CN111192235A - An Image Measurement Method Based on Monocular Vision Model and Perspective Transformation - Google Patents

Abstract

The invention provides an image measuring method based on a monocular vision model and perspective transformation, which comprises the steps of firstly, obtaining any image; then, perspective transformation is carried out on the image, and adjustment is carried out, so that the aspect ratio of a certain object in a new image obtained after transformation is the same as the actual aspect ratio of the object; establishing a monocular vision model; on the image, two auxiliary straight lines RO and ST are constructed, perspective transformation is carried out, a double parallel line projection model is established, and the conversion parameter f of the point between the coordinate systems of the monocular vision model is solved and obtained_x(ii) a And then the distance between any object D in the image and the photographer X is obtained. The invention has the following effects: simple and clear image taking device for measuring distance between photographerThe distance of a target point, and measurement accuracy is higher, has practicality and suitability.

Description

An Image Measurement Method Based on Monocular Vision Model and Perspective Transformation

technical field

The invention relates to the field of image processing, in particular to a monocular vision model and an image measurement method of perspective transformation.

Background technique

About 80% of the information obtained by ordinary people comes from vision. How to extract target information is a key problem in vision research. As a non-contact measurement, the machine vision system will not have any contact between the observer and the observed, thus improving the reliability of the system, and the advantages of being able to work stably for a long time have been widely used, especially the binocular. Vision Systems and Polyocular Vision Systems. Due to the particularity of the system, they can calibrate and analyze the system parameters through the high-precision calibration plate, and can obtain a lot of useful information. The main carriers of visual information are images and videos, and in general, photos or videos are only captured by ordinary mobile phones or cameras, not by binocular or multi-eye vision systems, and there is no calibration process. Even the parameters of the camera are not known in the photos. Most of the existing technologies are calibrated by complete camera parameters or by a high-precision calibration board, and then the precise internal and external parameters of the camera are obtained, and the calculation of the monocular vision model is performed on this basis. For example, the application number is: CN108317958A, The name is: a patent application for an image measurement method and a measuring instrument is to use the calibration plate to solve the parameters, and then solve the measurement results. However, the solution process in this way is cumbersome, and some old photos do not have any information except the pictures that are more observable by the human eye. At this time, any method based on calibration cannot be used.

SUMMARY OF THE INVENTION

In order to solve the above problems, the present invention provides an image measurement method based on a monocular vision model and perspective transformation. Parallel projection transformation is an improved model based on the monocular vision model and image perspective transformation. Starting from the calculated conclusion, by constructing a suitable auxiliary line, transforming the auxiliary line through perspective transformation, selecting a group of targets with length information in the vertical direction for simple calibration, calculating the distance corresponding to any pixel point on this plane, and then calculating the shooting according to the plane projection. distance from the target point on the image. The image measurement method based on the monocular vision model and perspective transformation mainly includes the following steps:

S1: First acquire any captured image; then perform perspective transformation on the image, and adjust an object in the perspective transformed image, the actual aspect ratio of the object is known, so that the perspective transformation and corresponding The aspect ratio of the certain object in the new image obtained after adjustment is the same as the actual aspect ratio of the certain object; at the same time, according to any known actual distance in the new image, the actual value corresponding to the unit pixel in the new image is obtained. distance D, in order to calculate the distance between any two points in the new image;

S2: Taking the optical center of the camera as the center, establish the pixel coordinate system, the image coordinate system and the camera coordinate system. The conversion relationship between these three coordinate systems is the established monocular vision model;

S3: On the image, construct two auxiliary straight lines RO and ST, both of which are parallel to the bottom edge of the image, wherein the auxiliary straight line RO passes through the center point O of the image, and the auxiliary straight line ST and the auxiliary straight line ST are parallel to the bottom edge of the image. The distance of the straight line RO is dis; perform perspective transformation on the image after adding the auxiliary line, and construct a double-parallel line projection model based on the two auxiliary lines RO and ST, and obtain the number of pixels of the auxiliary line RO and the line SP are n ₁ and 1 respectively. n ₂ ; P is the intersection of the straight line from the photographer F to the center point O of the image and the straight line ST, then in combination with the distance corresponding to the unit pixel obtained in step S1, the actual distance s ₁ of the straight line RO and the straight line SP are obtained. Actual distance s ₂ ; according to the distance difference between the photographer X to the two straight lines RO and SP equal to the distance dis, solve the conversion parameter f _x of the point between the coordinate systems of the monocular vision model;

计算得到光心到直线RO的深度

光心到直线RO的深度即是拍摄者X到直线RO的距离XO；其中，z为相机光心到任意两点连线的深度，s为所述任意两点之间的实际距离，d是像元尺寸，n为所述任意两点的连线上的像素个数；由于直线XO垂直于直线RO，结合直线RO的实际距离和XO的实际距离，根据勾股定理，计算得到直线XO与直线OS的夹角θ，进而可得到所述图像中任一目标点D与拍摄者X之间的距离。S4: According to the obtained conversion parameter f _x , the formula

Calculate the depth from the optical center to the straight line RO

The depth from the optical center to the straight line RO is the distance XO from the photographer X to the straight line RO; among them, z is the depth from the optical center of the camera to the line connecting any two points, s is the actual distance between the two arbitrary points, and d is the Pixel size, n is the number of pixels on the connecting line of any two points; since the straight line XO is perpendicular to the straight line RO, combined with the actual distance of the straight line RO and the actual distance of XO, according to the Pythagorean theorem, the straight line XO and the The included angle θ of the straight line OS, and then the distance between any target point D in the image and the photographer X can be obtained.

Further, the image is captured by a camera with length information in horizontal and vertical directions, but other camera parameters are unknown.

Further, the new image is a top view of the scene in the image.

Further, a trial and error method is used to adjust an object in the image after perspective transformation.

Further, the pixel coordinate system takes the upper left corner vertex of the image as the origin, the U axis and the V axis as the horizontal and vertical axes, and (u, v) to represent the coordinates, and the unit is pixel; the image coordinate system is based on the center of the image. The origin is the x-axis and the y-axis as the horizontal and vertical axes. The x-axis and the y-axis are parallel to the U and V axes respectively. The coordinates are represented by (x, y), and the unit is mm; the camera coordinate system is based on the optical center of the camera. As the origin, the X and Y axes are the horizontal and vertical axes, the X and Y axes are parallel to the x and y axes of the image coordinate system, the optical axis of the camera is the Z axis, and the image coordinate point P(x, y) corresponds to The coordinate point P(X, Y, Z) of the camera coordinate system;

The relationship between the pixel coordinate system and the image coordinate system is shown in formula (1):

Among them, dx and dy represent the size of each pixel in the x-axis and y-axis, respectively, and (u ₀ , v ₀ ) represent the coordinates corresponding to the origin of the image coordinate system in the pixel coordinate system.

由相似原理可知：

可得：Further, a straight line EQ is selected in the image, and the coordinates of points E and Q in the image coordinate system are E(u ₁ , v ₁ ) and Q (u ₂ , v ₂ ) respectively, and these two points are on the camera. The corresponding coordinates in the coordinate system are (X ₁ , Y ₁ , Z ₁ ) and (X ₂ , Y ₂ , Z ₂ ); from the number of pixels n between the two EQ points, combined with the actual length D of the unit pixel, we get The actual length s of the two EQ points, the depth from the camera optical center O' to the straight line EQ is z, and the camera focal length is f, then D=d n; where d is the pixel size, n is the number of pixels on the straight line EQ,

It follows from the similar principle that:

Available:

in,

The beneficial effects brought by the technical solution provided by the present invention are: in the case of unknown camera parameters, the distance between the photographer and any target point in the photographed image can be measured simply and clearly, and the measurement accuracy is high, which has practicality and practicability. applicability.

Description of drawings

The present invention will be further described below in conjunction with the accompanying drawings and embodiments, in which:

1 is a flowchart of an image measurement method based on a monocular vision model and perspective transformation in an embodiment of the present invention;

2 is a schematic diagram of an image in an embodiment of the present invention;

Fig. 3 is the schematic diagram of the perspective transformation principle in the embodiment of the present invention;

4 is a schematic diagram of four-point perspective transformation in an embodiment of the present invention;

5 is an image comparison diagram before and after perspective transformation in the embodiment of the present invention;

6 is a schematic diagram of a camera monocular vision model in an embodiment of the present invention;

7 is a schematic diagram of the conversion between an image coordinate system and a pixel coordinate system in an embodiment of the present invention;

8 is a schematic diagram of obtaining the linear distance from the optical center to the target in the embodiment of the present invention;

Fig. 9 is the image perspective transformation comparison schematic diagram after adding auxiliary line in the embodiment of the present invention;

FIG. 10 is a schematic diagram of a parallel line projection model in an embodiment of the present invention.

Detailed ways

In order to have a clearer understanding of the technical features, objects and effects of the present invention, the specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

Embodiments of the present invention provide an image measurement method based on a monocular vision model and perspective transformation.

Please refer to FIGS. 1 to 10. FIG. 1 is a flowchart of an image measurement method based on a monocular vision model and perspective transformation in an embodiment of the present invention, FIG. 2 is a schematic diagram of an image in an embodiment of the present invention, and FIG. 3 is the present invention. A schematic diagram of the perspective transformation principle in the embodiment, FIG. 4 is a schematic diagram of a four-point perspective transformation in an embodiment of the present invention, FIG. 5 is an image comparison diagram before and after perspective transformation in an embodiment of the present invention, and FIG. 6 is a camera monocular in an embodiment of the present invention. A schematic diagram of the visual model, FIG. 7 is a schematic diagram of the conversion of the image coordinate system and the pixel coordinate system in the embodiment of the present invention, FIG. 8 is a schematic diagram of the linear distance between the optical center and the target in the embodiment of the present invention, and FIG. 9 is an embodiment of the present invention. Figure 10 is a schematic diagram of a parallel line projection model in an embodiment of the present invention, which specifically includes the following steps:

S1: First acquire any captured image; then perform perspective transformation on the image, and perform trial-and-error adjustment on an object in the perspective-transformed image, so that a certain object in the new image obtained after the perspective transformation and corresponding adjustment is adjusted. The aspect ratio of an object is the same as the actual aspect ratio of the object; at the same time, according to any known actual distance in the new image, the actual distance D corresponding to the unit pixel in the new image is obtained, so as to calculate any two in the new image. The distance between points; the image is taken by a camera with length information in both horizontal and vertical directions, but other camera parameters are unknown; the new image is the top view of the scene in the image;

S2: Taking the optical center of the camera as the center, establish the pixel coordinate system, the image coordinate system and the camera coordinate system. The conversion relationship between these three coordinate systems is the established monocular vision model. A fixed-length information is used to solve the conversion parameter f _x of the point between the coordinate systems of the monocular vision model;

S3: On the image, construct two auxiliary lines RO and ST, both of which are parallel to the bottom edge of the image, wherein the auxiliary straight line RO passes through the center point O of the image, and the auxiliary straight line ST and the auxiliary line The distance of the straight line RO is dis; perform perspective transformation on the image after adding the auxiliary line, and construct a double-parallel line projection model based on the two auxiliary lines RO and ST, and obtain the number of pixels of the auxiliary line RO and the line SP are n ₁ and 1 respectively. n ₂ ; P is the intersection of the straight line from the photographer F to the center O of the image and the straight line ST, then in combination with the distance corresponding to the unit pixel obtained in step S1, the actual distance s ₁ of the straight line RO and the actual distance of the straight line SP are obtained. s ₂ ; according to the distance difference between the photographer X to the two straight lines RO and SP equal to the distance dis, solve the conversion parameter f _x of the point between the coordinate systems of the monocular vision model;

计算得到光心到直线RO的深度

光心到直线RO的深度即是拍摄者X到直线RO的距离XO；其中，z为相机光心到任意两点连线的深度，s为所述任意两点之间的实际距离，d是像元尺寸，n为所述任意两点的连线上的像素个数；由于直线XO垂直于直线RO，结合直线RO的实际距离和XO的实际距离，根据勾股定理，计算得到直线XO与直线OS的夹角θ，进而可得到所述图像中某一物体D与拍摄者X之间的距离。S4: According to the solved conversion parameter f _x , according to the formula

Calculate the depth from the optical center to the straight line RO

The depth from the optical center to the straight line RO is the distance XO from the photographer X to the straight line RO; among them, z is the depth from the optical center of the camera to the line connecting any two points, s is the actual distance between the two arbitrary points, and d is the Pixel size, n is the number of pixels on the connecting line of any two points; since the straight line XO is perpendicular to the straight line RO, combined with the actual distance of the straight line RO and the actual distance of XO, according to the Pythagorean theorem, the straight line XO and the The included angle θ of the straight line OS, and then the distance between an object D in the image and the photographer X can be obtained.

The selected image is shown in FIG. 2 , and the specific operation of the method is described by taking the solution of the distance between the photographer X and the left side of the road as an example.

Step 1: Perform perspective transformation on the image to obtain a top view of the ground, and perform corresponding calibration

The essence of perspective transformation is to project the image to a new viewing plane, which is reflected in the "straightness" of the image, that is, the straight line in the original image is still a straight line after perspective transformation, as shown in Figure 3. In this example, Mainly to transform the ground, the image is in a two-dimensional plane, and the transformed image is shown in Figure 4. After the transformation, the actual distance information of objects in the horizontal and vertical directions can be transformed into the picture to correctly show the distance between the objects. than the actual same image, which is also one of the purposes of perspective transformation.

移动到目标点为

透视变换的变换公式如下：

其中

为透视矩阵。设定透视矩阵中的某一个数值，通过试凑法调整得到透视矩阵中剩余的数值，即得到整个透视矩阵，试凑法的效果是要变换后的新图像中的某个物体的纵横比与该物体的实际纵横比相同。由于是二维空间变换到三维空间的转换，因为图像在二维平面，故除以Z得到(X',Y',Z')，这个坐标表示图像上的点，即The objects whose length and width are determined in the perspective transformed image are calibrated, and the distance corresponding to each pixel of the plane in the horizontal and vertical directions is calculated, and the distance between any two points on the plane can be calculated. The principle of perspective transformation is as follows, the original target point is

Move to the target point as

The transformation formula of perspective transformation is as follows:

in

is the perspective matrix. Set a certain value in the perspective matrix, and adjust the remaining values in the perspective matrix by trial and error, that is, to obtain the entire perspective matrix. The effect of trial and error is that the aspect ratio of an object in the new image to be transformed is the same as The actual aspect ratio of the object is the same. Since it is a transformation from two-dimensional space to three-dimensional space, because the image is in a two-dimensional plane, it is divided by Z to get (X', Y', Z'), and this coordinate represents a point on the image, that is

Let a ₃₃ =1, expand the formula, we can get:

The specific perspective transformation is shown in Figure 5. First, two auxiliary lines HI and JK, which are perpendicular to the double lines with a wide field of view in the image and appear at intervals in the middle of the road, are made, as shown in (a) of Figure 5. Use the trial and error method to correct the image until the aspect ratio of a reference object in the obtained new image is consistent with the actual aspect ratio of the reference object. For example, the length and width of the actual double yellow dotted line in the road in the image can be selected. In the actual road, the road signs on the road have corresponding standards. Here, its information is used as a reference for image calibration, and finally a corrected image as shown in (b) in Figure 5 is obtained.

After the corrected image is obtained, any known length information in the figure can be used to calculate the distance between any two points or any two straight lines in the plane. The information calculates the ratio information of the unit pixel and the actual size, that is, the actual distance D corresponding to the unit pixel, and uses this as a benchmark for subsequent calculations.

Step 2: Establish a monocular vision system model and analyze the depth of the optical center to the point and the line

To obtain the relevant distance based on ordinary images, first establish the monocular vision system model as shown in Figure 6, establish a real-world Cartesian three-dimensional coordinate system with the optical center O' as the center, and then select the fixed-length information pairs in the figure. The parameters of the monocular vision model are inversely solved to obtain the data in the model. To calculate the three-dimensional coordinates of the target relative to the camera, three coordinate systems are involved: the transformation of the pixel coordinate system, the image coordinate system (ICS), and the camera coordinate system (CCS). Among them, the pixel coordinate system takes the upper left corner of the image as the origin, the U axis and the V axis as the horizontal and vertical axes, and (u, v) represents the coordinates, and the unit is pixel; the image coordinate system takes the center of the image as the center The origin, with the x-axis and y-axis as the horizontal and vertical axes, the x-axis and the y-axis are parallel to the U and V axes respectively, and the coordinates are expressed by (x, y) in mm; the camera coordinate system is based on the optical center of the camera. The origin, with the X-axis and Y-axis as the horizontal and vertical axes, the X-axis and Y-axis are parallel to the x-axis and y-axis of the image coordinate system, the optical axis of the camera is the Z-axis, and the image coordinate point P(x, y) corresponds to the camera The coordinate point P (X, Y, Z) of the coordinate system; the point (x, y) in the image coordinate system corresponds to the point (x, y, z) in the actual space coordinate.

1) Calculate the depth of the actual point

According to the monocular vision model shown in Figure 6, the relationship between the image coordinate system and the pixel coordinate system as shown in Figure 7 is established. It can be seen from Figure 7 that in ICS, the relationship between the pixel coordinate system and the image coordinate system is The relationship is as follows, where dx and dy represent the dimensions of each pixel in the x- and y-axes, respectively, and f is the focal length of the camera:

The point P(X,Y,Z) in the camera coordinate system (CCS) corresponds to the point p(x,y) in the image coordinates. According to the similarity relationship, the following mathematical relationship can be obtained:

2) Calculate the distance from the optical center to a line in the image

Solving the distance from the optical center to a straight line in the image is to solve the depth from the optical center to the target line, as shown in Figure 8, where O' is the optical center of the camera, that is, the optical center of the monocular vision system, select a line in the image Straight line, the two end points are E(u ₁ , v ₁ ), Q(u ₂ , v ₂ ), and the coordinates of their two points in reality are (x ₁ , y ₁ , z ₁ ) and (x ₂ , y ₂ ,z ₂ ). When the optical center is far away from the straight line, the depth (ie, the distance) from the photographer to any point on the line segment EQ is considered to be the same, that is, the depth from O' to any point on the EQ line segment is considered unchanged.

由相似原理可知：

可得：In Figure 8, the actual length of the two EQ points is s, the actual length of the unit pixel is D, the depth from the camera optical center to the straight line is z, and the focal length is f, then D=d·n. Where d is the pixel size, and n is the number of pixels on the line, you can get:

It follows from the similar principle that:

Available:

where f _x is a simplified quantity,

Step 3: Build auxiliary lines and perform perspective transformation to obtain a parallel projection model, and calculate the distance to be obtained

In step 1, the corresponding perspective transformation has been done, and the distance corresponding to a single pixel point after the transformation has been obtained. In step 3, two auxiliary lines are constructed in the image, and then perspective transformation is performed to analyze the relevant model. Guided by the example of solving the distance between the photographer X and the left side of the road, the following steps are performed:

1) In the image shown in (a) of Figure 9, construct two auxiliary straight lines parallel to the bottom edge of the image, and one of them is required to pass through the center of the image, because such an angle starts from the optical center And perpendicular to the ICS. For the straight line RO and the straight line ST shown in (a) of FIG. 9 , the image obtained after perspective transformation is shown in (b) of FIG. 9 .

2) Build a double parallel line projection model. After the auxiliary line undergoes perspective transformation, a schematic top view of the auxiliary line is shown in FIG. 10 , and the number of pixels n ₁ and n ₂ of the line segments SP and SP can be directly obtained from the image. Since RO is a straight line passing through the center of the image and parallel to the bottom edge, the straight line XO from the photographer to the center point O is perpendicular to it, and SP is also perpendicular to XO. Since the actual distance represented by the unit pixel in the new image is known, the actual distances of the line segments RO and SP can be calculated as s ₁ and s ₂ , respectively.

和

直线RO和SP之间的实际距离dis已知，由于拍摄者X到两条直线的深度差即为两直线RO和SP之间的距离，所以

由此可以计算出

3) According to formula (4), the depths from the photographer X to the two straight lines RO and SP are respectively:

and

The actual distance dis between the straight lines RO and SP is known. Since the depth difference between the photographer X and the two straight lines is the distance between the two straight lines RO and SP, so

From this it can be calculated

也是拍摄者X距离直线RO的距离XO，最后根据图10中的几何信息，根据勾股定理，计算得到夹角θ，进而得到拍照者X到左侧马路的距离DX为：DX＝CD+CX＝CD+OX*sin(θ)；其中，CD为左侧马路与马路中间间隔出现的双线的距离，该距离可由马路设计时的信息得到，即该距离在马路设计时即已知，再结合计算出来的XO和θ，即可得到拍照者X到左侧马路的距离DX。4) Calculate the depth from the optical center to the straight line RO from the value of f _x obtained in step 3)

It is also the distance XO from the photographer X to the straight line RO. Finally, according to the geometric information in Figure 10 and the Pythagorean theorem, the included angle θ is calculated, and then the distance DX from the photographer X to the left road is: DX=CD+CX =CD+OX*sin(θ); where CD is the distance between the left road and the double line in the middle of the road, the distance can be obtained from the information during road design, that is, the distance is known when the road is designed, and then Combining the calculated XO and θ, the distance DX from the photographer X to the left road can be obtained.

Through the image measurement method based on the monocular vision model and perspective transformation described in the present invention, the distance between the photographer and any target point in the captured image can be simply and quickly obtained with high accuracy.

The beneficial effects of the present invention are: in the case of unknown camera parameters, the distance between the photographer and any target point in the captured image can be measured simply and clearly, and the measurement accuracy is high, and it has practicability and applicability.

The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection of the present invention. within the range.

Claims (6)

1. an image measurement method based on monocular vision model and perspective transformation, is characterized in that: comprise the following steps: S1: First acquire any captured image; then perform perspective transformation on the image, and adjust an object in the perspective transformed image, the actual aspect ratio of the object is known, so that the perspective transformation and corresponding The aspect ratio of the object in the new image obtained after adjustment is the same as the actual aspect ratio of the object; at the same time, according to any known actual distance in the new image, the actual distance D corresponding to the unit pixel in the new image is obtained , in order to calculate the distance between any two points in the new image; S2: Taking the optical center of the camera as the center, establish the pixel coordinate system, the image coordinate system and the camera coordinate system. The conversion relationship between these three coordinate systems is the established monocular vision model; S3: On the image, construct two auxiliary straight lines RO and ST, both of which are parallel to the bottom edge of the image, wherein the auxiliary straight line RO passes through the center point O of the image, and the auxiliary straight line ST and the auxiliary straight line ST are parallel to the bottom edge of the image. The distance of the straight line RO is dis; perform perspective transformation on the image after adding the auxiliary line, and construct a double-parallel line projection model based on the two auxiliary lines RO and ST, and obtain the number of pixels of the auxiliary line RO and the line SP are n ₁ and 1 respectively. n ₂ ; P is the intersection of the straight line from the photographer F to the center point O of the image and the straight line ST, then in combination with the distance corresponding to the unit pixel obtained in step S1, the actual distance s ₁ of the straight line RO and the straight line SP are obtained. Actual distance s ₂ ; according to the distance difference between the photographer X to the two straight lines RO and SP equal to the distance dis, solve the conversion parameter f _x of the point between the coordinate systems of the monocular vision model; S4: According to the obtained conversion parameter f _x , the formula

Calculate the depth from the optical center to the straight line RO

The depth from the optical center to the straight line RO is the distance XO from the photographer X to the straight line RO; among them, z is the depth from the optical center of the camera to the line connecting any two points, s is the actual distance between the two points, and d is the Pixel size, n is the number of pixels on the connection line between any two points; since the straight line XO is perpendicular to the straight line RO, combined with the actual distance of the straight line RO and the actual distance of XO, according to the Pythagorean theorem, the straight line XO and the The included angle θ of the straight line OS, and then the distance between any target point D in the image and the photographer X can be obtained. 2. An image measurement method based on a monocular vision model and perspective transformation as claimed in claim 1, wherein the image is taken by a camera with length information in both horizontal and vertical directions, but other camera parameters are unknown made. 3 . The image measurement method based on a monocular vision model and perspective transformation according to claim 1 , wherein the new image is a top view of the scene in the image. 4 . 4. a kind of image measurement method based on monocular vision model and perspective transformation as claimed in claim 1 is characterized in that: in step S1, adopt trial and error method to a certain object in the described image after perspective transformation make adjustments. 5. a kind of image measurement method based on monocular vision model and perspective transformation as claimed in claim 1, is characterized in that: in step S2, pixel coordinate system takes the upper left corner vertex of described image as origin, takes U axis and the V axis are the horizontal and vertical axes, and the coordinates are represented by (u, v), and the unit is pixel; the image coordinate system takes the center of the image as the origin, and the x and y axes are the horizontal and vertical axes, and the x and y axes are respectively Parallel to the U and V axes, the coordinates are represented by (x, y), and the unit is mm; the camera coordinate system is based on the optical center of the camera as the origin, and the X and Y axes are the horizontal and vertical axes, and the X and Y axes are respectively Parallel to the x-axis and y-axis of the image coordinate system, the optical axis of the camera is the Z-axis, and the image coordinate point P(x,y) corresponds to the coordinate point P(X,Y,Z) of the camera coordinate system; The relationship between the pixel coordinate system and the image coordinate system is shown in formula (1):

Among them, dx and dy represent the size of each pixel in the x-axis and y-axis, respectively, and (u ₀ , v ₀ ) represent the coordinates corresponding to the origin of the image coordinate system in the pixel coordinate system. 6. a kind of image measuring method based on monocular vision model and perspective transformation as claimed in claim 1, is characterized in that: in step S3, in described image, choose a straight line EQ, the in image of E and Q point The coordinates in the coordinate system are E(u ₁ , v ₁ ) and Q(u ₂ , v ₂ ), and the corresponding coordinates of these two points in the camera coordinate system are (X ₁ , Y ₁ , Z ₁ ) and ( X ₂ , Y ₂ , Z ₂ ); the actual length s of the two EQ points is obtained from the number of pixels n between the two EQ points, combined with the actual length D of the unit pixel, and the depth from the camera optical center O' to the straight line EQ is z , the focal length of the camera is f, then D=d·n; where d is the pixel size, n is the number of pixels on the line EQ,

It follows from the similar principle that:

Available:

in,

Images