CN111528579A - Non-contact foot measuring method based on machine vision - Google Patents

Abstract

The invention relates to a non-contact foot measurement method based on machine vision, which includes camera calibration and visual foot measurement. The camera calibration is used to determine measurement area and scale information and to calculate parameters of transmission transformation; the visual measurement foot is obtained from the camera calibration. Based on the parameters, measure the measured value of the person's foot. Based on machine vision and combined with image processing technology, the invention realizes non-contact measurement of foot length and foot width by extracting the complete foot contour of the foot, which not only greatly shortens the measurement time, improves the measurement accuracy, but also has strong environmental adaptability Ability to work normally in different lighting environments.

Description

A non-contact foot measurement method based on machine vision

technical field

The invention relates to a non-contact foot measuring method, in particular to a non-contact foot measuring method based on machine vision.

Background technique

With the development of the times and the progress of society, people's requirements for the comfort of shoes and boots are getting higher and higher. In the 1960s, our country conducted two nationwide foot surveys, using manual measurement methods. The foot shape of 250,000 people was measured and analyzed, and shoe size standards and shoe last samples suitable for Chinese were formulated. Traditional shoes and boots can be divided into men's shoes, women's shoes and children's shoes according to gender and age. According to the length of the human foot The yardage is divided, and the adjacent yardage differs by 5 mm. However, the shape of the feet in the foot length range corresponding to the same yard length is very different. Even the left and right feet of the same person are not completely mirror-symmetrical, which makes it difficult to determine the appropriate shoe size based solely on the foot length indicator. . Under this circumstance, many industrial production lines are transforming from large-scale batch production to user-customized, personalized, and small batch production. The personalized customization of shoes and boots relies on the accurate measurement of the characteristic data of the foot shape. The traditional measurement method of foot shape parameters mainly relies on manual measurement. This measurement method is not only inefficient, cumbersome in the process, and high in labor costs, but also with different measurement personnel. There are certain differences in the measured dimensions, so it cannot meet the needs of batch measurement, precise measurement and fast measurement. In order to solve this problem, foot measurement is gradually changing from contact measurement to non-contact measurement, among which non-contact measurement mainly includes line laser scanning method and stereo vision method.

The advantage of the line laser scanning method is that it has better 3D measurement accuracy, but there are still some bottlenecks in application and promotion, mainly because the current 3D foot measurement equipment is expensive, which is not conducive to the use of small and medium-sized enterprises. The scanning time is long, resulting in poor user experience; based on the stereo vision method, the 3D information of the foot shape can also be obtained, that is, the 3D model is constructed by image analysis, image reconstruction and other methods. The shooting task can be completed in less than a second, which is suitable for occasions requiring fast measurement. However, the stereo vision method needs to match two images, and the matching and measurement accuracy are easily affected when the surface grayscale and surface shape of the object change little. Moreover, in real life, only the data of foot length and foot width can meet the needs of ordinary users for shoe selection. Other size parameters such as arch height, foot circumference, and toe length are also characteristic information of foot shape, but It has almost no effect on the selection of shoes for ordinary users. Therefore, in view of the current requirements of automatic, low-cost, high-precision and easy-to-operate foot shape parameter measurement, the present invention is based on the machine vision method, combined with image processing technology, and realizes foot length and foot shape by automatically extracting the complete foot shape contour of the foot. Wide non-contact measurement.

There are the following two types in the prior art:

1) Line laser foot measuring instrument

The scheme uses infrared scanning technology to collect laser scanning images of the contour of the foot surface through binocular cameras, uses the scanner to scan the images of the soles of the feet, and then uses image processing technology to reconstruct the foot shape and calculate the parameters of the foot shape, so as to measure the length of the human foot, Height, width, arch height and other data (Figure 18). This method of using line laser scanning to acquire the shape of the foot shape has the advantage that the acquired model has high precision. However, a large amount of data needs to be stored and processed during the movement of the scanner, resulting in a long measurement time and a scan time of up to 10 seconds for a single foot, resulting in poor user experience. At the same time, during a long measurement process, the position of the human foot may move slightly, which will inevitably cause measurement errors. In addition, high-quality laser generators and their transmission devices are very expensive, which is not conducive to popularization and use in small and medium-sized enterprises. Therefore, the method of measuring foot shape by line laser scanning is limited in application.

2) 3D foot measurement technology based on stereo vision

The foot shape images are collected from different angles mainly through the binocular camera, and then the 3D model of the foot shape is restored by means of image analysis, visual reconstruction and geometric processing, so as to realize the measurement of the foot shape parameters. This stereo vision-based three-dimensional foot measurement solution has greatly improved the image capture time, and can complete the image acquisition of the foot without blind spots in an instant, avoiding measurement errors caused by jitter. However, the two key technologies of calibration technology and stereo matching technology in this scheme are easily affected by the light source, the environmental adaptability is weak, and the surface of the human foot is relatively smooth, which contains less texture features, and the calibration and matching accuracy is good. Not guaranteed.

To sum up, in view of the defects in the current technical solutions for foot measurement, such as high price, too long measurement time, unstable accuracy and weak environmental adaptability, it is necessary to further improve or innovate the measurement technology.

SUMMARY OF THE INVENTION

In view of the problems existing in the above background technology, the present invention proposes a non-contact measurement of foot length and foot width by extracting the complete foot contour of the foot based on machine vision and image processing technology, which not only greatly shortens the measurement time , which improves the measurement accuracy, has strong environmental adaptability, and can work normally in different lighting environments.

The technical scheme of the present invention is as follows:

The above-mentioned non-contact foot measurement method based on machine vision includes camera calibration and visual foot measurement. The camera calibration is used to determine the measurement area and scale information and calculate the parameters of the transmission transformation; the visual foot measurement is based on the parameters obtained by the camera calibration. Measure the measurement of the person's feet.

The non-contact foot measurement method based on machine vision, wherein: the measured value of the measured person's foot includes the foot length and the foot width; Standing with both feet in an inward stance or an outward stance of more than 10° will be identified as an abnormal situation; the visual foot measurement does not require the use of a calibration board, and the subject stands with both feet on the footboard.

In the non-contact foot measurement method based on machine vision, the main execution process of the camera calibration is: (1.1) placing a calibration board marked with a specific color on the foot pedal; (1.2) using the camera to take a picture The image with the calibration plate, then convert the image from RGB color space to HSV color space, and extract the green area of the calibration plate as the measurement area through the color space threshold segmentation method; (1.3) Use the contour detection algorithm to find the green calibration According to the four corner points of the contour and the actual size of the calibration plate, the image perspective transformation matrix parameters and scale information are calculated; (1.4) The transmission transformation matrix parameters and scale information of the calibration plate image are saved.

The machine vision-based non-contact foot measurement method, wherein: the contour detection algorithm in the step (1.3) adopts the FindContours() function in OpenCV.

The non-contact foot measurement method based on machine vision, wherein: in the step (1.1), the color of the foot pedal is red, and the calibration plate is green.

In the non-contact foot measuring method based on machine vision, the camera calibration only needs to be performed once during the camera initialization process before the foot measurement, unless the camera is offset or the foot area changes, the calibration does not need to be performed again.

The non-contact foot measurement method based on machine vision, wherein: the specific processing steps of the visual foot measurement are:

(2.1) Use a camera to take pictures with the feet to be tested, and the pictures taken are oblique views;

(2.2) Perform a transmission transformation on the picture taken in the above step (2.1) according to the parameters obtained by the camera calibration, and convert it into a top view to ensure that the pixel scale of the picture is uniform in the top view, and then according to the length and width of the foot area and The ratio of pixel points to calculate the value of foot length and foot width;

(2.3) Use the Canny operator to extract the edge information of the foot from the image after transmission transformation, and use the Gaussian blur algorithm to remove the noise next to the edge, and convert the image into the contour map of the foot. When extracting the edge information, the Canny operator is maximized The threshold is set to 0, that is, all edge information is retained;

(2.4) After extracting the outline of the foot, then extract the minimum circumscribed rectangle of the outline, extract the width of the left and right foot regions and the distance from the toes of the two feet to the upper boundary of the area, and judge various measurements through the information of the minimum circumscribed rectangle abnormal situation;

(2.5) After the processing of the above steps (2.1)-(2.4), the detection and distance calculation model can be obtained, taking the right foot as an example, where W ₁ represents the width of the standing foot area, W ₂ is the width of the right foot to be measured, H ₁ is the length of the standing area, H ₂ is the distance from the toe of the right foot to the upper boundary of the area; C ₁ , C ₂ , D ₁ and D ₂ represent the number of pixels; since W ₁ and H ₁ are known quantities, and the pixels The scale is fixed, and the calculation method of foot width and foot length can be obtained:

The length of the foot is:

H _foot = H ₁ -H ₂ (6).

In the non-contact foot measurement method based on machine vision, the determination of various abnormal measurement conditions in the step (2.4) specifically includes:

First of all, according to the number of detected minimum circumscribed rectangles, two anomalies such as no feet and obstacles can be detected; if the minimum circumscribed rectangle is not detected at this time, it can be judged that the camera has not captured an image with feet; if If the detected minimum circumscribed rectangle is greater than 2, it can be judged that the collected image has other obstacles similar to feet;

Secondly, if the number of the minimum circumscribed rectangles is correct, the standing situation of the tested person can be judged according to the angle of the minimum circumscribed rectangle. If the angle of the rectangle is within the range of -80°～-45°, it is judged that the subject's feet are standing in the outer eight position.

In the non-contact foot measurement method based on machine vision, in the step (2.4), the MinAreaRect() method of OpenCV is used to extract the minimum circumscribed rectangle of the contour.

Beneficial effects:

The non-contact foot measurement method based on machine vision of the present invention has a reasonable conception and simple operation process. Based on machine vision and combined with image processing technology, the non-contact measurement of foot length and foot width can be realized by extracting the complete foot contour of the foot, which can effectively Meet the low-cost, high-efficiency, high-precision foot parameter measurement requirements.

Compared with line laser scanning and stereo vision methods, the non-contact foot measurement method based on machine vision of the present invention uses optical imaging technology, does not involve complex algorithms such as three-dimensional matching, and optimizes image processing algorithms. The invention of the non-contact foot measurement method based on machine vision not only greatly shortens the measurement time (the time for measuring feet is 1 second), but also improves the measurement accuracy (accuracy can reach 0.2mm), and has strong environmental adaptability, It can work normally in different lighting environments; in addition, the foot measuring device involved in the machine vision-based non-contact foot measuring method of the present invention has a simple structure, and only needs a common camera and a foot pedal to measure, which greatly reduces the It reduces the production cost and is conducive to the promotion and use of small and medium-sized enterprises.

The present invention is a non-contact foot measuring method based on machine vision, uses a single high-precision color camera to collect the original image, and accurately extracts the length and width information of the foot through the image processing algorithm. Due to the defects of lower, unstable accuracy and weak environmental adaptability, the invention reduces the measurement cost and improves the measurement efficiency and accuracy. In addition, by adding a light source, it can ensure that the brightness of the image collected each time is uniform and sufficient, and the environmental adaptability is improved.

The non-contact foot measuring method based on machine vision of the present invention, compared with the existing foot measuring method, greatly improves the measurement efficiency, at the same time ensures the measurement accuracy and reduces the measurement cost. The foot measuring device involved in the foot measuring method is easy to maintain and has a long service life.

Description of drawings

Fig. 1 is the flow chart of camera calibration in the non-contact foot measuring method based on machine vision of the present invention;

Fig. 2 is the original picture taken by the calibration plate in step (1.2) of camera calibration in the non-contact foot measuring method based on machine vision of the present invention;

3 is a schematic diagram of the green calibration plate area in step (1.2) of camera calibration in the non-contact foot measuring method based on machine vision of the present invention;

FIG. 4 is a corner detection diagram in step (1.3) of camera calibration in the non-contact foot measurement method based on machine vision of the present invention;

Fig. 5 is the flow chart of visual foot measurement in the non-contact foot measurement method based on machine vision of the present invention;

6 is the original picture taken with feet in step (2.1) of visual foot measurement in the non-contact foot measurement method based on machine vision of the present invention;

Fig. 7 is the top view picture after perspective transformation in step (2.2) of visual foot measurement in the non-contact foot measurement method based on machine vision of the present invention;

Fig. 8 is a foot outline diagram in step (2.3) of visual foot measurement in the non-contact foot measurement method based on machine vision of the present invention;

9 is a minimum circumscribed rectangle diagram in step (2.4) of visual foot measurement in the non-contact foot measurement method based on machine vision of the present invention;

10 is a schematic diagram of distance calculation in step (2.5) of visual foot measurement in the non-contact foot measurement method based on machine vision of the present invention;

11 is a flowchart of the implementation of visual foot measurement in the non-contact foot measurement method based on machine vision of the present invention;

12 is a low-brightness stance diagram in Example 1 of the machine vision-based non-contact foot measurement method of the present invention;

13 is a block diagram of a low-brightness foot shape in Example 1 of the machine vision-based non-contact foot measurement method of the present invention;

14 is a middle-brightness stance diagram in Example 2 of the non-contact foot measurement method based on machine vision of the present invention;

15 is a block diagram of a mid-brightness foot shape in Example 2 of the machine vision-based non-contact foot measurement method of the present invention;

FIG. 16 is a high-brightness standing foot diagram in Example 3 of the non-contact foot measurement method based on machine vision of the present invention;

17 is a block diagram of a high-brightness foot shape in Example 3 of the machine vision-based non-contact foot measurement method of the present invention;

FIG. 18 is a schematic structural diagram of a conventional line laser foot type measuring instrument.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments.

The non-contact foot measurement method based on machine vision of the present invention includes camera calibration and visual foot measurement. The camera calibration is used to determine the measurement area and scale information. At the same time, because most foot measurement devices cannot install the camera vertically, the camera calibration also needs to be calculated. The parameters of transmission transformation; based on the parameters obtained by camera calibration, the visual measuring foot measures the measured value of the subject's feet (including foot length and foot width). Parallel standing (inner stance or outer stance over 10° will be identified as abnormal); the image processing algorithm of the present invention can be implemented using OpenCV.

As shown in Figure 1, the main execution flow of the above camera calibration is as follows:

(1.1) Place a calibration board marked with a specific color on the foot pedal; in the present invention, the color of the foot pedal is red, and the calibration board is green, and other distinguishable colors can also be used;

(1.2) Use the camera to take an image with a calibration board (as shown in Figure 2), and then convert the image from the RGB (Red, Green, Blue) color space to the HSV (Hue, Saturation, Value) color space, Through the color space threshold segmentation method, the green area of the calibration plate is extracted (as shown in Figure 3), and this area is the measurement area;

(1.3) Use the contour detection algorithm (FindContours() function in OpenCV) to find the outline of the green calibration plate, and then extract the four corner points of the contour (as shown in Figure 4) and the actual size of the calibration plate, and calculate the image perspective transformation matrix parameters and scale information;

(1.4) Save transmission transformation matrix parameters and scale information.

The above camera calibration only needs to be performed once during the camera initialization process before the foot measurement. Unless the camera is offset or the foot area changes, there is no need to perform the calibration again.

The above-mentioned visual foot measurement does not require the use of a calibration board, and the subject stands with both feet on the red (other colors can also be used) footboards;

As shown in Figure 5, the main processing steps of the above-mentioned visual foot measurement process are:

(2.1) Use the camera to take a picture with the feet to be tested, as shown in Figure 6, the picture taken is an oblique view.

(2.2) Perform a transmission transformation on the picture taken in the above step (2.1) according to the parameters obtained by the camera calibration, and convert it into a top view (as shown in Figure 7), so as to ensure that in the top view, the pixel scale of the picture is uniform, so The values of foot length and foot width (ie, W ₂ and H ₂ in step 2.5) can be calculated according to the ratio of the length and width of the standing foot area to the pixel points.

(2.3) Use the Canny operator to extract the edge information of the foot from the transmission transformed picture, and use Gaussian blur to remove the noise next to the edge; convert the picture into the outline of the foot (as shown in Figure 8); when extracting the edge information , the maximum threshold of Canny operator is set to 0, that is, all edge information is retained, which effectively improves the accuracy of edge information.

(2.4) After extracting the outline of the foot, extract the minimum circumscribed rectangle of the outline (using the MinAreaRect() method of OpenCV, the result is shown in Figure 9); extract the width of the left and right foot areas and the toe-to-toe area of the two feet The distance of the upper boundary; through the information of the minimum circumscribed rectangle, various measurement anomalies can be judged. Among them, judging a variety of measurement anomalies specifically includes: first, according to the number of detected minimum circumscribed rectangles, two anomalies can be detected: no feet and obstacles; if the minimum circumscribed rectangle is not detected at this time, the camera can be judged There is no image with feet collected; if the detected minimum circumscribed rectangle is more than 2, it can be judged that the collected image has other obstacles similar to feet; secondly, when the number of the minimum circumscribed rectangle is correct, it can be determined according to the minimum circumscribed rectangle If the angle of the rectangle is within the range of (10°, 45°), it is judged that the subject's feet are standing in the inner eight position; if the angle of the rectangle is within ( -80°, -45°), it is judged that the subject's feet are standing in the outer eight position.

(2.5) After the previous processing steps, the detection and distance calculation model as shown in Figure 10 can be obtained (taking the right foot as an example), where W ₁ represents the width of the standing foot area, W ₂ is the width of the right foot to be measured, and H ₁ is the length of the standing area, H ₂ is the distance from the toe of the right foot to the upper boundary of the area (because of the occlusion problem at the heel position, it cannot be directly measured); C ₁ , C ₂ , D ₁ and D ₂ represent the number of pixels; W ₁ and H ₁ are known quantities, and the scale of pixels is fixed, so the calculation methods of foot width and foot length can be obtained:

The length of the foot is:

H _foot = H ₁ -H ₂ (9)

The foot measurement process algorithm of the machine vision-based non-contact foot measurement method of the present invention can use the Modbus communication protocol to communicate with external application software (host computer), and the specific implementation flowchart is shown in FIG. 11 .

As shown in Figure 11, the implementation process of visual foot measurement in the non-contact foot measurement method based on machine vision of the present invention is divided into two parts:

(1) Camera calibration part: First, after the foot measurement program receives the calibration command sent by Modbus, it calls the calibration algorithm inside the foot measurement program, collects pictures with the calibration board, and determines whether the calibration is successful according to the detected corner positions. , and return to the host computer whether the calibration is successful or not;

(2) Foot measurement part: After the foot measurement program receives the foot measurement instruction sent by Modbus, it calls the foot measurement algorithm inside the foot measurement program to control the camera to collect images every 5 frames. A total of 5 frames of images are collected for measurement, and in the algorithm The results that do not meet the requirements are deleted internally, and the average value of 5 correct measurement values is calculated and returned to the upper computer.

The non-contact foot measurement method based on machine vision of the present invention can perform normal measurement under different lighting conditions, and examples under the following three lighting conditions are provided:

Example 1: Under the condition of low brightness, the person to be measured stands on both feet (as shown in Figure 12), and the obtained foot block diagram is shown in Figure 13. The measurement result is that the left foot is 229mm long, the left foot is 91mm wide, the right foot is 230mm long, and the right foot is 230mm long. The foot is 94mm wide.

Example 2: Under the condition of medium brightness, the person to be measured stands on both feet (as shown in Figure 14), and the obtained foot block diagram is shown in Figure 15. The measurement result is that the left foot is 228mm long, the left foot is 91mm wide, the right foot is 231mm long, and the right foot is 231mm long. The foot is 94mm wide.

Example 3: Under the condition of high brightness, the person to be measured stands on both feet (as shown in Figure 16), and the obtained foot block diagram is shown in Figure 17. The measurement result is that the left foot is 228mm long, the left foot is 91mm wide, the right foot is 231mm long, and the right The foot is 94mm wide.

Based on machine vision and combined with image processing technology, the invention realizes non-contact measurement of foot length and foot width by extracting the complete foot contour of the foot, which not only greatly reduces the measurement cost, shortens the measurement time, improves the measurement accuracy, and has the advantages of Strong environmental adaptability, can work normally in different lighting environments.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the scope of rights of the present invention; at the same time, the above descriptions should be understood and implemented by those skilled in the relevant technical fields. Any modifications, equivalent replacements and improvements made within the scope of the present invention shall be included within the protection scope of the present invention.

Claims (9)

1. a non-contact foot measurement method based on machine vision, is characterized in that: comprise camera calibration and visual measurement foot, camera calibration is used to determine measurement area and scale information and calculate the parameter of transmission transformation; visual measurement foot is calibrated in camera. On the basis of the obtained parameters, the measurements of the feet of the subject are measured. 2. The non-contact foot measuring method based on machine vision as claimed in claim 1, characterized in that: the measured value of the measured person's foot comprises foot length and foot width; the visual foot measurement needs to ensure that the measured foot is measured The subjects stand with their feet parallel, and the subjects standing with their feet in a splayed figure or standing with an outer figure of more than 10° will be identified as abnormal; the visual foot measurement does not require the use of a calibration board, and the subject stands with both feet on the pedals. board. 3. the non-contact foot measuring method based on machine vision as claimed in claim 1 is characterized in that, the main execution flow of described camera calibration is: (1.1) Place a calibration board marked with a specific color on the pedal; (1.2) Use a camera to shoot an image with a calibration plate, then convert the image from RGB color space to HSV color space, and extract the green area of the calibration plate as a measurement area through the color space threshold segmentation method; (1.3) Use the contour detection algorithm to find the outline of the green calibration plate, and extract the four corner points of the contour and the actual size of the calibration plate accordingly, and calculate the image perspective transformation matrix parameters and scale information; (1.4) Save the transmission transformation matrix parameters and scale information of the calibration plate image. 4. The non-contact foot measurement method based on machine vision as claimed in claim 3, wherein the contour detection algorithm in the step (1.3) adopts the FindContours() function in OpenCV. 5 . The non-contact foot measurement method based on machine vision according to claim 3 , wherein in the step (1.1), the color of the foot pedal is red, and the calibration plate is green. 6 . 6. The non-contact foot measurement method according to claim 1 or 3, wherein the camera calibration only needs to be performed once during the camera initialization process before the foot measurement, unless the camera is offset or the foot area is changed , otherwise there is no need to calibrate again. 7. the non-contact foot measuring method based on machine vision as claimed in claim 1 is characterized in that: the concrete processing step of described visual foot measuring is: (2.1) Use a camera to take pictures with the feet to be tested, and the pictures taken are oblique views; (2.2) Perform a transmission transformation on the picture taken in the above step (2.1) according to the parameters obtained by the camera calibration, and convert it into a top view to ensure that the pixel scale of the picture is uniform in the top view, and then according to the length and width of the foot area and The ratio of pixel points to calculate the value of foot length and foot width; (2.3) Use the Canny operator to extract the edge information of the foot from the transmission transformed picture, and use the Gaussian blur algorithm to remove the noise next to the edge, and convert the picture into the contour map of the foot. When extracting the edge information, the Canny operator is maximized The threshold is set to 0, that is, all edge information is retained; (2.4) After extracting the outline of the foot, then extract the minimum circumscribed rectangle of the outline, extract the width of the left and right foot regions and the distance from the toes of the two feet to the upper boundary of the area, and judge various measurements through the information of the minimum circumscribed rectangle abnormal situation; (2.5) After the processing of the above steps (2.1)-(2.4), the detection and distance calculation model can be obtained, taking the right foot as an example, where W ₁ represents the width of the standing foot area, W ₂ is the width of the right foot to be measured, H ₁ is the length of the standing area, H ₂ is the distance from the toe of the right foot to the upper boundary of the area; C ₁ , C ₂ , D ₁ and D ₂ represent the number of pixels; since W ₁ and H ₁ are known quantities, and the pixels The scale is fixed, and the calculation method of foot width and foot length can be obtained:

The length of the foot is: H _foot = H ₁ -H ₂ (3). 8. The non-contact foot measuring method based on machine vision as claimed in claim 7, characterized in that, in the step (2.4), judging various abnormal conditions of measurement specifically comprises: First of all, according to the number of detected minimum circumscribed rectangles, two anomalies such as no feet and obstacles can be detected; if the minimum circumscribed rectangle is not detected at this time, it can be judged that the camera has not captured an image with feet; if If the detected minimum circumscribed rectangle is greater than 2, it can be judged that the collected image has other obstacles similar to feet; Secondly, when the number of the minimum circumscribed rectangles is correct, the standing situation of the tested person can be judged according to the angle of the minimum circumscribed rectangle. If the angle of the rectangle is within the range of -80°～-45°, it is judged that the subject's feet are standing in the outer eight position. 9 . The non-contact foot measurement method based on machine vision according to claim 7 , wherein, in the step (2.4), the MinAreaRect() method of OpenCV is used to extract the minimum circumscribed rectangle of the outline. 10 .

Images