CN112767304B - A Vision-Based Detection Method for the Position and Orientation of Sunflower Modules - Google Patents

Abstract

The invention belongs to the technical field of detection methods, and particularly relates to a sunflower module position and direction detection method based on vision. A vision-based sunflower module position and direction detection method comprises the following steps: s1, identifying the position of a whole-layer module; s2, identifying the position of a single module, wherein the identification is carried out after the identification of the position of the 12 holes in the step S1 is carried out; and S3, identifying the direction to realize the searching of the sunflower module direction. The invention provides a vision-based sunflower module position and direction detection method which can detect the position and direction of a sunflower module through image processing, realize that a sunflower can be aligned to a sensor when being placed and facilitate the execution of a subsequent glue filling process.

Description

A Vision-Based Detection Method for the Position and Orientation of Sunflower Modules

technical field

The invention belongs to the technical field of detection methods, in particular to a method for detecting the position and direction of a sunflower module based on vision.

Background technique

As domestic labor costs continue to rise, enterprises and factories have an increasingly urgent need for machine replacement. my country is a big manufacturing country, and industrial manufacturing is inseparable from the assistance of machines. The higher the degree of automation, the more obvious our advantages in the field of manufacturing.

Vision-guided robots apply machine vision technology to robots to realize the intelligence of robots. Machine vision is to collect the image signal of the target through the image sensor and other devices, convert the image into a digital signal in the image processing system for analysis, and obtain the characteristic information of the target, such as pixel distribution, color, brightness, etc.

In the production process of street lamps, since the position of the robot arm needs to be determined according to the position of the module radiator when loading materials on the production line, and when the batch modules are piled up, the current position cannot be directly judged, so machine vision technology is needed to assist the realization Module positioning detection.

Contents of the invention

The purpose of the present invention is to provide a vision-based sunflower module position and direction detection method that detects the position and direction of the sunflower module through image processing, realizes that the sensor can be aligned when the sunflower is placed, and facilitates the subsequent glue filling process.

In order to achieve the above-mentioned purpose, the present invention takes the following technical solutions:

A method for detecting the position and direction of a sunflower module based on vision, comprising the following steps:

S1. Identifying the position of the entire layer of modules, including the following steps:

S11, take a picture, the camera shoots the whole sunflower module;

S12, preprocessing, performing histogram equalization and median filtering to remove noise on the photos obtained in step S11;

S13, threshold segmentation, forming a binarized image through threshold segmentation on the picture obtained in step S12;

S14, etching, forming closed holes by etching the image obtained in step S13;

S15, judge and return the data, and find the target area by judging the area and circularity of the hole contour for the image obtained in step S14;

S2. Position identification of a single module. After 12 hole positions are identified in step S1, the following steps are performed:

S21, take a picture, the camera shoots a single sunflower module;

S22, preprocessing, performing median filtering on the photos obtained in step S21 to remove noise;

S23, detecting, performing Hough circle detection on the sunflower module in the image obtained in step S22;

S24, finding the smallest circle, traversing all the circles for the image obtained in step S23 to find the smallest circle and returning the center position and radius;

S3. Direction recognition, realizing the search for the direction of the Sunflower module, including the following steps:

S31, highlighting the edge of the image, using the difference between the binarized image and the corrosion map for the image obtained in step S21 to represent the edge profile of the image;

S32, mask processing, for displaying the largest gap, setting the position of the center of the smallest circle obtained in step S24 as the position of the center of the circle and multiplying it with the image to be processed to obtain the ROI image;

S33, edge detection, using the Canny edge detection algorithm to identify the image obtained in step S32 to detect the actual edge;

S34. Expansion processing, expanding the highlighted part of the image obtained in step S33;

S35. Find the maximum contour, traverse all the contours in the image obtained in step S34 and obtain the maximum contour;

S36. Convex hull and convex defect detection, performing convex hull and convex defect detection on the image of the largest contour obtained in step S35, to obtain the feature value of the convex defect area;

S37. Determine the maximum opening, calculate and compare the feature quantities obtained in step S36, and find the opening direction of the sunflower module.

Among them, the Canny edge detection algorithm is a multi-level edge detection algorithm developed by John F.Canny in 1986. Usually, the purpose of edge detection is to significantly reduce the data size of the image while retaining the original image attributes. There are various algorithms for edge detection, although the Canny algorithm is old, it can be said that it is a standard algorithm for edge detection, and it is still widely used in research.

On the basis of adopting the above-mentioned technical solution, the present invention can also adopt the following further technical solutions:

The step S11 also includes: the camera moves with the mechanical arm to a place 1 meter above the entire sunflower module, triggering the camera to take pictures;

The size of the corroded convolution kernel in step S14 is 120*120.

The step S12 also includes:

S121, histogram equalization, performing histogram equalization processing on the photos obtained in step S11 to enhance image contrast;

S122, median filtering, performing median filtering processing on the image obtained in step S121 to eliminate image noise.

The step S13 also includes: calculating the total number of pixels and the probability of occurrence of the gray value of the picture obtained in the step S12, and performing binarization processing on the image, making the pixel value greater than the threshold value 255, and making the pixel value smaller than the threshold value 0 .

The step S15 also includes:

S151. Detect all contours in the image obtained in step S14 and save them in the container, calculate the area and perimeter of the contours and calculate the circularity based on this, if the area is greater than 12500 and less than 92500, and the circularity is greater than 0.65, Then it is judged that the contour is the target area;

S152. Traversing all the target areas satisfying the conditions obtained in step S151, finding the smallest circle enclosing the outline, and returning the center position and radius of the smallest circle.

The step S21 also includes: the camera moves to the center position of each circle sequentially under the drive of the mechanical arm and takes pictures, and the captured image is a grayscale image;

In the step S22, a 3*3 convolution kernel is used to perform median filtering.

The expanded convolution kernel in step S34 is 19*19.

The step S35 also includes: finding all the contours in the image obtained in the step S34 and storing them in the container sequentially, and traversing all the contours to find the maximum value.

The feature quantity in the step S36 includes the starting point, the ending point, the furthest point, and the distance from the furthest point to the convex defect.

The step S37 also includes: calculating the distance between the starting point and the ending point of each convex defect, and storing it in the container, and comparing the calculated distance to find the area where the maximum distance is located, which is the area of the Sunflower module. Opening direction.

The present invention has the following technical characteristics:

The invention automatically recognizes the position and direction of the sunflower module, which is convenient for grabbing and rotating to a designated direction, and can realize that the sunflower can be aligned with the sensor when it is placed, and is convenient for the execution of the subsequent glue filling process. The method of the invention has high precision, strict process and extremely low error rate.

Description of drawings

Fig. 1 is an image obtained in step S11 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

FIG. 2 is a gray histogram of step S121 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

FIG. 3 is an image obtained in step S121 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

FIG. 4 is an image obtained in step S122 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

Fig. 5 is an image obtained in step S13 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

FIG. 6 is an image obtained in step S14 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

FIG. 7 is an image obtained in step S21 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

FIG. 8 is an image obtained in step S22 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

FIG. 9 is an image obtained in step S23 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

FIG. 10 is an image obtained in step S31 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

FIG. 11 is a mask image obtained in step S32 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

FIG. 12 is an ROI image obtained in step S32 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

FIG. 13 is an image obtained in step S33 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

FIG. 14 is an image obtained in step S34 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

Fig. 15 is an image obtained in step S35 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

Fig. 16 is a maximum contour convex hull diagram obtained in step S36 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

Fig. 17 is a maximum contour convex defect map obtained in step S36 of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

Fig. 18 is a flowchart of a method for detecting the position and direction of a sunflower module based on vision in the present invention.

Fig. 19 is a schematic diagram of the sunflower module structure of a method for detecting the position and direction of the sunflower module based on vision in the present invention.

Detailed ways

The following specific examples are further descriptions of the methods and technical solutions provided by the present invention, but should not be construed as limiting the present invention.

Embodiment 1, as shown in Figures 18 and 19, a method for detecting the position and direction of a sunflower module based on vision includes the following steps:

S1. Identifying the position of the entire layer of modules, including the following steps:

S11. Taking pictures, the camera moves with the mechanical arm to 1m above the entire layer of the module, triggering the camera to take pictures. The camera selection is obtained by detecting the position and direction of a single module, and the shooting effect on the workbench is shown in Figure 1.

S12, preprocessing, performing histogram equalization and median filtering to remove noise on the photos obtained in step S11, including the following steps:

S121. Histogram equalization, performing histogram equalization processing on the photos obtained in step S11 to enhance image contrast.

Histogram equalization is to convert the input image into an output image with approximately the same number of pixels on each gray level through some grayscale mapping. As shown in Figure 1, it can be seen that the overall image is dark and the details are not clear. Prominent, so first use histogram equalization to increase the contrast of the image, so that you can see the details of the image. The gray histogram is a two-dimensional image, as shown in Figure 2, the abscissa is the gray level of each pixel in the image, and the ordinate is the number or frequency of pixels with each gray level appearing in the image.

Specifically, it is first assumed that the gray scale of the image ranges from 0 to 1 and is continuous, then the normalized histogram of the image is the probability density function.

p(x),0≤x≤1

According to the properties of the probability density function, we can get:

Suppose the image probability density function before conversion is p _r (r), the image probability density function after conversion is p _s (s), and the mapping relationship of the conversion function is s=f(r). Among them, r and s are the gray value of the input image and the output image at any point, respectively. Then from probability theory:

If you want to make the converted probability density function p _s (s)=1, 0≤s≤1 (that is, the histogram is uniform), it must satisfy:

Integrating both sides of the equation with respect to r, we get:

For the case where the gray value is in [0,255], it needs to be multiplied by the maximum gray value D _max , at this time, the conversion formula of gray balance is

Among them, D _B is the gray value after conversion, and D _A is the gray value before conversion.

S122, median filtering, performing median filtering processing on the image obtained in step S121 to eliminate image noise.

After the histogram equalization, it can be seen that the image noise is obvious, so the median filter is used to eliminate the image noise, which is convenient for the subsequent processing of the image. Median filtering is a statistical sorting filter. For a point in the original image, median filtering takes the median of the statistical sorting of all pixels in the neighborhood centered on the changed point as the response of the point.

In this embodiment, a 3*3 median filter is used to scan each pixel of the image, and the value at the fifth place replaces the original pixel value. Compared with the linear filter, this method can better eliminate the influence of salt and pepper noise, because the linear filter needs to calculate the noise, but in the median filter, the noise is directly ignored.

S13. Threshold segmentation. The image obtained in step S12 is subjected to threshold segmentation to form a binary image. Here, threshold segmentation is performed based on the method of maximum variance between classes.

It can be seen that the gray value of the target area is very different from the surrounding black circle, so threshold segmentation can be used to filter out the target area. However, due to uneven illumination, for example, under the strong light on the right, the gray value of the black area has obvious changes compared with other similar areas, so it is not possible to use a uniform threshold for segmentation of all captured images.

各灰度值出现的概率为

把图像中的像素用阈值T分为两类，A₁和A₂，A₁是由灰度值在0～T-1的像素组成，概率为

平均灰度

其中μ是整幅图像的平均灰度。A₂是由灰度值在T～255的像素组成，概率为

平均灰度

则

A₁和A₂两个区域的总方差σ＝P₁P₂(μ₁-μ₂)²。通过计算，求出使得σ最大的T值。对图像进行二值化，大于阈值的令像素值为255，小于阈值的为0。Specifically, as shown in FIG. 5 , the total number of pixels and the occurrence probability of each gray value are calculated. Traverse the image, count the number of pixels whose gray level is i, and record it as n _i , the range of i is 0-255, the total number of pixels

The probability of occurrence of each gray value is

Divide the pixels in the image into two categories with a threshold T, A ₁ and A ₂ , A ₁ is composed of pixels with gray values between 0 and T-1, and the probability is

average gray

where μ is the average gray level of the entire image. A ₂ is composed of pixels whose gray value is T~255, the probability is

average gray

but

The total variance σ=P ₁ P ₂ (μ ₁ −μ ₂ ) ² of the two areas of A ₁ and A ₂ . By calculation, find the T value that makes σ the largest. The image is binarized, and the pixel value greater than the threshold is 255, and the pixel value smaller than the threshold is 0.

S14, etching, forming closed holes by etching the image obtained in step S13.

As shown in Figure 6, the binarized graph is corroded to obtain the closed area of the circular target. The corroded convolution kernel size is 120*120.

S15, judge and return the data, and find the target area by judging the area and circularity of the hole contour on the image obtained in step S14.

S151. Detect all contours in the image obtained in step S14 and save them in a container.

The contour shape judgment method can be based on the equiperimeter theorem, in closed geometric shapes with equal perimeter lengths, the area of the circle is the largest. Inequality can be expressed as: 4πS≤L^2. In the formula, S is the area enclosed by the curve, and L is the perimeter of the closed curve. Define the circularity as 4πS/L^2. The area S of the contour can be calculated by the number of pixels occupied by the connected domain.

For contour i, calculate its area S and perimeter L, so as to calculate the circularity R=4πS/L ² , if S can satisfy 12500<S<92500, and R satisfies R>0.65, then the contour is considered to be circular , the target area.

S152. Traversing all the target areas satisfying the conditions obtained in step S151, finding the smallest circle enclosing the outline, and returning the center position and radius of the smallest circle.

S2. Position identification of a single module. After 12 hole positions are identified in step S1, the following steps are performed:

S21. Taking pictures, the camera shoots a single sunflower module.

After identifying the positions of the 12 holes, the camera moves to the center of each circle in turn under the drive of the robotic arm and takes pictures.

Specifically, camera selection: resolution calculation: the detection accuracy is 0.1mm, and the diameter of the sunflower module is 160mm, so the best field of view is about 200*200mm, and the minimum resolution is 4 million. Due to practical applications, the calculated value is generally selected 3-5 times to ensure the accuracy, the actual resolution is at least 12 million.

Therefore, the model of Daheng Image was finally selected as MER-1220-32U3M-C, a GigE industrial camera with a resolution of 12.2 million pixels.

The captured image is a grayscale image, and the shooting effect at the working position 20cm above the module is shown in Figure 7.

S22. Preprocessing, performing median filtering on the photo obtained in step S21 to remove noise. As shown in FIG. 8 , a 3*3 convolution kernel is used to perform median filtering, specifically as described in step S1.

S23, detection, performing Hough circle detection on the sunflower module in the image obtained in step S22.

Carry out Hough circle detection on the Sunflower module, and after the test, it is determined that the minimum distance between the circle and the center of the circle is 45, the minimum radius of the circle is 200, and the maximum is 250. The effect after detection is shown in Figure 9.

S24. Find the smallest circle, traverse all the circles in the image obtained in step S23 to find the smallest circle and return the center position and radius of the circle.

It can be seen that after the Hough circle detection, all circles surround the target area, and the smallest circle is the desired one. Therefore, traverse all found circles R ₁ , R ₂ ...R _n , find the smallest circle R _i , and return its center position (xi _, y _i ) and radius r _i .

So far, the location of a single Sunflower module has been found, and the following steps are performed to find the direction of the Sunflower module.

S3. Direction recognition is carried out synchronously with step S2 to realize the search for the direction of the Sunflower module, including the following steps:

S31. Highlighting the edge of the image, using the difference between the binarized image and the erosion map to represent the edge profile of the image obtained in step S21.

In order to reduce unnecessary noise and obtain a clear outline of the image, the edges of the clumps are highlighted, and the difference between the binarized image and the erosion map is used to represent the internal gradient of the image, that is, the edge outline of the image. As shown in Figure 10, the internal gradient is a kind of morphological gradient, which is a difference image obtained by subtracting the corroded image from the original image, which is called the internal gradient of the image.

S32. Mask processing, for displaying the largest gap, setting the position of the center of the smallest circle obtained in step S24 as the position of the center of the circle and multiplying it with the image to be processed to obtain an ROI image.

In order to show the largest gap and remove the redundant sunflowers, the central area of the image is taken to prepare for the subsequent search for the largest opening. The mask operation can extract the ROI area. When making a mask, as shown in Figure 11, the center position is the minimum circle center position ( _xi , y _i ) after Hough circle detection. Multiply it with the image to be processed, as shown in Figure 12, to get the ROI image, the image value in the ROI remains unchanged, and the outer pixel value is set to 0.

S33. Edge detection. As shown in FIG. 13 , the Canny edge detection algorithm is used to identify the image obtained in step S32 to detect actual edges.

In order to identify the edge of the image, the Canny edge detection algorithm can be used to detect the actual edge as much as possible, while reducing the false positives caused by noise as much as possible.

Specifically, to remove noise, use Gaussian smoothing filter convolution denoising

Calculate the gradient magnitude and direction The edges of the image can point to different directions, so the classic Canny algorithm uses four gradient operators to calculate the gradients in the horizontal, vertical and diagonal directions respectively. However, four gradient operators are usually not used to calculate the four directions separately. In this paper, the Sobel operator is used to calculate the horizontal and vertical gradients.

A is the original image pixel matrix.

Finally get the gradient magnitude

gradient direction

Non-maximum suppression, non-maximum suppression is an edge refinement method. Usually the obtained gradient edge is not only one pixel wide, but multiple pixels wide, so such a gradient map is still very "fuzzy". Non-maximum suppression can help preserve the local maximum gradient while suppressing all other gradient values. This means that only the sharpest locations in the gradient change are preserved. Algorithm: 1) Compare the gradient strength of the current point with the gradient strength of the positive and negative gradient direction points. 2) If the gradient strength of the current point is the largest compared with the gradient strength of other points in the same direction, keep its value. Otherwise suppressed, ie set to 0. For example, the direction of the current point points to the 90° direction directly above, then it needs to be compared with the vertical direction, the pixels directly above and directly below it.

hysteresis threshold. The hysteresis threshold requires two thresholds, a high threshold and a low threshold: 1) If the magnitude of a certain pixel position exceeds the high threshold, the pixel remains as an edge pixel. 2) If the magnitude of a certain pixel position is less than the low threshold, the pixel is excluded. 3) If the magnitude is between two thresholds, the pixel is only kept if it is connected to a pixel above the high threshold. Generally, the ratio of high and low thresholds is between 2:1 and 3:1. In this embodiment, the effect is better when the high threshold is 59 and the low threshold is 13.

S34. Expansion processing, expanding the highlighted part of the image obtained in step S33.

Because there are many breakpoints in the processed edge, dilation is used. Expansion is the expansion of the highlighted part of the image, as shown in Figure 14, which has a larger highlighted area than the original image. The expanded convolution kernel is 19*19.

S35. Find the maximum contour, traverse all the contours in the image obtained in step S34 and obtain the maximum contour.

Currently there are several contours in the image, and the largest contour needs to be found for convex hull and convex defect detection. A contour generally corresponds to a series of points, that is, a curve in the image. Firstly, it is necessary to find all the contours C ₁ , C ₂ ...C _n and store them in the container in turn, then traverse all the contours and find the maximum value C _i to get the maximum contour, as shown in Figure 15.

S36. Convex hull and convex defect detection, performing convex hull and convex defect detection on the image of the largest contour obtained in step S35, so as to obtain the feature value of the convex defect area.

The largest contour convex hull diagram shown in Figure 16 and the largest contour convex defect diagram shown in Figure 17, it can be seen that the largest contour contains many gaps and protrusions. In order to get the maximum opening and then the direction of the sunflower, the convex hull and convex defect detection are performed on the image.

Convex Hull and Convex Defect Detection provide a good understanding of the shape or contour of an object. For a given two-dimensional plane point set, the convex hull is a convex polygon formed by connecting the outermost points, which can contain all points in the point set. A convex defect refers to a concave part in a section of contour. Each convex defect area has four characteristic quantities: start point, end point, distance from the farthest point, and distance from the farthest point to the convex defect.

S37. Determine the maximum opening, calculate and compare the feature quantities obtained in step S36, and find the opening direction of the sunflower module.

Calculate the distance L ₁ , L ₂ ...L _n between the start point and the end point of each convex defect, store it in the container, compare these distances, find the maximum distance L _i , and the area where L _i is located is the Sunflower module Opening direction. The descriptions of the above embodiments are only used to help understand the method of the present invention and its core idea. It should be pointed out that for those skilled in the art, without departing from the principles of the present invention, some improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

Claims (10)

1. A vision-based sunflower module position and direction detection method is characterized by comprising the following steps:

s1, identifying the position of the whole-layer module, comprising the following steps:

s11, photographing, wherein a camera shoots the whole layer of sunflower module;

s12, preprocessing, namely performing histogram equalization and median filtering noise removal processing on the picture obtained in the step S11;

s13, threshold segmentation, namely forming a binary image on the picture obtained in the step S12 through threshold segmentation;

s14, etching, namely etching the image obtained in the step S13 to form a closed hole;

s15, judging and returning data, and finding a target area for the image obtained in the step S14 by judging the area and the circularity of the hole outline;

s2, identifying the position of a single module, and after identifying the position of 12 holes in the step S1, performing the following steps:

s21, taking a picture, wherein a camera shoots a single sunflower module;

s22, preprocessing, namely performing median filtering on the picture obtained in the step S21 to remove noise;

s23, detecting, namely performing Hough circle detection on the sunflower module in the image obtained in the step S22;

s24, finding the smallest circle, traversing all the circles of the image obtained in the step S23 to find the smallest circle, and returning to the circle center position and the radius; and

s3, direction identification is carried out, the sun flower module direction searching is achieved, and the method comprises the following steps:

s31, highlighting the edge of the image, and representing the edge contour of the image by using the difference between the binarized image and the erosion map for the image obtained in the step S21;

s32, mask processing is carried out, the maximum notch is shown, the position of the center of the smallest circle obtained in the step S24 is set as the position of the circle center, and the position is multiplied by the image to be processed to obtain an ROI image;

s33, edge detection, namely identifying the image obtained in the step S32 by adopting a Canny edge detection algorithm to detect an actual edge;

s34, expansion processing, namely expanding the highlight part in the image obtained in the step S33;

s35, searching a maximum contour, traversing all contours in the image obtained in the step S34 and obtaining the maximum contour;

s36, detecting convex hulls and convex defects, namely detecting the convex hulls and the convex defects of the image with the maximum outline obtained in the step S35 to obtain the characteristic quantity of a convex defect area;

and S37, determining the maximum opening, calculating the characteristic quantity obtained in the step S36, comparing the characteristic quantity, and finding the opening direction of the sunflower module.

2. The vision-based sunflower module position and orientation detection method of claim 1, wherein said step S11 further comprises: the camera moves to a position 1m above the whole layer of sunflower module along with the mechanical arm, and the camera is triggered to take a picture;

the size of the convolution kernel eroded in step S14 is 120 × 120.

3. The vision-based sunflower module position and orientation detection method of claim 1, wherein said step S12 further comprises:

s121, histogram equalization, namely performing histogram equalization processing on the photo obtained in the step S11 to enhance the image contrast;

and S122, median filtering, namely performing median filtering processing on the image obtained in the step S121 to eliminate image noise.

4. The vision-based sunflower module position and orientation detection method of claim 1, wherein said step S13 further comprises: and (4) calculating the total pixel number and the probability of occurrence of the gray value of the picture obtained in the step (S12), and performing binarization processing on the picture, wherein the pixel value larger than the threshold is 255, and the pixel value smaller than the threshold is 0.

5. The vision-based sunflower module position and orientation detection method of claim 1, wherein said step S15 further comprises:

s151, detecting all contours in the image obtained in the step S14, storing the contours in a container, calculating the area and the perimeter of the contours, calculating the circularity according to the area and the perimeter, and if the area is larger than 12500 and smaller than 92500 and the circularity is larger than 0.65, judging that the contours are target areas;

s152, traversing all the target areas meeting the conditions obtained in the step S151, finding the minimum circle surrounding the outline, and returning to the center position and the radius of the minimum circle.

6. The vision-based sunflower module position and orientation detection method of claim 1, wherein step S21 further comprises: the camera is driven by the mechanical arm to sequentially move to the circle center position of each circle and take a picture, and the taken picture is a gray scale image;

in step S22, a convolution kernel of 3*3 is used for median filtering.

7. The vision-based sunflower module position and orientation detection method of claim 1, wherein the convolution kernel of the expansion in step S34 is 19 x 19.

8. The vision-based sunflower module position and orientation detection method of claim 1, wherein said step S35 further comprises: and finding all the contours in the image obtained in the step S34, sequentially storing the contours in a container, and traversing all the contours to find the maximum value.

9. The vision-based sunflower module position and orientation detection method of claim 1, wherein said feature parameters in step S36 include a starting point, an ending point, a farthest point, and a distance from the farthest point to the convex defect.

10. The vision-based sunflower module position and orientation detection method of claim 9, wherein step S37 further comprises: and calculating the distance between the starting point and the ending point of each convex defect, storing the distance into a container, and comparing the calculated distances to find the area where the maximum distance is located, wherein the area is the opening direction of the sunflower module.

Images