CN117876468B - Convex hull-based detection method and system for low-beam light type characteristic points of automobile lamplight - Google Patents
Convex hull-based detection method and system for low-beam light type characteristic points of automobile lamplight Download PDFInfo
- Publication number
- CN117876468B CN117876468B CN202410275293.8A CN202410275293A CN117876468B CN 117876468 B CN117876468 B CN 117876468B CN 202410275293 A CN202410275293 A CN 202410275293A CN 117876468 B CN117876468 B CN 117876468B
- Authority
- CN
- China
- Prior art keywords
- point
- value
- pixel
- image
- points
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims description 85
- 238000000034 method Methods 0.000 claims abstract description 64
- 238000009499 grossing Methods 0.000 claims abstract description 11
- 238000004422 calculation algorithm Methods 0.000 claims description 25
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000001914 filtration Methods 0.000 claims description 11
- 238000009877 rendering Methods 0.000 claims description 3
- 230000036039 immunity Effects 0.000 abstract description 3
- 230000000694 effects Effects 0.000 description 23
- 238000010586 diagram Methods 0.000 description 12
- 230000008859 change Effects 0.000 description 10
- 238000004590 computer program Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000003190 augmentative effect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a convex hull-based method and a convex hull-based system for detecting low beam light type characteristic points of automobile lights, wherein the method comprises the steps of sequentially carrying out smoothing filter processing, opening operation and binarization processing on an initial image, and utilizing Canny operator processing to obtain a first contour point set; drawing a connecting line of a first contour point set and adjacent points thereof in a first all-0 image; processing the first contour point set to obtain a first convex hull point set; drawing the connecting lines of the first convex hull point set and adjacent points in a second all-0 image; respectively comparing pixel values of the first full-0 image and the second full-0 image column by column and point by point to respectively obtain a first point column and a second point column; performing difference between the ordinate of the corresponding point in the first point row and the second point row to obtain a difference value sequence; solving each item of the difference value sequence to obtain a first sequence; and comparing each numerical value of the first number column with a preset threshold value to obtain corner points and characteristic points of the dipped beam patterns. The invention has rotational invariance, dimensional invariance and noise immunity.
Description
Technical Field
The invention belongs to the technical field of automobile dipped headlight light detection, and particularly relates to an automobile dipped headlight light characteristic point detection method and system based on convex hulls.
Background
The automobile headlamp is an important guarantee for safe driving of an automobile, and the detection of a dipped headlight light type image occupies an important position in the detection of the dipped headlight of the automobile, wherein the detection of characteristic points is very important. However, the existing detection method for the dipped beam type characteristic points commonly used at present has obvious defects in terms of operability, practicability and detection efficiency and accuracy of detection results, and cannot meet the actual demands of the current automobile flow line production.
For the detection of the characteristic points of the low beam pattern, researchers have achieved a certain result after many years of research. For example, the maximum level calculation method and the photocell method are combined to detect the light distribution performance of the front-illuminated dipped headlight, so that some problems in definition of the cut-off line and determination of the characteristic points are discussed, but due to the influence of actual physical equipment, the cut-off line and the characteristic points of the dipped headlight have certain ambiguity, and accurate positions are difficult to directly find; aiming at the light type image with blurred cut-off line, the Gaussian Laplace operator is adopted to carry out edge detection on the image, the edges detected for many times are overlapped, the point with the largest overlapping times is the position of the characteristic point, and obviously, the method has to be improved in the precision of extracting the characteristic point; the improved Canny operator extracts the optical axis angle based on the Canny algorithm, the gray center of gravity is used as the light spot center, and finally, a linear fitting method of a Tukey weighting function is used for extracting the characteristic point information, but due to the problems of aperture overexposure, pixel distortion and the like of a camera, a larger error still exists in characteristic point detection.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a convex hull-based method and a convex hull-based system for detecting the low beam characteristic points of automobile lights.
In a first aspect, the invention provides a convex hull-based detection method for a low beam light type characteristic point of automobile lamplight, which comprises the following steps:
Acquiring an initial image of a dipped beam pattern;
smoothing filtering processing is carried out on the initial image to obtain a first image;
performing open operation processing on the first image to obtain a second image;
performing binarization processing on the second image to obtain a third image;
Processing the third image by using a Canny operator to obtain a first contour point set of low beam patterns;
Drawing a connecting line of the first contour point set and the adjacent points of the first contour point set in a first all-0 image; the point, where the connecting line of the first contour point set and the adjacent point of the first contour point set passes through in the first all-0 image, is used as a second contour point set;
processing the first contour point set by utilizing a convex hull algorithm to obtain a first convex hull point set of the dipped beam type contour;
drawing a connecting line of the first convex hull point set and the adjacent points of the first convex hull point set in a second all-0 image; the point, where the connecting line of the first convex hull point set and the adjacent point of the first convex hull point set passes through in the second all-0 image, is used as a second convex hull point set;
The method comprises the steps of performing row-by-row from left to right and point-by-point comparison of pixel values and preset pixel value thresholds from top to bottom in a first all-0 image; under the condition that the pixel value of a first target point in one column of pixel points is equal to a first preset pixel threshold value, determining that the first target point meets the comparison condition and comparing the pixel values of the next column of pixel points; the pixel value of a first target point in a column of pixel points of the first all-0 image is equal to a first preset pixel threshold value to serve as a first comparison condition, at most one point in each column of pixel points meets the first comparison condition, and all pixel points meeting the first comparison condition serve as a first point column;
In the second all-0 image, comparing pixel values with a preset pixel value threshold value from left to right in a row-by-row mode and from top to bottom in a point-by-point mode; under the condition that the pixel value of a second target point in one column of pixel points is equal to a second preset pixel threshold value, determining that the second target point meets the comparison condition and comparing the pixel values of the next column of pixel points; wherein, the pixel value of the second target point in a row of pixel points of the second all-0 image is equal to a second preset pixel threshold value and is used as a second comparison condition, at most one point in each row of pixel points meets the second comparison condition, and all pixel points meeting the second comparison condition are used as a second point row;
Performing difference calculation on the ordinate of the corresponding point in the first point row and the second point row to obtain a difference value number row;
determining each numerical value of the first number column according to each numerical value of the difference number column;
Comparing each numerical value in the first number column with a preset item threshold v; taking q 1+ε1 points in the first point column as first corner points of the low beam type of the automobile lamplight; taking q 2+p-ε1 points in the first point column as second corner points of the low beam type of the automobile lamplight; taking q 3+ε2 points in the first point row as characteristic points of the low beam type of the automobile lamplight; taking q 4+p-ε2 points in the first point column as third corner points of the low beam type of the automobile lamplight; wherein q 1 is the sequence number of the first item in the first sequence satisfying the item value greater than the preset item threshold v; epsilon 1 is a first compensation value; q 2 is the sequence number of the last item in the first array that satisfies the item value greater than the preset item threshold v; epsilon 2 is a second compensation value; p is a preset term spacing; q 3 is the sequence number of the first item in the first array satisfying the item value less than-v; q 4 is the sequence number of the last item in the first array that satisfies the item value less than-v; epsilon 1 ≤ p;1≤ ε2 is more than or equal to 1 and p is more than or equal to p.
Further, the performing a difference operation on the ordinate of the corresponding point in the first point row and the second point row to obtain a difference value row includes:
the value of each term of the difference series is calculated according to the following formula:
；
wherein d 1n is the value of the nth term of the difference sequence; An ordinate value for an nth point in the first point column; /(I) An ordinate value for an nth point in the second point row; n=1, 2, …, s; s is the length of the first or second dot column.
Further, the determining each value of the first number sequence according to each value of the difference number sequence includes:
Each value of the first array is calculated according to the following formula:
d2m=d1m+p-d1m；
wherein d 2m is the value of the m-th term of the first series; d 1m+p is the value of the m+p term of the difference sequence; d 1m is the value of the m-th term of the difference sequence; p is a preset term spacing; m=1, 2, …, s -p.
Further, the drawing the first contour point set and the connection line of the adjacent points of the first contour point set in the first all 0 image includes:
calculating the pixel value of each pixel point of the first all 0 image according to the following formula:
；
Wherein P i j is the second set of contour points A point with a median coordinate of (i, j ); /(I)The pixel value of the pixel point P ij in the first all 0 image; v 1 is the second set of contour points/>Pixel values for each point in (a).
Further, the drawing the first convex hull point set and the connection line of the adjacent points of the first convex hull point set in the second all 0 image includes:
Calculating the pixel value of each pixel point of the second full 0 image according to the following formula:
；
Wherein P i'j' is the second convex hull point set A point with a median coordinate of (i', j'); /(I)The pixel value of the pixel point P i'j' in the second full 0 image; v 2 is the second convex hull point set/>Pixel values for each point in (a).
In a second aspect, the present invention provides a convex hull-based system for detecting a low beam characteristic point of an automobile light, including:
the image acquisition module is used for acquiring an initial image of the dipped beam pattern;
The first image processing module is used for carrying out smoothing filter processing on the initial image to obtain a first image;
the second image processing module is used for performing open operation processing on the first image to obtain a second image;
The third image processing module is used for carrying out binarization processing on the second image to obtain a third image;
the fourth image processing module is used for processing the third image by utilizing a Canny operator to obtain a first contour point set of low beam patterns;
The first drawing module is used for drawing the first contour point set and the connecting lines of the adjacent points of the first contour point set in the first all-0 image; the point, where the connecting line of the first contour point set and the adjacent point of the first contour point set passes through in the first all-0 image, is used as a second contour point set;
the contour point set processing module is used for processing the first contour point set by utilizing a convex hull algorithm to obtain a first convex hull point set of the dipped beam type contour;
The second drawing module is used for drawing the first convex hull point set and the connecting line of the adjacent points of the first convex hull point set in a second all-0 image; the point, where the connecting line of the first convex hull point set and the adjacent point of the first convex hull point set passes through in the second all-0 image, is used as a second convex hull point set;
The first pixel value comparison module is used for comparing the pixel value with a preset pixel value threshold value from top to bottom in a first all-0 image from left to right in a row-by-row manner; under the condition that the pixel value of a first target point in one column of pixel points is equal to a first preset pixel threshold value, determining that the first target point meets the comparison condition and comparing the pixel values of the next column of pixel points; the pixel value of a first target point in a column of pixel points of the first all-0 image is equal to a first preset pixel threshold value to serve as a first comparison condition, at most one point in each column of pixel points meets the first comparison condition, and all pixel points meeting the first comparison condition serve as a first point column;
The second pixel value comparison module is used for comparing the pixel value with a preset pixel value threshold value from top to bottom in a row-by-row manner from left to right in the second all-0 image; under the condition that the pixel value of a second target point in one column of pixel points is equal to a second preset pixel threshold value, determining that the second target point meets the comparison condition and comparing the pixel values of the next column of pixel points; wherein, the pixel value of the second target point in a row of pixel points of the second all-0 image is equal to a second preset pixel threshold value and is used as a second comparison condition, at most one point in each row of pixel points meets the second comparison condition, and all pixel points meeting the second comparison condition are used as a second point row;
the difference calculating module is used for carrying out difference calculating on the ordinate of the corresponding point in the first point row and the second point row to obtain a difference value sequence;
the determining module is used for determining each numerical value of the first number sequence according to each numerical value of the difference number sequence;
The feature point detection module is used for comparing each numerical value in the first number sequence with a preset item threshold v; taking q 1+ε1 points in the first point column as first corner points of the low beam type of the automobile lamplight; taking q 2+p-ε1 points in the first point column as second corner points of the low beam type of the automobile lamplight; taking q 3+ε2 points in the first point row as characteristic points of the low beam type of the automobile lamplight; taking q 4+p-ε2 points in the first point column as third corner points of the low beam type of the automobile lamplight; wherein q 1 is the sequence number of the first item in the first sequence satisfying the item value greater than the preset item threshold v; epsilon 1 is a first compensation value; q 2 is the sequence number of the last item in the first array that satisfies the item value greater than the preset item threshold v; epsilon 2 is a second compensation value; p is a preset term spacing; q 3 is the sequence number of the first item in the first array satisfying the item value less than-v; q 4 is the sequence number of the last item in the first array that satisfies the item value less than-v; epsilon 1 ≤ p;1≤ ε2 is more than or equal to 1 and p is more than or equal to p.
Further, the difference module includes:
a first calculation unit for calculating a value of each term of the difference sequence according to the following formula:
；
wherein d 1n is the value of the nth term of the difference sequence; An ordinate value for an nth point in the first point column; /(I) An ordinate value for an nth point in the second point row; n=1, 2, …, s; s is the length of the first or second dot column.
Further, the determining module includes:
a second calculation unit for calculating each numerical value of the first series according to the following formula:
d2m=d1m+p-d1m；
wherein d 2m is the value of the m-th term of the first series; d 1m+p is the value of the m+p term of the difference sequence; d 1m is the value of the m-th term of the difference sequence; p is a preset term spacing; m=1, 2, …, s -p.
Further, the first drawing module includes:
A third calculation unit for calculating a pixel value of each pixel point of the first all 0 image according to the following formula:
；
Wherein P i j is the second set of contour points A point with a median coordinate of (i, j ); /(I)The pixel value of the pixel point P ij in the first all 0 image; v 1 is the second set of contour points/>Pixel values for each point in (a).
Further, the second drawing module includes:
a fourth calculation unit for calculating a pixel value of each pixel point of the second all 0 image according to the following formula:
；
Wherein P i'j' is the second convex hull point set A point with a median coordinate of (i', j'); /(I)The pixel value of the pixel point P i'j' in the second full 0 image; v 2 is the second convex hull point set/>Pixel values for each point in (a).
The invention provides a convex hull-based method and a convex hull-based system for detecting low beam light type characteristic points of automobile lamplight, wherein the method comprises the steps of smoothing and filtering an initial image to obtain a first image; performing open operation on the first image to obtain a second image; performing binarization processing on the second image to obtain a third image; processing the third image by using a Canny operator to obtain a first contour point set of low beam patterns; drawing a connecting line of a first contour point set and adjacent points thereof in a first all-0 image; processing the first contour point set by utilizing a convex hull algorithm to obtain a first convex hull point set of the dipped beam type contour; drawing the connecting lines of the first convex hull point set and adjacent points in a second all-0 image; pixel value comparison is carried out on the first full 0 image column by column and point by point to obtain a first point column; pixel value comparison is carried out on the second full-0 image column by column and point by point to obtain a second point column; performing difference calculation on the ordinate of the corresponding point in the first point row and the second point row to obtain a difference value number row; solving the difference value sequence and each item thereof to obtain a first sequence; and comparing each numerical value of the first number column with a preset threshold value, and detecting to obtain the corner points and the characteristic points of the dipped beam patterns. The invention has the advantages of rotation invariance, scale invariance and noise immunity, and has higher running speed.
Drawings
In order to more clearly illustrate the technical solutions of the present invention, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a convex hull-based method for detecting low beam light type characteristic points of automobile lights according to an embodiment of the invention;
FIG. 2 is a schematic view of a bathtub-type low beam light cutoff according to an embodiment of the present invention;
FIG. 3 is a diagram of a bathtub-type low beam pattern according to an embodiment of the present invention;
FIG. 4 is a diagram of a low beam pattern with noise for a bathtub according to an embodiment of the present invention;
FIG. 5 is a sample down graph of FIG. 3 provided by an embodiment of the present invention;
FIG. 6 is a further down-sampling plot of FIG. 3 provided by an embodiment of the present invention;
FIG. 7 is a view of FIG. 3 rotated 30 according to an embodiment of the present invention;
FIG. 8 is a view of the FIG. 3 rotated-30 provided in accordance with an embodiment of the present invention;
FIG. 9 is an effect diagram of the smoothing filter processing of FIG. 3 according to an embodiment of the present invention;
FIG. 10 is an effect diagram of the open operation processing of FIG. 9 according to the embodiment of the present invention;
FIG. 11 is a diagram showing the effect of the binarization processing of FIG. 10 according to an embodiment of the present invention;
FIG. 12 is a representation of the calculated low beam profile of FIG. 11 according to an embodiment of the present invention;
Fig. 13 is a convex hull display diagram calculated for the low beam profile of fig. 12 according to an embodiment of the present invention;
FIG. 14 is a graph showing the variation of the difference sequence according to the embodiment of the present invention;
FIG. 15 is a graph showing a variation of a first series provided by an embodiment of the present invention;
FIG. 16 is a graph showing the detection effect of the method according to the present invention;
FIG. 17 is a graph of the effect of detecting a cutoff line according to the method of the present invention;
fig. 18 is a detection effect diagram of a harris detection algorithm provided by an embodiment of the present invention;
Fig. 19 is a detection effect diagram of SHI-Tomasi detection algorithm provided in the embodiment of the present invention;
FIG. 20 is a diagram showing the detection effect of the fast detection algorithm according to the embodiment of the present invention;
FIG. 21 is a graph showing the detection effect of the Moravec detection algorithm according to the embodiment of the present invention;
FIG. 22 is a graph showing the detection effect of the method of the present invention on FIG. 5 according to the embodiment of the present invention;
FIG. 23 is a graph showing the detection effect of the method of the present invention on FIG. 6 according to the embodiment of the present invention;
FIG. 24 is a graph showing the effect of the method of the present invention on the cut-off line detection of FIG. 5 according to the embodiment of the present invention;
FIG. 25 is a graph showing the effect of the method of the present invention on the cut-off line detection of FIG. 6 according to the embodiment of the present invention;
FIG. 26 is a graph showing the detection effect of the method of the present invention on FIG. 7 according to an embodiment of the present invention;
FIG. 27 is a graph showing the effect of the method of the present invention on the cut-off line detection of FIG. 7 according to an embodiment of the present invention;
FIG. 28 is a graph showing the detection effect of the method of the present invention on FIG. 8 according to an embodiment of the present invention;
FIG. 29 is a graph showing the effect of the method of the present invention on the cut-off line detection of FIG. 8 according to an embodiment of the present invention;
FIG. 30 is a graph showing the detection effect of the method according to the present invention under the influence of noise;
FIG. 31 is a graph showing the effect of detecting the cut-off line of the method according to the present invention under the influence of noise;
Fig. 32 is a detection effect diagram of a harris detection algorithm under the influence of noise according to an embodiment of the present invention;
fig. 33 is a diagram of detection effect of SHI-Tomasi detection algorithm under noise influence according to an embodiment of the present invention;
FIG. 34 is a graph showing the detection effect of fast detection algorithm under noise influence according to the embodiment of the present invention;
FIG. 35 is a graph of the detection effect of the Moravec detection algorithm under the influence of noise according to an embodiment of the present invention;
fig. 36 is a schematic structural diagram of a convex hull-based system for detecting low beam light characteristic points of automobile lights according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In an embodiment, as shown in fig. 1, an embodiment of the present invention provides a convex hull-based method for detecting a low beam characteristic point of an automobile light. The characteristic point mentioned in the embodiment of the present invention is the intersection of the low beam cutoff horizontal portion and the inclined portion, i.e., the C point in fig. 2 is the characteristic point. The first corner of the bathtub type from left to right is defined as the first corner in the embodiment of the present invention, as point a in fig. 2; the second corner point from left to right of the bathtub type is defined as the second corner point in the embodiment of the present invention, as point B in fig. 2; the third corner of the bathtub type from left to right is defined as the third corner in the embodiment of the present invention, as the D point in fig. 2. Wherein the corner points belong to feature points, which include corner points and other points. Therefore, in this embodiment, the corner points may be regarded as the feature points, but the feature points cannot be regarded as the corner points.
The height of the bathtub refers to the vertical distance or pixel difference between a first corner point and a lower horizontal cut-off line of the dipped beam, namely the connection line between a second corner point and a characteristic point.
The method for detecting the characteristic points of the low beam light of the automobile light comprises the following steps:
step 101, an initial image of the low beam pattern is acquired.
In this step, an actual low beam pattern photographed by an industrial camera is taken as an initial image; as shown in fig. 3, the resolution is 1600×1200.
And 102, performing smoothing filtering processing on the initial image to obtain a first image.
The initial image is subjected to smoothing filtering processing by using a uniform filtering operator to obtain a first image, and as shown in fig. 9, the smoothing filtering is helpful for improving the smoothness of the image edges. In the embodiment of the invention, the filter kernel size adopted by the uniform filter operator is 11×11. The smooth filtering is side window mean filtering or side window Gaussian filtering, and the edge protection capability of the algorithm is further improved.
Step 103, performing an open operation process on the first image to obtain a second image.
As shown in fig. 10, the open operation may further smooth the edge contour while eliminating edge burr points. In this embodiment, the open operation has a structure core size of 19×19.
And 104, performing binarization processing on the second image to obtain a third image.
In this step, illustratively, as shown in fig. 11, the binarization threshold value is set to 50. It should be noted that, the setting of the binarization threshold affects the accuracy of the detection result, and the actual setting of the magnitude of the binarization threshold needs to be actually adjusted according to parameters such as the exposure parameter of the camera, the brightness of the lamp, the attenuation rate of the attenuation piece, or a binarization method capable of sampling the local self-adaptive threshold.
And 105, processing the third image by using a Canny operator to obtain a first contour point set of the dipped beam type.
Step 106, drawing the first contour point set and the connection line of the adjacent points of the first contour point set in the first all-0 image, as shown in fig. 12; and the points, where the connecting lines of the adjacent points of the first contour point set pass through in the first all-0 image, are used as the second contour point set.
The size of the first all 0 image is identical to the initial image size. The pixel value of each point in the second contour point set is set to V 1; illustratively, the pixel value for each pixel point of the first all 0 image is calculated according to the following formula:
。
Wherein P i j is the second set of contour points A point with a median coordinate of (i, j ); /(I)The pixel value of the pixel point P i j in the first all 0 image; v 1 is the second set of contour points/>Pixel values for each point in (a). V 1 is an integer between 1 and 255; in the embodiment of the present invention, V 1 is 255.
And step 107, processing the first contour point set by using a convex hull algorithm to obtain a first convex hull point set of the dipped beam type contour.
Step 108, drawing the first convex hull point set and the connection line of the adjacent points of the first convex hull point set in the second full 0 image, as shown in fig. 13, wherein the point where the connection line of the first convex hull point set and the adjacent points of the first convex hull point set passes through in the second full 0 image is used as the second convex hull point set.
The size of the second all 0 image is identical to the initial image size. The pixel value of each point in the second convex hull point set is set to be V 2; illustratively, the pixel value for each pixel of the second all 0 image is calculated according to the following formula:
。
Wherein P i'j' is the second convex hull point set A point with a median coordinate of (i', j'); /(I)The pixel value of the pixel point P i'j' in the second full 0 image; v 2 is the second convex hull point set/>Pixel values for each point in (a). V 2 is an integer between 1 and 255; in the embodiment of the present invention, V 2 is 255.
Step 109, comparing the pixel value with a preset pixel value threshold (the preset pixel value threshold in this step is V 1) from left to right and from top to bottom in the first all-0 image; under the condition that the pixel value of a first target point in one column of pixel points is equal to a first preset pixel threshold value, determining that the first target point meets the comparison condition and immediately comparing the pixel values of the next column of pixel points; the pixel value of the first target point in a column of pixel points of the first all-0 image is equal to a first preset pixel threshold value to serve as a first comparison condition, at most one point in each column of pixel points meets the first comparison condition, and all pixel points meeting the first comparison condition serve as a first point column. The arrangement order of the dots in the first dot column coincides with the detection order of the dots in the first dot column.
Step 1010, comparing the pixel value with a preset pixel value threshold (the preset pixel value threshold in this step is V 2) from left to right and from top to bottom in the second all-0 image; under the condition that the pixel value of a second target point in one column of pixel points is equal to a second preset pixel threshold value, determining that the second target point meets the comparison condition and immediately comparing the pixel values of the next column of pixel points; and the pixel value of a second target point in a row of pixel points of the second all-0 image is equal to a second preset pixel threshold value to serve as a second comparison condition, at most one point in each row of pixel points meets the second comparison condition, and all pixel points meeting the second comparison condition serve as second point rows. The arrangement order of the points in the second point row is identical to the detection order of the points in the second point row.
Step 1011, performing a difference operation on the ordinate of the corresponding point in the first point row and the second point row to obtain a difference value row; the trend of the difference series is shown in fig. 14.
Illustratively, the value of each term of the differential series is calculated according to the following formula:
。
wherein d 1n is the value of the nth term of the difference sequence; An ordinate value for an nth point in the first point column; /(I) An ordinate value for an nth point in the second point row; n=1, 2, …, s; s is the length of the first or second dot column.
Step 1012, determining each value of the first series of values from each value of the series of difference values.
Illustratively, each term value of the first series is calculated according to the following formula:
d2m=d1m+p-d1m。
wherein d 2m is the value of the m-th term of the first series; d 1m+p is the value of the m+p term of the difference sequence; d 1m is the value of the m-th term of the difference sequence; p is a preset term spacing; m=1, 2, …, s -p.
In order to avoid the influence of noise when the terms are differenced, the term distance p between the two terms of the difference is greater than 1, and in the embodiment of the invention, p is 30.
Step 1013, comparing each numerical value in the first number row with a preset threshold v; taking q 1+ε1 points in the first point column as first corner points of the low beam type of the automobile lamplight; taking q 2+p-ε1 points in the first point column as second corner points of the low beam type of the automobile lamplight; taking q 3+ε2 points in the first point row as characteristic points of the low beam type of the automobile lamplight; taking q 4+p-ε2 points in the first point column as third corner points of the low beam type of the automobile lamplight; wherein q 1 is the sequence number of the first item in the first sequence satisfying the item value greater than the preset item threshold v; epsilon 1 is a first compensation value; q 2 is the sequence number of the last item in the first array that satisfies the item value greater than the preset item threshold v; epsilon 2 is a second compensation value; p is a preset term spacing; q 3 is the sequence number of the first item in the first array satisfying the item value less than-v; q 4 is the sequence number of the last item in the first array that satisfies the item value less than-v; epsilon 1 ≤ p;1≤ ε2 is more than or equal to 1 and p is more than or equal to p.
In this step, v.gtoreq.2, illustratively, v is 4, and ε 1 and ε 2 are 15.
In order to make the solution of the present invention clearer, the embodiment of the present invention further discloses specific examples.
Under normal conditions (meaning that the light pattern is not inverted upside down or its inclination is within ±90°), the instantaneous rate of change of the difference between the convex hull and the contour suddenly increases at a first angle, abrupt change from near 0 to a larger value V4 (0→v4), as shown by the point a in fig. 15; the instantaneous rate of change of the difference between the corresponding convex hull and the contour at the second corner point will be steeply reduced, and the value from V4 suddenly changes to the vicinity of 0 (V4-0), as shown by the point B in FIG. 15; the instantaneous rate of change of the difference between the corresponding convex hull and the contour at the feature point will be steeply reduced from about 0 value to-V 5(0→-V), as shown at point C in fig. 15; the instantaneous rate of change of the difference between the corresponding convex hull and the contour at the fourth corner will increase abruptly from-V to around 0 (-V 5→0), as shown at D in fig. 15.
Therefore, only one value V is needed to be taken, 0< v < V4 and V <V are met, when the point that the instantaneous change rate of the difference value between the first and last met convex hulls and the contour is larger than V is detected, the point can be regarded as a first corner point and a second corner point after a certain compensation is carried out, and when the point that the instantaneous change rate of the difference value between the first and last met convex hulls and the contour is smaller than-V is detected, the point can be regarded as a characteristic point and a fourth corner point after a certain compensation is carried out.
The characteristic points and all the angular points of the dipped beam are all detected, and the cut-off line can be obtained only by taking the first point and the last point in the contour point set and sequentially connecting the first point and the last point with the measured characteristic points and the angular points.
Fig. 3 is a "bathtub" type image, which is a low beam type image of the mainstream in the industry at present, and is used as an input image, and feature point detection is performed by using a harris corner detection method, a SHI-Tomasi corner detection method, a fast corner detection method and a Moravec corner detection algorithm, respectively, and compared with the detection result of the low beam type feature point detection method (hereinafter referred to as the method of the present invention) of the automobile lamplight provided by the present invention.
As shown in fig. 16, in the result of the detection method for the low beam characteristic points and the angular points of the automotive lamplight according to the present invention, it can be seen that the method of the present invention can accurately detect the low beam characteristic points, and simultaneously accurately detect other angular points included in the low beam, such as the cross points in fig. 16. The corner points are connected with each other, and two points with the maximum change of edge gradients are respectively taken at two sides, so that a low beam cutoff can be obtained, as shown in fig. 17. As can be seen from comparing the results of fig. 18, 19, 20 and 21, the harris corner detection method, the SHI-Tomasi corner detection method and the fast corner detection method cannot accurately detect the feature points of the low beam, and the algorithm has more parameters, which is inconvenient for debugging. For the result of fig. 21, although the Moravec corner detection algorithm detects the low beam feature points, a plurality of interference points are detected at the same time, which is inconvenient for accurately positioning the actually required feature points and has low operation efficiency. In the embodiment of the invention, the running speed of the method is about 68.6 percent faster than that of the Moravec corner detection algorithm.
To further verify the detection performance of the method of the present invention, embodiments of the present invention are described with reference to fig. 3, 5, 6, 22 and 23.
Using fig. 5 and 6 as input images, fig. 5 and 6 are downsampled images of fig. 3 with downsampling factors of 1/2 and 1/4, respectively, i.e. resolutions 800 x 600 and 400 x 300, respectively.
The "bathtub" height in FIG. 3 is about 22 pixels; the "bathtub" height in FIG. 5 is 10 pixels; in fig. 6, the "bathtub" is only 6 pixels, the cut-off line is almost a horizontal line, and in such a small scale, it is difficult for the conventional corner detection algorithm to accurately detect the characteristic points and cut-off lines of the low beam pattern.
Fig. 22 shows the result of the detection of the low beam type corner points of fig. 5 by the method of the present invention, and it can be seen from fig. 22 that the method of the present invention can accurately detect the low beam characteristic points and other corner points (cross points in fig. 22) included in the low beam type even in the case of low resolution. Fig. 23 shows the result of the detection of the low beam type corner point of fig. 6 by the method of the present invention, and it can be seen from fig. 23 that the characteristic points and the corner points of the low beam type can be accurately detected by the method of the present invention even if the cut-off line has reached the nearly horizontal state.
According to the embodiment, the method can accurately detect the characteristic points and the angular points of the low beam patterns of different bathtub heights and the low beam patterns of different resolutions, as shown in fig. 24 and 25, and accurately draw the low beam cut-off line, so that the method is verified to have scale invariance.
The rotational invariance of the method of the present invention was further verified in conjunction with figures 3, 7, 8, 26 and 28.
Using fig. 7 and 8 as input images, fig. 7 and 8 are rotated images of fig. 3, wherein fig. 7 is a 30 ° change in clockwise rotation angle of fig. 3, and fig. 8 is a 30 ° change in clockwise rotation angle of fig. 3.
Fig. 26 shows the result of the detection of the low beam type corner points of fig. 7 by the method according to the present invention, and it can be seen from fig. 26 that after the image is rotated 30 ° clockwise, the method according to the present invention can still accurately detect the low beam characteristic points and other corner points (cross points in fig. 26) included in the low beam type. Fig. 28 shows the result of the detection of the low beam type corner point of fig. 8 by the method of the present invention, and it can be seen from fig. 28 that the method of the present invention can accurately detect the feature points and the corner points of the low beam type even after the image is rotated clockwise by-30 °.
According to the embodiment, the method can accurately detect the characteristic points and the angular points of the low beam patterns rotating at different angles, as shown in fig. 27 and 29, and accurately draw the cut-off line of the low beam, and the method is verified to have rotation invariance, so that the method can be well applied to a complex car lamp online adjustment detection system.
The noise immunity of the inventive method was verified in conjunction with fig. 3 and 30-35.
The low beam pattern for fig. 3 is augmented with Jiang Gaosi random noise with a mean of 0 and a variance of 0.01, as shown in fig. 4. The detection is carried out by the method of the invention, and the detection is compared with the traditional corner detection algorithm, and the results are shown in fig. 30 to 35. As can be seen from fig. 30, 31, 32, 33, 34 and 35, the method of the present invention can accurately identify the low beam characteristic points even in the case of strong noise. The harris corner detection method, the SHI-Tomasi corner detection method and the fast corner detection method are seriously affected by noise, and a plurality of noise points are identified as corner points; for the Moravec corner detection algorithm, the Moravec corner detection algorithm is less affected by noise, but the problem of interference points is still difficult to solve, and the operation efficiency is far lower than that of the method.
Based on the same inventive concept, the embodiment of the invention also provides a convex hull-based automobile light low beam type feature point detection system, and because the principle of solving the problem of the system is similar to that of a convex hull-based automobile light low beam type feature point detection method, the implementation of the system can be referred to the implementation of the convex hull-based automobile light low beam type feature point detection method, and repeated parts are omitted.
In another embodiment, the convex hull-based system for detecting the characteristic points of low beam light of the automobile light according to the embodiment of the present invention, as shown in fig. 36, includes:
The image acquisition module 10 is used for acquiring an initial image of the dipped beam type.
The first image processing module 20 is configured to perform smoothing filtering processing on the initial image to obtain a first image.
The second image processing module 30 is configured to perform an open operation on the first image to obtain a second image.
And a third image processing module 40, configured to perform binarization processing on the second image to obtain a third image.
The fourth image processing module 50 is configured to process the third image by using a Canny operator to obtain a first contour point set of low beam patterns.
A first drawing module 60, configured to draw a first contour point set and a connection line of adjacent points of the first contour point set in a first all 0 image; and the points, where the connecting lines of the adjacent points of the first contour point set pass through in the first all-0 image, are used as the second contour point set.
The contour point set processing module 70 is configured to process the first contour point set by using a convex hull algorithm to obtain a first convex hull point set of the low beam contour.
The second drawing module 80 is configured to draw the first convex hull point set and a connection line of the adjacent points of the first convex hull point set in the second all 0 image; and the point, where the connecting line of the adjacent points of the first convex hull point set passes through in the second all-0 image, is used as a second convex hull point set.
A first pixel value comparison module 90, configured to perform a row-by-row from left to right and a point-by-point comparison of a pixel value and a preset pixel value threshold from top to bottom in a first all-0 image; under the condition that the pixel value of a first target point in one column of pixel points is equal to a first preset pixel threshold value, determining that the first target point meets the comparison condition and comparing the pixel values of the next column of pixel points; the pixel value of the first target point in a column of pixel points of the first all-0 image is equal to a first preset pixel threshold value to serve as a first comparison condition, at most one point in each column of pixel points meets the first comparison condition, and all pixel points meeting the first comparison condition serve as a first point column.
A second pixel value comparison module 100, which performs a row-by-row from left to right and a point-by-point comparison of the pixel value and a preset pixel value threshold from top to bottom in the second all-0 image; under the condition that the pixel value of a second target point in one column of pixel points is equal to a second preset pixel threshold value, determining that the second target point meets the comparison condition and comparing the pixel values of the next column of pixel points; and the pixel value of a second target point in a row of pixel points of the second all-0 image is equal to a second preset pixel threshold value to serve as a second comparison condition, at most one point in each row of pixel points meets the second comparison condition, and all pixel points meeting the second comparison condition serve as second point rows.
The difference calculating module 110 is configured to perform a difference calculating operation on the ordinate of the corresponding point in the first point row and the second point row to obtain a difference value array;
The determining module 120 is configured to determine each value of the first number sequence according to each value of the difference number sequence.
The feature point detection module 130 is configured to compare each numerical value in the first array with a preset term threshold v; taking q 1+ε1 points in the first point column as first corner points of the low beam type of the automobile lamplight; taking q 2+p-ε1 points in the first point column as second corner points of the low beam type of the automobile lamplight; taking q 3+ε2 points in the first point row as characteristic points of the low beam type of the automobile lamplight; taking q 4+p-ε2 points in the first point column as third corner points of the low beam type of the automobile lamplight; wherein q 1 is the sequence number of the first item in the first sequence satisfying the item value greater than the preset item threshold v; epsilon 1 is a first compensation value; q 2 is the sequence number of the last item in the first array that satisfies the item value greater than the preset item threshold v; epsilon 2 is a second compensation value; p is a preset term spacing; q 3 is the sequence number of the first item in the first array satisfying the item value less than-v; q 4 is the sequence number of the last item in the first array that satisfies the item value less than-v; epsilon 1 ≤ p;1≤ ε2 is more than or equal to 1 and p is more than or equal to p.
Illustratively, the differencing module comprises:
a first calculation unit for calculating a value of each term of the difference sequence according to the following formula:
。
wherein d 1n is the value of the nth term of the difference sequence; An ordinate value for an nth point in the first point column; /(I) An ordinate value for an nth point in the second point row; n=1, 2, …, s; s is the length of the first or second dot column.
Illustratively, the determining module includes:
a second calculation unit for calculating each numerical value of the first series according to the following formula:
d2m=d1m+p-d1m。
wherein d 2m is the value of the m-th term of the first series; d 1m+p is the value of the m+p term of the difference sequence; d 1m is the value of the m-th term of the difference sequence; p is a preset term spacing; m=1, 2, …, s -p.
Illustratively, the first rendering module includes:
A third calculation unit for calculating a pixel value of each pixel point of the first all 0 image according to the following formula:
。
wherein P ij is the second set of contour points A point with a median coordinate of (i, j ); /(I)The pixel value of the pixel point P i j in the first all 0 image; v 1 is the second set of contour points/>Pixel values for each point in (a).
Illustratively, the second rendering module includes:
a fourth calculation unit for calculating a pixel value of each pixel point of the second all 0 image according to the following formula:
。
Wherein P i'j' is the second convex hull point set A point with a median coordinate of (i', j'); /(I)The pixel value of the pixel point P i'j' in the second full 0 image; v 2 is the second convex hull point set/>Pixel values for each point in (a).
For more specific working procedures of the above modules, reference may be made to the corresponding contents disclosed in the foregoing embodiments, and no further description is given here.
In another embodiment, the invention provides a computer device comprising a processor and a memory; the method comprises the steps of realizing the convex hull-based detection method for the low beam light type characteristic points of the automobile light when a processor executes a computer program stored in a memory.
For more specific procedures of the above method, reference may be made to the corresponding contents disclosed in the foregoing embodiments, and no further description is given here.
In another embodiment, the present invention provides a computer-readable storage medium storing a computer program; the computer program is executed by the processor to realize the steps of the convex hull-based automobile light low beam type feature point detection method.
For more specific procedures of the above method, reference may be made to the corresponding contents disclosed in the foregoing embodiments, and no further description is given here.
In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, so that the same or similar parts between the embodiments are referred to each other. For the system, apparatus and storage medium disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the relevant points refer to the description of the method section.
It will be apparent to those skilled in the art that the techniques of embodiments of the present invention may be implemented in software plus a necessary general purpose hardware platform. Based on such understanding, the technical solutions in the embodiments of the present invention may be embodied in essence or what contributes to the prior art in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the embodiments or some parts of the embodiments of the present invention.
The invention has been described in detail in connection with the specific embodiments and exemplary examples thereof, but such description is not to be construed as limiting the invention. It will be understood by those skilled in the art that various equivalent substitutions, modifications or improvements may be made to the technical solution of the present invention and its embodiments without departing from the spirit and scope of the present invention, and these fall within the scope of the present invention. The scope of the invention is defined by the appended claims.
Claims (10)
1. The method for detecting the low beam light type characteristic points of the automobile lamplight based on the convex hull is characterized by comprising the following steps of:
Acquiring an initial image of a dipped beam pattern;
smoothing filtering processing is carried out on the initial image to obtain a first image;
performing open operation processing on the first image to obtain a second image;
performing binarization processing on the second image to obtain a third image;
Processing the third image by using a Canny operator to obtain a first contour point set of low beam patterns;
Drawing a connecting line of the first contour point set and the adjacent points of the first contour point set in a first all-0 image; the point, where the connecting line of the first contour point set and the adjacent point of the first contour point set passes through in the first all-0 image, is used as a second contour point set;
processing the first contour point set by utilizing a convex hull algorithm to obtain a first convex hull point set of the dipped beam type contour;
drawing a connecting line of the first convex hull point set and the adjacent points of the first convex hull point set in a second all-0 image; the point, where the connecting line of the first convex hull point set and the adjacent point of the first convex hull point set passes through in the second all-0 image, is used as a second convex hull point set;
The method comprises the steps of performing row-by-row from left to right and point-by-point comparison of pixel values and preset pixel value thresholds from top to bottom in a first all-0 image; under the condition that the pixel value of a first target point in one column of pixel points is equal to a first preset pixel threshold value, determining that the first target point meets the comparison condition and comparing the pixel values of the next column of pixel points; the pixel value of a first target point in a column of pixel points of the first all-0 image is equal to a first preset pixel threshold value to serve as a first comparison condition, at most one point in each column of pixel points meets the first comparison condition, and all pixel points meeting the first comparison condition serve as a first point column;
In the second all-0 image, comparing pixel values with a preset pixel value threshold value from left to right in a row-by-row mode and from top to bottom in a point-by-point mode; under the condition that the pixel value of a second target point in one column of pixel points is equal to a second preset pixel threshold value, determining that the second target point meets the comparison condition and comparing the pixel values of the next column of pixel points; wherein, the pixel value of the second target point in a row of pixel points of the second all-0 image is equal to a second preset pixel threshold value and is used as a second comparison condition, at most one point in each row of pixel points meets the second comparison condition, and all pixel points meeting the second comparison condition are used as a second point row;
Performing difference calculation on the ordinate of the corresponding point in the first point row and the second point row to obtain a difference value number row;
determining each numerical value of the first number column according to each numerical value of the difference number column;
comparing each numerical value in the first number column with a preset item threshold v; taking q 1+ε1 points in the first point column as first corner points of the low beam type of the automobile lamplight; taking q 2+p-ε1 points in the first point column as second corner points of the low beam type of the automobile lamplight; taking q 3+ε2 points in the first point row as characteristic points of the low beam type of the automobile lamplight; taking q 4+p-ε2 points in the first point column as third corner points of the low beam type of the automobile lamplight; wherein q 1 is the sequence number of the first item in the first sequence satisfying the item value greater than the preset item threshold v; epsilon 1 is a first compensation value; q 2 is the sequence number of the last item in the first array that satisfies the item value greater than the preset item threshold v; epsilon 2 is a second compensation value; p is a preset term spacing; q 3 is the sequence number of the first item in the first array satisfying the item value less than-v; q 4 is the sequence number of the last item in the first array that satisfies the item value less than-v; epsilon 1 ≤ p;1≤ ε2 is more than or equal to 1 and p is more than or equal to p.
2. The convex hull-based method for detecting low beam light type feature points of automobile lights according to claim 1, wherein the performing a difference operation on the ordinate of the corresponding point in the first point row and the second point row to obtain a difference value row includes:
the value of each term of the difference series is calculated according to the following formula:
；
Wherein d1 n is the value of the nth term of the difference sequence; An ordinate value for an nth point in the first point column; /(I) An ordinate value for an nth point in the second point row; n=1, 2, …, s; s is the length of the first or second dot column.
3. The convex hull-based method for detecting low beam light characteristic points of automobile lights according to claim 1, wherein the determining each value of the first series according to each value of the difference series comprises:
Each value of the first array is calculated according to the following formula:
d2m=d1m+p-d1m；
Wherein d2 m is the value of the m-th term of the first series; d1 m+p is the value of the m+p term of the difference sequence; d1 m is the value of the m-th term of the difference sequence; p is a preset term spacing; m=1, 2, …, s-p.
4. The convex hull-based low beam light feature point detection method for automobile lights according to claim 1, wherein the drawing the first contour point set and the connection line of the adjacent points of the first contour point set in the first all 0 image includes:
calculating the pixel value of each pixel point of the first all 0 image according to the following formula:
；
Wherein P i j is the second set of contour points A point with intermediate coordinates (i, j); /(I)The pixel value of the pixel point P i j in the first all 0 image; v1 is the second set of contour points/>Pixel values for each point in (a).
5. The convex hull-based low beam light type feature point detection method for the automobile light according to claim 1, wherein the drawing the connection line between the first convex hull point set and the adjacent point of the first convex hull point set in the second all 0 images includes:
Calculating the pixel value of each pixel point of the second full 0 image according to the following formula:
；
Wherein P i'j' is the second convex hull point set A point with intermediate coordinates (i ', j'); /(I)The pixel value of the pixel point P i'j' in the second full 0 image; v2 is the second convex hull point set/>Pixel values for each point in (a).
6. Automobile lamplight low beam type characteristic point detection system based on convex hull, which is characterized by comprising:
the image acquisition module is used for acquiring an initial image of the dipped beam pattern;
The first image processing module is used for carrying out smoothing filter processing on the initial image to obtain a first image;
the second image processing module is used for performing open operation processing on the first image to obtain a second image;
The third image processing module is used for carrying out binarization processing on the second image to obtain a third image;
the fourth image processing module is used for processing the third image by utilizing a Canny operator to obtain a first contour point set of low beam patterns;
The first drawing module is used for drawing the first contour point set and the connecting lines of the adjacent points of the first contour point set in the first all-0 image; the point, where the connecting line of the first contour point set and the adjacent point of the first contour point set passes through in the first all-0 image, is used as a second contour point set;
the contour point set processing module is used for processing the first contour point set by utilizing a convex hull algorithm to obtain a first convex hull point set of the dipped beam type contour;
The second drawing module is used for drawing the first convex hull point set and the connecting line of the adjacent points of the first convex hull point set in a second all-0 image; the point, where the connecting line of the first convex hull point set and the adjacent point of the first convex hull point set passes through in the second all-0 image, is used as a second convex hull point set;
The first pixel value comparison module is used for comparing the pixel value with a preset pixel value threshold value from top to bottom in a first all-0 image from left to right in a row-by-row manner; under the condition that the pixel value of a first target point in one column of pixel points is equal to a first preset pixel threshold value, determining that the first target point meets the comparison condition and comparing the pixel values of the next column of pixel points; the pixel value of a first target point in a column of pixel points of the first all-0 image is equal to a first preset pixel threshold value to serve as a first comparison condition, at most one point in each column of pixel points meets the first comparison condition, and all pixel points meeting the first comparison condition serve as a first point column;
The second pixel value comparison module is used for comparing the pixel value with a preset pixel value threshold value from top to bottom in a row-by-row manner from left to right in the second all-0 image; under the condition that the pixel value of a second target point in one column of pixel points is equal to a second preset pixel threshold value, determining that the second target point meets the comparison condition and comparing the pixel values of the next column of pixel points; wherein, the pixel value of the second target point in a row of pixel points of the second all-0 image is equal to a second preset pixel threshold value and is used as a second comparison condition, at most one point in each row of pixel points meets the second comparison condition, and all pixel points meeting the second comparison condition are used as a second point row;
the difference calculating module is used for carrying out difference calculating on the ordinate of the corresponding point in the first point row and the second point row to obtain a difference value sequence;
the determining module is used for determining each numerical value of the first number sequence according to each numerical value of the difference number sequence;
The feature point detection module is used for comparing each numerical value in the first number sequence with a preset item threshold v; taking q 1+ε1 points in the first point column as first corner points of the low beam type of the automobile lamplight; taking q 2+p-ε1 points in the first point column as second corner points of the low beam type of the automobile lamplight; taking q 3+ε2 points in the first point row as characteristic points of the low beam type of the automobile lamplight; taking q 4+p-ε2 points in the first point column as third corner points of the low beam type of the automobile lamplight; wherein q 1 is the sequence number of the first item in the first sequence satisfying the item value greater than the preset item threshold v; epsilon 1 is a first compensation value; q 2 is the sequence number of the last item in the first array that satisfies the item value greater than the preset item threshold v; epsilon 2 is a second compensation value; p is a preset term spacing; q 3 is the sequence number of the first item in the first array satisfying the item value less than-v; q 4 is the sequence number of the last item in the first array that satisfies the item value less than-v; epsilon 1 ≤ p;1≤ ε2 is more than or equal to 1 and p is more than or equal to p.
7. The convex hull-based low beam light pattern feature point detection system of an automotive lamp according to claim 6, wherein the difference module comprises:
a first calculation unit for calculating a value of each term of the difference sequence according to the following formula:
；
Wherein d1 n is the value of the nth term of the difference sequence; An ordinate value for an nth point in the first point column; /(I) An ordinate value for an nth point in the second point row; n=1, 2, …, s; s is the length of the first or second dot column.
8. The convex hull based low beam light feature point detection system of an automotive light of claim 6, wherein the determining module comprises:
a second calculation unit for calculating each numerical value of the first series according to the following formula:
d2m=d1m+p-d1m；
Wherein d2 m is the value of the m-th term of the first series; d1 m+p is the value of the m+p term of the difference sequence; d1 m is the value of the m-th term of the difference sequence; p is a preset term spacing; m=1, 2, …, s-p.
9. The convex hull based low beam light feature point detection system of an automotive light of claim 6, wherein the first rendering module comprises:
A third calculation unit for calculating a pixel value of each pixel point of the first all 0 image according to the following formula:
；
wherein P i j is the second set of contour points A point with intermediate coordinates (i, j); /(I)The pixel value of the pixel point P ij in the first all 0 image; v1 is the second set of contour points/>Pixel values for each point in (a).
10. The convex hull-based low beam light feature point detection system of an automotive lamp according to claim 6, wherein the second drawing module comprises:
a fourth calculation unit for calculating a pixel value of each pixel point of the second all 0 image according to the following formula:
；
Wherein P i'j' is the second convex hull point set A point with intermediate coordinates (i ', j'); /(I)The pixel value of the pixel point P i'j' in the second full 0 image; v2 is the second convex hull point set/>Pixel values for each point in (a).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410275293.8A CN117876468B (en) | 2024-03-12 | 2024-03-12 | Convex hull-based detection method and system for low-beam light type characteristic points of automobile lamplight |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410275293.8A CN117876468B (en) | 2024-03-12 | 2024-03-12 | Convex hull-based detection method and system for low-beam light type characteristic points of automobile lamplight |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117876468A CN117876468A (en) | 2024-04-12 |
| CN117876468B true CN117876468B (en) | 2024-05-10 |
Family
ID=90583345
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410275293.8A Active CN117876468B (en) | 2024-03-12 | 2024-03-12 | Convex hull-based detection method and system for low-beam light type characteristic points of automobile lamplight |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117876468B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH10289319A (en) * | 1997-04-16 | 1998-10-27 | Sumitomo Electric Ind Ltd | Method and device for checking stripe-shaped defect |
| CN117444441A (en) * | 2023-09-07 | 2024-01-26 | 桂林电子科技大学 | Intelligent lap joint welding method |
| CN117576219A (en) * | 2023-10-21 | 2024-02-20 | 东北石油大学 | Camera calibration equipment and calibration method for single shot image of large wide-angle fish-eye lens |
-
2024
- 2024-03-12 CN CN202410275293.8A patent/CN117876468B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH10289319A (en) * | 1997-04-16 | 1998-10-27 | Sumitomo Electric Ind Ltd | Method and device for checking stripe-shaped defect |
| CN117444441A (en) * | 2023-09-07 | 2024-01-26 | 桂林电子科技大学 | Intelligent lap joint welding method |
| CN117576219A (en) * | 2023-10-21 | 2024-02-20 | 东北石油大学 | Camera calibration equipment and calibration method for single shot image of large wide-angle fish-eye lens |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117876468A (en) | 2024-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110866924B (en) | Line structured light center line extraction method and storage medium | |
| JP6358351B1 (en) | Surface defect inspection method and surface defect inspection apparatus | |
| CN109003258B (en) | High-precision sub-pixel circular part measuring method | |
| CN108369650B (en) | Method for identifying possible characteristic points of calibration pattern | |
| CN111968144B (en) | Image edge point acquisition method and device | |
| Maini et al. | Study and comparison of various image edge detection techniques | |
| CN116309510B (en) | Numerical control machining surface defect positioning method and device | |
| CN109001757B (en) | Parking space intelligent detection method based on 2D laser radar | |
| CN107797560B (en) | Visual recognition system and method for robot tracking | |
| CN110660072B (en) | Method and device for identifying straight line edge, storage medium and electronic equipment | |
| CN110634137A (en) | Bridge deformation monitoring method, device and equipment based on visual perception | |
| CN106875430B (en) | Single moving target tracking method and device based on fixed form under dynamic background | |
| CN115294099A (en) | Method and system for detecting hairline defect in steel plate rolling process | |
| CN111524139B (en) | Bilateral filter-based corner detection method and system | |
| CN116503462A (en) | Method and system for quickly extracting circle center of circular spot | |
| CN116862910A (en) | Visual detection method based on automatic cutting production | |
| CN115018785A (en) | Hoisting steel wire rope tension detection method based on visual vibration frequency identification | |
| CN114674826A (en) | Visual detection method and detection system based on cloth | |
| CN116883446B (en) | Real-time monitoring system for grinding degree of vehicle-mounted camera lens | |
| CN117876468B (en) | Convex hull-based detection method and system for low-beam light type characteristic points of automobile lamplight | |
| CN113310426A (en) | Thread parameter measuring method and system based on three-dimensional profile | |
| CN117611525A (en) | Visual detection method and system for abrasion of pantograph slide plate | |
| CN110348363B (en) | Vehicle tracking method for eliminating similar vehicle interference based on multi-frame angle information fusion | |
| US7027637B2 (en) | Adaptive threshold determination for ball grid array component modeling | |
| CN114842041A (en) | Embedded-based steel rail profile detection light bar extraction method |
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 |