CN111442817A - Non-contact structured light binocular vision sewage level measuring device and method - Google Patents

Abstract

The invention discloses a non-contact structured light binocular vision sewage water level measuring device and method, comprising a rotating robot, the rotating robot includes a base, a horizontal rotating base is connected to the base, and a vertical rotating base is connected to the shaft. The pitch axis is rotated, the pitch axis is connected with a horizontal movement axis, and two vision devices in different positions and a structured light generator are fixed on the horizontal movement axis. The device of the invention adopts an integrated structure design, which is convenient for on-site installation, has high reliability, strong adaptability, and is easy to popularize. Adapting to different needs of sewage treatment environment, the whole set of device can measure by changing different angles, and realize self-verification of sewage water level measurement according to different measurement results to ensure the accuracy and reliability of measurement results.

Description

Non-contact structured light binocular vision sewage water level measurement device and method

technical field

The invention relates to a non-contact structured light binocular vision sewage water level measurement device and method, belonging to the technical field of social network analysis.

background content

With the rapid development of the economy and the expansion of the urban circle, the urban population has increased sharply, and the amount of urban sewage is increasing. Simply relying on automatic water purification can no longer meet the actual needs. The sewage treatment level of the sewage treatment plant is directly related to the living environment of the entire urban environment. condition. In the process of sewage treatment, sewage water level measurement is an important part of sewage treatment, which is directly related to the instantaneous flow of sewage and the amount of sewage that needs to be matched. At the same time, with the acceleration of the urban digitalization process, the real-time information of sewage water level needs to be updated in time, and it is imperative to study an economical and feasible digital sewage water level measurement device.

At present, the commonly used water level monitoring mainly consists of contact monitoring and non-contact monitoring. Contact monitoring mainly adopts several methods such as water gauge, float type and pressure type, and non-contact measurement mainly adopts ultrasonic water level gauge and radar water level gauge. Due to the factors such as strong corrosion, high adhesion, strong impact and volatile gas in urban sewage, and the float type level gauge cannot measure viscous media, the high adhesion environment of sewage can easily cause the floating ball or flap (column) to get stuck, sewage The impact of the equipment will also damage the mechanical structure of the equipment, resulting in errors in liquid level measurement. The pressure type liquid level gauge needs to be installed at the bottom. Due to the adhesion of sewage and changes in viscosity, the measurement is inaccurate, requiring frequent maintenance, and it is inconvenient to install and debug. . For non-contact measurement, ultrasonic measurement is greatly affected by the environment, and the cost is high, which is not conducive to large-scale delivery. The method of image recognition water gauge can realize liquid level measurement, but the sewage treatment environment is complex, volatile gas is large, and light is not enough. Therefore, the measurement accuracy and accuracy cannot be guaranteed.

SUMMARY OF THE INVENTION

The invention overcomes the shortcomings of the prior art, and discloses a non-contact structured light binocular vision sewage water level measuring device and method. The device of the invention adopts an integrated structure design, which is convenient for on-site installation, has high reliability, strong adaptability, and is easy to popularize. Adapting to different needs of sewage treatment environment, the whole set of device can measure by changing different angles, and realize self-verification of sewage water level measurement according to different measurement results to ensure the accuracy and reliability of measurement results.

In order to solve the above-mentioned technical problems, the technical scheme adopted in the present invention is:

A non-contact structured light binocular vision sewage water level measurement device, comprising a rotating robot, the rotating robot includes a base 1, the base 1 is connected with a horizontally rotating rotating base 2, and the rotating base 2 is connected with a vertically rotating shaft. The tilt axis 3 is connected with a horizontal movement axis 4 , and two vision devices 5 and a structured light generator 6 are fixed on the horizontal movement axis 4 in different positions.

As a further improvement, the rotating base 2 is driven to rotate by the first motor 9 ; the pitch axis 3 is driven to rotate by the second motor 10 .

A further improvement, the horizontal moving shaft 4 includes a sliding cylinder 41, and the sliding cylinder 41 is slidably connected with a terminal connecting rod 42, one end of the terminal connecting rod 42 is threadedly connected with a screw rod 43, and the other end is connected with a mounting seat 7, and the screw rod 43 is connected with a A limit switch 45 is installed on the third motor 44 and the sliding cylinder 41 .

As a further improvement, the mounting base 7 is formed with a long-hole guide rail 8 for mounting the vision device 5 and the structured light generator 6 .

As a further improvement, the first motor 9 , the second motor 10 , the third motor 44 , the vision device 5 and the structured light generator 6 are wired or wirelessly connected to the equipment control cabinet 11 .

A further improvement, the visual device 5 is an industrial camera.

A non-contact structured light binocular visual sewage water level measurement method, comprising the following steps:

Step 1, align two visual devices 5 and a structured light generator 6 at different positions on the water surface, the structured light generator 6 emits characteristic stripe structured light 12 to the water surface, and the two visual devices 5 take pictures respectively and draw the water surface. The center point of the characteristic stripe structured light;

Step 2. The coordinate points of the center point of the characteristic stripe structured light on the images obtained by the two visual devices 5 are respectively A(x ₁ , y ₁ , z ₁ ) and B(x ₂ , y ₂ , z ₂ ), then,

θ represents the angle between the characteristic stripe structured light and the sewage liquid surface;

Knowing the actual coordinates of point A and point B in the two visual devices, the distance z _to the liquid surface corresponding to the center point H ( _{in x,} y _, z) _of the binocular vision baseline:

The vertical distance from the top of the rotating robot to the liquid surface is:

a represents the telescopic length of the horizontal moving axis 4;

According to the angle of the current pitch axis, the angle between the characteristic fringe structured light emitted by the current structured light generator 6 and the horizontal plane is obtained as θ ₁ , and according to the right triangle, it is obtained:

l ₁ represents the height of the device vertex from the sewage liquid surface;

When the horizontal movement axis of the entire rotating robot is kept horizontal, the height from the center to the bottom surface is l. Assuming that the height of the liquid level is l ₂ , the actual height of the liquid level is obtained as:

A further improvement is to change the positions of the visual device 5 and the structured light generator 6, and obtain l ₂ by continuous measurement for many times, and then take the average value of l ₂ as the actual height of the final liquid level.

For a further improvement, the central point of the characteristic stripe structured light is obtained in the following manner: filter the image captured by the vision device 5, and after median filtering, repeat the spatial filtering of the image for 2 to 3 times, remove the coarse extraction, and leave it behind. Then, the sub-pixel precision level extraction is performed on the area of the image occupied by the structured coded light, and each pixel is subdivided, and the subdivision is filled by the bilinear interpolation method. The back area, and then cyclically search to find the grayscale maximum pixel point of each row, according to the principle of least squares, for the grayscale maximum value pixel point and the two adjacent pixels on both sides, through parabolic fitting or The Gaussian curve is fitted with a quadratic curve, and the maximum value point is obtained as the center point of the encoded structured light, that is, the center point of the characteristic stripe structured light is obtained. location to find out.

In a further improvement, the characteristic stripe structured light is the only characteristic stripe structured light.

The advantages of the present invention are as follows:

1. Device innovation: The non-contact structured light stereo vision sewage water level measuring device designed by the present invention is a sewage water level measuring device composed of a binocular vision unit, a structural feature coding light, a horizontal moving axis, a pitch axis, a rotating base, etc. No related devices were found. The device adopts an integrated structure design, which is convenient for on-site installation, with high reliability, strong adaptability and easy promotion.

2. Method innovation:

(1) The structured light generator emits coded light with characteristic values, and adjusts the power of the structured light generator according to environmental changes to adapt to different environmental needs of sewage treatment.

(2) Use the rotating robot combined with binocular vision to establish a measurement mathematical model, extract the eigenvalues of the coded light according to image acquisition and image matching, and analyze the current angle of the rotating robot, and compare the distance of the software processing result with the hardware The actual movement distance should be combined to improve the detection accuracy of sewage water level.

(3) The whole set of device can measure by changing different angles, and realize self-verification of sewage water level measurement according to different measurement results, so as to ensure the accuracy and reliability of measurement results.

Description of drawings

Figure 1 is a block diagram of the system structure.

Figure 2 is a structural diagram of a non-contact structured light binocular vision sewage water level measuring device.

Figure 3 shows the internal layout of the equipment control cabinet.

Figure 4 is a schematic diagram of the measurement.

Detailed ways

Example 1

A non-contact structured light binocular vision sewage water level measurement device, which combines a rotating robot with a computer binocular vision system, and uses feature-encoded structured light to increase the target eigenvalue, adopts binocular vision, robot automatic control, image matching, intelligent transmission The combination of sensor technology improves the accuracy and reliability of sewage water level detection. At the same time, the non-contact measurement method is used to install the binocular vision and structured light generator on the end execution structure of the rotating robot arm, without on-site maintenance, reducing Labor costs.

The non-contact structured light binocular vision sewage water level measurement device mainly includes a data processing unit, a structured light generator, a binocular vision unit, and a rotating robot. The data processing unit includes a central processing unit, a memory, a user terminal and a communication module. The central processing unit is connected to the memory, the user terminal and the communication module, and is placed in the equipment control cabinet. The binocular vision unit includes two industrial cameras, two The industrial camera is installed on the guide rail through the slider and moves along the guide rail, and the guide rail is perpendicular to the lens direction of the digital camera; the structured light generator is installed on the top surface of the guide rail, and the structured light generator is oriented in the same direction as the digital camera lens; the structured light generator and The two digital cameras are connected with the communication module of the data processing unit; the rotating robot unit consists of a base, a rotating base, a pitch axis, a horizontal moving axis and related mechanical connection structures, the rotating robot is fixed on the base, and the end of the horizontal moving axis executes The structure is equipped with a structured light stereo vision unit, and the horizontal movement of the structured light stereo vision unit module is realized through the movement of the motor; the horizontal movement axis is installed on the pitch axis, and the pitch angle is changed by controlling the motors related to the pitch axis; the pitch axis It is installed on the rotating base, and the rotation angle is changed through the rotating base. The motion control mechanism of the rotating robot is connected to the central processing unit of the data processing unit, and the motion process of each axis is fed back to the central processing unit through the encoder to achieve The overall device is safe and controllable, and constitutes a non-contact structured light binocular visual sewage water level measurement device.

The system block diagram of the non-contact structured light binocular vision sewage water level measurement device is shown in Figure 1. The system controls the stepping motor of the rotating base and the pitch axis to move a certain angle through the central processing unit, and can see the position of the sewage liquid level, control the stepping motor of the horizontal moving axis, and drive the structured light binocular installed at the end of the mechanical arm. The vision unit moves to a clear imaging position, and records the current position information, including the angle of the rotating base, the angle of the pitch axis, and the telescopic distance of the horizontal movement axis. The central processor controls the structured light generator to generate coded information with specific stripes Irradiate to the junction of the sewage liquid surface, control the binocular vision system to perform image extraction, preprocessing, image feature analysis and feature matching, and three-dimensional information acquisition. Through analysis, combined with the current position information, the actual distance of the current sewage liquid level is obtained. The central processor transmits to the memory and the user terminal displays.

The characteristic stripe structured light is a stripe structured light with specific coding information generated by a structured light generator. The coding information is expressed by the width of the light and dark stripes. Different structured light generators have different characteristic stripe structured lights.

The non-contact structured light binocular vision sewage water level measurement device adopts an integrated structure design, which is easy to install and maintain. It is mainly used in the field of smart city sewage treatment, especially in the field of sewage treatment with harsh environment, difficult installation and maintenance, and high precision. The schematic diagram of the three-dimensional device structure is shown in Figure 2.

1. Structured light binocular vision system

Industrial cameras mainly use CMOS or CCD chips as the core components to convert the received optical signals into electrical signals, with high image stability, high transmission ability and high anti-interference ability. This device uses two industrial cameras to form parallel binoculars Stereo vision system, the axis is parallel and fixed on the guide rail, and the distance between the cameras can be adjusted by sliding the guide rail. This device adopts a binocular vision system with the camera axis parallel but not limited to the axis parallel. The feature-encoded structured light with different brightness can be emitted by adjusting the power to adapt to different working environments.

2. Feature-encoded structured light

The characteristic stripe structured light is the stripe structured light with encoded information. The structured light generator generates the striped structured light with specific encoded information. The encoded information is expressed by the width of the light and dark stripes. Different structured light generators have different The characteristic fringes of , this time, but not limited to the only characteristic fringe structured light, other matching characteristic fringe structured light can also be used.

3. Horizontal movement axis

The horizontal moving shaft is mainly composed of stepper motor, coupling, sliding module, end connecting mechanism, limit switch, motor base, encoder, etc. The motor is installed on the motor base, and one end of the sliding module is connected to the motor through the coupling , one end is connected to the end connection structure, the end execution structure is connected to the structured light binocular vision unit, and the linear motion of the structured light binocular vision unit is realized through the movement of the motor, and the limit switch is used to ensure the safety and reliability of the movement process. The encoder records the pulses in real time and sends them to the central processing unit for processing.

4. Pitch axis

The pitch axis mainly includes the motor, the motor base, the mechanical arm, the encoder and the horizontal moving shaft base. The mechanical arm is connected to the rotating base, the motor is installed on the motor base, and the horizontal moving shaft is connected to the XZ plane through the movement of the motor. The angle changes inside, so that the sewage treatment liquid surface is in the field of vision of the structured light binocular vision system.

5. Rotary base

The rotating base mainly includes a motor, a motor base, a mechanical arm and an encoder. By adjusting the rotating base, the angle change of the measuring device in the XY plane is realized, so that the sewage treatment liquid level is in a reasonable position of the structured light binocular vision system.

Equipment control cabinet

The equipment control cabinet adopts industrial grade and has good airtight performance. The power module, central processing unit, memory and user terminal are placed inside. Power to the processor and memory. The central processing unit mainly completes the control of the rotating robot and the distance measurement of the structured light binocular vision system. The user terminal mainly completes the display of sewage level data and the display of the current robot posture. The internal layout of the equipment box is shown in Figure 3. base

For stationary rotating robots.

The above are the basic functions and functions of various hardware connection units. The measurement principle and test method of the non-contact structured light stereo vision sewage water level measurement device are described below.

1. Non-contact structured light binocular vision sewage water level measurement principle

According to the principle of triangulation, the binocular vision system can be used to measure the three-dimensional information of the target. The environment of the sewage treatment plant is complex, the target point cannot be accurately identified, and the eigenvalues are few, and the binocular vision cannot be matched. The structured coded light is used to increase the eigenvalues. Before extracting the central stripe of the coded light, the image is filtered. After value filtering, repeat the spatial filtering of the image for 2 to 3 times, remove the noise points left after the rough extraction, and obtain the area of the image occupied by the structured coded light, and then perform sub-pixel precision level extraction on the area of the image occupied by the structured coded light. , subdivide each pixel in it, fill the subdivided area through bilinear interpolation, and then perform a circular search to find the grayscale maximum pixel point of each row, and according to the principle of least squares, for the grayscale maximum value Value pixel point and its two adjacent pixels on both sides, perform quadratic curve fitting through parabola fitting or Gaussian curve, and find the maximum value point as the center point of the encoded structured light, and then perform image matching. The position of a target point in the two cameras is found, as shown in Figure 4, the two characteristic points of the measured characteristic encoded light and the liquid surface are A(x ₁ , y ₁ , z ₁ ) and B(x ₂ , y ₂ , z ₂ ), where z ₁ is the distance of CD, z ₂ is the distance of AE, and ΔABC is a right-angled triangle, measured by the formula:

The horizontal axis of the current rotating robot moves a, according to the similar principle, if the actual coordinates of point A and point B in binocular vision are known, then the center point H ( _{in x,} y _, z) _of the baseline center point of binocular vision corresponds to Distance to liquid level:

According to the formula, the vertical distance from the top of the rotating robot to the liquid surface is:

According to the angle of the current pitch axis, the angle between the current structured light stereo vision unit and the horizontal plane is obtained as θ ₁ , and according to the right triangle, we get:

To sum up, when the horizontal movement axis of the entire rotating robot is kept in a horizontal state, the height from the center to the bottom surface is l. Assuming that the height of the liquid level is l ₂ , the actual height of the liquid level is obtained through the above formula:

2. Non-contact structured light stereo vision sewage water level test method

The central processor controls the rotating robot to move the rotating base. After moving to the designated position, the current rotation angle is recorded through the encoder, and the pitch axis is controlled to move at a certain angle, so that the sewage level is in the field of vision of binocular stereo vision. Control the structured light generator to produce characteristic stripes, and place the structured light stripes at the junction of the sewage and liquid surface, record the angle of the current pitch axis, and control the movement of the horizontal axis, so that the binocular vision system can clearly capture the characteristic structured light and obtain the current The target point and the width of the characteristic coded light, according to the pitch axis angle and the related calculation formula, get the current sewage liquid level height, change the relevant axis of the rotating robot, repeat the above measurement steps, measure the current liquid level height, continuously measure more The average value is obtained every time, and the mechanical movement is combined with the software measurement results to achieve accurate and reliable measurement results.

The above example is only a specific embodiment of the present invention, and simple transformation and replacement thereof are also within the protection scope of the present invention.

Claims (10)

1. a non-contact structured light binocular vision sewage water level measuring device, is characterized in that, comprises a rotating robot, and the rotating robot comprises a base (1), and the base (1) is pivotally connected with a horizontally rotating rotating base (2), A vertical rotating pitch axis (3) is connected to the rotating frame (2), the pitch axis (3) is connected to a horizontal moving axis (4), and two vision axes at different positions are fixed on the horizontal moving axis (4). device (5) and a structured light generator (6). 2. The non-contact structured light binocular vision sewage water level measuring device according to claim 1, wherein the rotating base (2) is driven to rotate by the first motor (9); the pitch axis (3) Rotation is driven by the second motor (10). 3. The non-contact structured light binocular vision sewage water level measuring device according to claim 2, wherein the horizontal moving shaft (4) comprises a sliding cylinder (41), and a terminal end is slidably connected in the sliding cylinder (41). A connecting rod (42), one end of the end connecting rod (42) is threadedly connected with a screw rod (43), the other end is connected with a mounting seat (7), the screw rod (43) is connected with a third motor (44), and the sliding cylinder (41) is Install limit switch (45). 4. The non-contact structured light binocular vision sewage water level measuring device according to claim 3, wherein the mounting seat (7) is formed with a visual device (5) and a structured light generator (6) for mounting the slotted guide rail (8). 5. The non-contact structured light binocular vision sewage water level measuring device according to claim 3, wherein the first motor (9), the second motor (10), the third motor (44), the visual device (5) and the structured light generator (6) are wired or wirelessly connected with an equipment control cabinet (11). 6. The non-contact structured light binocular vision sewage water level measuring device according to claim 1, wherein the vision device (5) is an industrial camera. 7. a non-contact structured light binocular vision sewage water level measurement method, is characterized in that, comprises the steps: Step 1. Align two visual devices (5) and a structured light generator (6) at different positions on the water surface, the structured light generator (6) emits characteristic stripe structured light (12) to the water surface, and the two visual devices (5) Take pictures respectively and obtain the center point of the characteristic stripe structured light on the water surface; Step 2: The coordinate points of the center point of the characteristic stripe structured light on the images obtained by the two visual devices (5) are A(x ₁ , y ₁ , z ₁ ) and B(x ₂ , y ₂ , z ₂ ), respectively, but,

θ represents the angle between the characteristic stripe structured light and the sewage liquid surface; Knowing the actual coordinates of point A and point B in the two visual devices, the distance z _to the liquid surface corresponding to the center point H ( _{in x,} y _, z) _of the binocular vision baseline:

The vertical distance from the top of the rotating robot to the liquid surface is:

a represents the telescopic length of the horizontal moving axis (4); According to the angle of the current pitch axis, the angle between the characteristic fringe structured light emitted by the current structured light generator (6) and the horizontal plane is θ ₁ , and according to the right triangle, it is obtained:

l ₁ represents the height of the device vertex from the sewage liquid surface; When the horizontal movement axis of the entire rotating robot is kept horizontal, the height from the center to the bottom surface is l. Assuming that the height of the liquid level is l ₂ , the actual height of the liquid level is obtained as:

8. non-contact structured light binocular vision sewage water level measurement method as claimed in claim 7, is characterized in that, the position of the vision device (5) and structured light generator (6) that changes, continuously measures and obtains many times _1.2 , and then take the average value of _l2 as the actual height of the final liquid level. 9. The non-contact structured light binocular vision sewage water level measurement method as claimed in claim 7, wherein the central point of the characteristic stripe structured light is obtained in the following manner: the image captured by the vision device (5) is filtered and processed , after median filtering, repeat the spatial filtering of the image for 2 to 3 times, remove the noise points left after rough extraction, and obtain the area of the image occupied by the structured coded light, and then perform sub-pixel analysis on the area of the image occupied by the structured coded light. Accuracy level extraction, subdivide each pixel in it, fill the subdivided area through bilinear interpolation, and then perform a cyclic search to find the grayscale maximum pixel point of each row, according to the principle of least squares for grayscale The maximum value pixel point and its two adjacent pixels on both sides are fitted with a quadratic curve by parabolic fitting or Gaussian curve, and the maximum value point is obtained as the center point of the encoded structured light to obtain the characteristic fringe structure. The center point of the light, and then image matching is performed to find out the position of the same target point in the two visual devices (5). 10 . The non-contact structured light binocular visual sewage water level measurement method according to claim 7 , wherein the characteristic stripe structured light is the only characteristic stripe structured light. 11 .

Images