight line and at least two other second straight lines.
3. The method according to claim 1, wherein the method further comprises:
and if the ith second straight line is intersected with the jth second straight line, determining an included angle between the ith second straight line and the jth second straight line.
4. The method of claim 1, wherein the step of determining the position of the substrate comprises,
the classifying the straight line according to the equation of the first straight line includes:
determining a rectangular coordinate equation of the first straight line according to coordinates of the head point and the tail point of the first straight line;
determining the slope and intercept of the first straight line according to the rectangular coordinate equation;
determining a paradigm distance between the two first lines based on the slope and the intercept;
And classifying the two first straight lines with the normal form distance smaller than a second distance threshold value.
5. The method of claim 1, wherein determining the similarity of the detection object to the anchor based on the characteristics of the detected line comprises:
matching an included angle between an xth second line in the second line set of the detected object and other second lines with an included angle between a y-th line in the template of the anchor and other lines;
based on the matching result of the included angle, determining whether the x second straight line and the y straight line meet an included angle matching condition;
determining whether the x second straight line and the y straight line meet a length matching condition;
the second line and the y line simultaneously meet the included angle matching condition and the length matching condition, and an affine matrix for affine transformation of the second line to the y line is determined according to the coordinates of the head and tail points of the second line and the coordinates of the head and tail points of the y line;
transforming the second straight lines except the x second straight line according to the affine matrix to obtain a second straight line set after affine transformation of the detection object;
Determining the similarity of the second straight line set and the straight lines in the template;
and determining whether the detection object is the anchor according to the similarity.
6. The method according to claim 5, further comprising:
the x second straight line of the detection object and a plurality of straight lines in the template simultaneously meet the included angle matching condition and the length matching condition to generate a combined set, wherein the combined set comprises a plurality of combinations of the x second straight line and at least two straight lines;
the determining similarity of the detection object and the anchoring object based on the characteristics of the detected straight line comprises:
traversing each of the set of combinations, determining a similarity of the assay to the anchor based on each of the combinations.
7. The method of claim 5, wherein the method further comprises:
determining an included angle range according to included angles of all straight lines in the template;
and if the included angle between the second straight line of the detected object and other second straight lines is out of the included angle range, removing the second straight lines of the detected object, of which the included angles with other second straight lines are out of the included angle range, from the second straight line set.
8. The method of claim 5, wherein the step of determining the position of the probe is performed,
the determining whether the x-th second straight line and the y-th straight line meet a length matching condition includes:
determining whether a length difference between the xth second line and the yth line is less than a length threshold;
and if the length difference between the x second straight line and the y straight line is smaller than the length threshold value, determining that the x second straight line and the y straight line meet the length matching condition.
9. The method according to any one of claims 1 to 8, further comprising:
acquiring the image and obtaining an image comprising candidate contour regions imaged by the anchor;
selecting a contour region with the largest area from the candidate contour regions as a target contour region;
the detecting a straight line in the image includes:
detecting a straight line in the target contour area.
10. An anchor detection device, comprising:
an image module for acquiring an image containing an anchor for imaging;
a straight line module for detecting a straight line in the image;
a similarity module for determining a similarity of the detection object and the anchor based on the characteristics of the detected straight line;
A positioning module for positioning a position of the anchor in the image based on the similarity;
the linear module is used for detecting a linear in the image and obtaining at least one first linear;
performing straight line classification according to the equation of the first straight line;
forming a second straight line based on the first straight line belonging to the same class according to the classification result of the first straight line;
the similarity module is used for determining the similarity between the detection object and the anchoring object according to the intersection point and the included angle of at least two second straight lines of the detection object and/or the length of the second straight lines.
11. An electronic device, comprising:
a memory;
a processor, coupled to the memory, for implementing the method provided in any one of claims 1 to 9 by executing computer-executable instructions located on the memory.
12. A computer storage medium having stored thereon computer executable instructions; the computer-executable instructions, when executed, implement the method provided in any one of claims 1 to 9.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811575222.0A CN111354038B (en) | 2018-12-21 | 2018-12-21 | Anchor detection method and device, electronic equipment and storage medium |
| PCT/CN2019/124678 WO2020125528A1 (en) | 2018-12-21 | 2019-12-11 | Anchor object detection method and apparatus, electronic device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811575222.0A CN111354038B (en) | 2018-12-21 | 2018-12-21 | Anchor detection method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111354038A CN111354038A (en) | 2020-06-30 |
| CN111354038B true CN111354038B (en) | 2023-10-13 |
Family
ID=71102011
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811575222.0A Active CN111354038B (en) | 2018-12-21 | 2018-12-21 | Anchor detection method and device, electronic equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111354038B (en) |
| WO (1) | WO2020125528A1 (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112070142B (en) * | 2020-09-02 | 2024-05-10 | 平安科技（深圳）有限公司 | Grouping method and device for vehicle accessories, electronic equipment and storage medium |
| CN112859189A (en) * | 2020-12-31 | 2021-05-28 | 广东美的白色家电技术创新中心有限公司 | Workpiece detection device, detection method, and computer-readable storage medium |
| CN113362290B (en) * | 2021-05-25 | 2023-02-10 | 同济大学 | Method, storage device and device for quickly identifying collinear features of dot matrix planes |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103837135A (en) * | 2012-11-27 | 2014-06-04 | 国际商业机器公司 | Workpiece detecting method and system |
| CN104240227A (en) * | 2013-06-24 | 2014-12-24 | 富泰华工业（深圳）有限公司 | Image analysis system and method |
| CN105069786A (en) * | 2015-07-31 | 2015-11-18 | 小米科技有限责任公司 | Straight line detection method and straight line detection device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6179592B2 (en) * | 2013-05-31 | 2017-08-16 | 日本電気株式会社 | Image recognition apparatus, processing method thereof, and program |
-
2018
- 2018-12-21 CN CN201811575222.0A patent/CN111354038B/en active Active
-
2019
- 2019-12-11 WO PCT/CN2019/124678 patent/WO2020125528A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103837135A (en) * | 2012-11-27 | 2014-06-04 | 国际商业机器公司 | Workpiece detecting method and system |
| CN104240227A (en) * | 2013-06-24 | 2014-12-24 | 富泰华工业（深圳）有限公司 | Image analysis system and method |
| CN105069786A (en) * | 2015-07-31 | 2015-11-18 | 小米科技有限责任公司 | Straight line detection method and straight line detection device |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020125528A1 (en) | 2020-06-25 |
| CN111354038A (en) | 2020-06-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108985137B (en) | License plate recognition method, device and system | |
| Bansal et al. | Ultrawide baseline facade matching for geo-localization | |
| EP2783328B1 (en) | Text detection using multi-layer connected components with histograms | |
| US9014432B2 (en) | License plate character segmentation using likelihood maximization | |
| CN109753953B (en) | Method and device for positioning text in image, electronic equipment and storage medium | |
| CN111354038B (en) | Anchor detection method and device, electronic equipment and storage medium | |
| CN109255802B (en) | Pedestrian tracking method, device, computer equipment and storage medium | |
| KR101742115B1 (en) | An inlier selection and redundant removal method for building recognition of multi-view images | |
| CN110288040B (en) | Image similarity judging method and device based on topology verification | |
| CN111695373A (en) | Zebra crossing positioning method, system, medium and device | |
| Maalek et al. | Robust detection of non-overlapping ellipses from points with applications to circular target extraction in images and cylinder detection in point clouds | |
| CN110929782B (en) | River channel abnormity detection method based on orthophoto map comparison | |
| CN110673607B (en) | Feature point extraction method and device under dynamic scene and terminal equipment | |
| KR101151435B1 (en) | Apparatus and method of recognizing a face | |
| US9830532B1 (en) | Object detection in images using distance maps | |
| US7440636B2 (en) | Method and apparatus for image processing | |
| JP2011107878A (en) | Position detection apparatus and position detection method | |
| CN114926508B (en) | Visual field boundary determining method, device, equipment and storage medium | |
| CN110896469B (en) | Resolution testing method for three-shot photography and application thereof | |
| Noury et al. | How to overcome perceptual aliasing in ASIFT? | |
| Quach et al. | A model-based approach to finding tracks in SAR CCD images | |
| KR20220101373A (en) | Apparatus and method for detecting defects in pcb based on video | |
| Sabater et al. | Rejecting wrong matches in stereovision | |
| US11816878B1 (en) | Object detection in an image based on one or more oriented projection spaces | |
| RU2778906C1 (en) | Method for automatically recognizing scenes and objects in an image |
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 |