Cylinder roll angle vision measurement algorithm based on full-surrounding target

Abstract

The invention discloses a cylinder roll angle vision measurement algorithm based on a full-surrounding target, which utilizes a camera to capture a target pattern and accurately analyzes the roll angle of a cylinder through an image processing technology, and comprises the following steps of: adopting a uniformly distributed target design method, and through an image feature enhancement technology and a proper threshold value, calculating the roll angle of the cylinder; determining the accurate positions of the positioning strip and the distance measuring strip; performing closed operation and de-noising processing on the positioning strip, and determining the position of a positioning point by using polygon fitting and a vector rotation method; positioning points of the positioning strips are analyzed, the two middle points are connected to form a central axis, distance measurement points are calculated in combination with the distance measurement strips and the central axis, and stable feature points are selected as distance measurement references; and accurately calculating the roll angle by using a deformation compensation algorithm and a similar triangle principle. The dynamic motion state of the cylinder can be captured and analyzed in real time, the measurement precision and real-time performance are remarkably improved, dependence on the external environment is reduced, and the equipment installation and maintenance process is simplified.

Description

Technical Field

[0001] The invention relates to a measurement technology of a cylinder roll angle, and in particular to a cylinder roll angle visual measurement algorithm based on a fully surrounded target. Background Art

[0002] At present, in modern industrial production and laboratory testing, the accurate measurement of the roll angle of a cylinder is crucial and widely used in automated production, quality control and experimental research. Traditional measurement methods mostly rely on mechanical sensors, photoelectric sensors or laser measurement equipment, but these methods have shortcomings in accuracy, real-time and environmental adaptability.

[0003] Prior art 1, a technical solution using a mechanical sensor for measurement:

[0004] Mechanical sensors are a common device used to measure the angle and attitude of an object, and are widely used in industries such as industry and aviation. This type of sensor usually senses the angle change of the object being measured through physical contact. For example, a gyroscope or inertial measurement unit (IMU), as a type of mechanical sensor, can detect the angular velocity and attitude changes of an object through the interaction between the internal mass block and the spring system. They can provide accurate measurement data of roll angle (Roll), pitch angle (Pitch), and yaw angle (Yaw). In a system using a mechanical sensor, the sensor is usually fixed on the object being measured, and the corresponding roll angle is calculated by sensing the attitude changes of the object in real time.

[0005] Disadvantages of prior art 1:

[0006] Mechanical sensors require higher precision to process roll angle changes, and because they are based on contact measurement, they are affected by environmental factors such as object vibration and temperature changes during measurement.

[0007] 1) Susceptible to external environmental influences: Mechanical sensors rely on physical contact, and environmental factors such as vibration and temperature changes may cause measurement errors.

[0008] 2) Complex installation and maintenance: Mechanical sensors need to be installed on the object being measured, requiring high-precision fixtures and complex installation processes. In addition, since mechanical sensors have moving parts, they are prone to wear after long-term use, resulting in reduced accuracy and increased maintenance requirements.

[0009] Prior art 2, using photoelectric sensors for measurement:

[0010] Photoelectric sensors mainly rely on optical systems and electronic detection elements to measure the attitude of an object. Common photoelectric measurement methods include the use of photoelectric encoders, infrared rangefinders, or optical attitude measuring instruments. Photoelectric encoders calculate the rotation angle of an object by reading the changes in reflected or transmitted light signals, and are particularly widely used in rotating mechanical equipment. In addition, infrared rangefinders can also be used to detect the distance and angle changes of an object, thereby indirectly measuring the roll angle.

[0011] The advantages of photoelectric sensors are non-contact measurement, which can avoid mechanical wear, fast measurement speed, quick response, and suitable for use in high dynamic environments. Although photoelectric sensors perform well in angle measurement, their accuracy is often limited by factors such as light source stability and ambient light interference.

[0012] Disadvantages of the second prior art:

[0013] 1) Limited scope of application: The measurement distance and angle range of photoelectric sensors are limited, and they are usually only suitable for posture measurement at a shorter distance or a smaller range.

[0014] 2) Limited measurement accuracy: Although photoelectric sensors respond quickly, their measurement accuracy is often lower than that of laser or mechanical sensors, especially in the detection of small angle changes.

[0015] Prior art three, using laser measuring instruments for measurement:

[0016] Laser measuring instruments mainly rely on high-energy laser beams and precision detectors to measure objects. Common laser measurement methods include laser radar, laser interferometer, and laser triangulation. Laser radar obtains the distance and angle information of objects by emitting lasers and receiving reflected signals, calculating the optical path difference, and is widely used in unmanned driving, mapping, and industrial inspection. Laser interferometers are based on the principle of interference and are used for high-precision distance measurement or deformation monitoring, while laser triangulation is suitable for contour and position detection of complex object surfaces.

[0017] The advantages of laser measurement technology are extremely high measurement accuracy, fast response speed, and efficient measurement in complex environments. Since the laser beam has excellent directionality and high energy density, it performs well when measuring small angle changes or objects with complex shapes. In addition, laser measuring instruments can achieve non-contact measurement, which is suitable for posture detection of fragile objects or special environments.

[0018] Disadvantages of the prior art three:

[0019] 1) High cost: Laser measuring instruments are complex to manufacture, expensive, and have relatively high maintenance costs.

[0020] 2) Sensitive to environmental conditions: Laser measurement has high requirements for the external environment. Factors such as dust, haze, and strong light interference may cause measurement errors or signal loss.

[0021] 3) Complex installation and operation: Laser measurement equipment usually requires precise calibration and professional operation, and has high requirements on the use environment and operators.

[0022] In summary, the existing technologies have problems such as poor environmental adaptability, insufficient real-time performance, complex installation and maintenance, and limited measurement accuracy.

[0023] In view of this, the present invention is proposed. Summary of the invention

[0024] The purpose of the present invention is to provide a cylinder roll angle visual measurement algorithm based on a fully surrounded target to solve the above technical problems existing in the prior art.

[0025] The objective of the present invention is achieved through the following technical solutions:

[0026] The cylinder roll angle visual measurement algorithm based on the fully surrounded target of the present invention uses a camera to capture the target pattern and accurately analyzes the roll angle of the cylinder through image processing technology, including the following steps:

[0027] Step 1: Use a uniformly distributed target design method to ensure that the target features are obvious and have anti-interference capabilities;

[0028] Step 2: Determine the exact position of the positioning bar and the ranging bar through image feature enhancement technology and appropriate thresholds;

[0029] Step 3: Perform closing operation and denoising on the positioning bar, and use polygon fitting and vector rotation method to determine the position of the positioning point;

[0030] Step 4: By analyzing the positioning points of the positioning bar, connecting the two midpoints to form the central axis, and solving the relevant geometric constraints and axis equations;

[0031] Step 5: Calculate the distance measurement points by combining the distance measurement strip and the central axis, and select stable feature points as distance measurement references;

[0032] Step 6: Apply the deformation compensation algorithm and the principle of similar triangles to accurately calculate the roll angle.

[0033] Compared with the prior art, the cylinder roll angle visual measurement algorithm based on the fully surrounded target provided by the present invention adopts a non-contact sensor combined with image processing technology, which can capture and analyze the dynamic motion state of the cylinder in real time. It can significantly improve the measurement accuracy and real-time performance, reduce the dependence on the external environment, and simplify the equipment installation and maintenance process. In addition, the flexibility of the system enables it to be widely used in many fields such as automated production lines, scientific research experiments and quality monitoring, meeting the high standards of modern industry for measurement technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A flow chart of a cylinder roll angle visual measurement algorithm based on a fully surrounded target provided by an embodiment of the present invention;

[0035] Figure 2 This is a schematic diagram of a target fully surrounded by a cylinder according to an embodiment of the present invention;

[0036] Figure 3 This is a schematic diagram of a target fully surrounded by a cylinder in three-dimensional space according to an embodiment of the present invention;

[0037] Figure 4 A schematic diagram of determining positioning points and ranging points according to an embodiment of the present invention;

[0038] Figure 5 This is a schematic diagram of a radial plane of a cylinder according to an embodiment of the present invention;

[0039] Figure 6 It is a schematic diagram of the side expansion of a cylindrical target according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention; it is obvious that the described embodiments are only part of the embodiments of the present invention, not all of the embodiments, which does not constitute a limitation of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the protection scope of the present invention.

[0041] First, the terms that may be used in this article are explained as follows:

[0042] The term “and / or” means that either or both of them can be realized at the same time. For example, X and / or Y means both “X” or “Y” and “X and Y”.

[0043] The terms "include", "comprises", "contains", "has" or other descriptions with similar semantics should be interpreted as non-exclusive inclusion. For example, including certain technical feature elements (such as raw materials, components, ingredients, carriers, dosage forms, materials, dimensions, parts, components, mechanisms, devices, steps, procedures, methods, reaction conditions, processing conditions, parameters, algorithms, signals, data, products or products, etc.) should be interpreted as including not only certain technical feature elements explicitly listed, but also other technical feature elements known in the art that are not explicitly listed.

[0044] The term "consisting of..." means excluding any technical feature elements not explicitly listed. If this term is used in a claim, it will make the claim closed, so that it does not contain technical feature elements other than the technical feature elements explicitly listed, except for the conventional impurities related to them. If this term only appears in a clause of a claim, it only limits the elements explicitly listed in the clause, and the elements recorded in other clauses are not excluded from the overall claim.

[0045] The contents not described in detail in the examples of the present invention belong to the prior art known to professionals in the field. If no specific conditions are specified in the examples of the present invention, the conditions are carried out according to the conventional conditions in the field or the conditions recommended by the manufacturer. If the manufacturer is not specified in the reagents or instruments used in the examples of the present invention, they are all conventional products that can be purchased commercially.

[0046] The cylinder roll angle visual measurement algorithm based on the fully surrounded target of the present invention uses a camera to capture the target pattern and accurately analyzes the roll angle of the cylinder through image processing technology, including the following steps:

[0047] Step 1: Use a uniformly distributed target design method to ensure that the target features are obvious and have anti-interference capabilities;

[0048] Step 2: Determine the exact position of the positioning bar and the ranging bar through image feature enhancement technology and appropriate thresholds;

[0049] Step 3: Perform closing operation and denoising on the positioning bar, and use polygon fitting and vector rotation method to determine the position of the positioning point;

[0050] Step 4: By analyzing the positioning points of the positioning bar, connecting the two midpoints to form the central axis, and solving the relevant geometric constraints and axis equations;

[0051] Step 5: Calculate the distance measurement points by combining the distance measurement strip and the central axis, and select stable feature points as distance measurement references;

[0052] Step 6: Apply the deformation compensation algorithm and the principle of similar triangles to accurately calculate the roll angle.

[0053] In the step 1, the fully enclosed target is designed to be a "Z" shape composed of two red positioning bars and a yellow distance measuring bar on the unfolded rectangle on the side of the cylinder.

[0054] The step 2, determining the positions of the positioning bar and the distance measuring bar, includes any one or both of the following:

[0055] Method 1: Use the RGB(R) minus Gray feature enhancement method to make the red positioning bar more prominent in the image, and then use the threshold segmentation method to process the positioning bar. By setting an appropriate threshold, the RGB(R)-Gray mode part in the image is extracted.

[0056] Method 2: In the CMYK color space, the Y channel represents yellow, the M channel represents magenta, and the K channel represents black; the feature enhancement method of CMYK (YKM) is used to further enhance the recognizability of the yellow ranging bar;

[0057] Through this color space conversion, the edges of the ranging bars are ensured to be clear, which improves the accuracy of the subsequent ranging algorithm. Then, a suitable threshold is set to extract the corresponding part of the image, ensuring that the features of the ranging bars are clearly visible.

[0058] The step 3, solving the positioning point to accurately locate the positioning point on the target, includes:

[0059] First, the closing operation is performed to denoise the image after threshold segmentation, and the morphological closing operation is performed on the image;

[0060] Through dilation and corrosion operations, small noise is eliminated and the connectivity of features is enhanced, making the edge of the target smoother and more continuous, thereby improving the stability of subsequent feature point extraction;

[0061] A polygon fitting algorithm is used to fit the characteristic points of the positioning strip after the closing operation;

[0062] This method simplifies the representation of feature points, extracts the geometric shape of the positioning bar, and obtains a more accurate determination of the position of the feature points;

[0063] Finally, the positioning point is determined by using the vector rotation method, and the vector angle is calculated through the known joint centroid coordinates of the two positioning strips and the polygon fitting feature points of each positioning strip to form a fan-shaped area;

[0064] The two required positioning points are on both sides of the area, and the positions of the positioning points are calculated.

[0065] The step 4, determining the central axis, includes:

[0066] First, the positioning point analysis is performed to extract the characteristic points on the positioning strips. By analyzing the spatial distribution of the points, the basic data for calculating the central axis is prepared to avoid calculating the line perpendicular to the central axis. Then, the midpoint of each positioning strip is calculated, and the central axis is formed by connecting the two midpoints.

[0067] This process can use the coordinate average method to calculate the midpoint. By calculating the midpoint coordinates of each positioning point, a central axis is determined and expressed as a mathematical equation:

[0068]

[0069] The step 5, performing the calculation of the ranging points to accurately calculate the rolling angle, includes:

[0070] Firstly, the intersection of the central axis and the distance measuring strip is calculated, and the intersection point between the central axis and the distance measuring strip is detected by using a geometric algorithm;

[0071] By intersecting the straight line with the ranging strip mask, all the intersection points are obtained, and the calculated intersection points are collected to form a point set. The coordinates of each intersection point will be used for subsequent ranging point selection;

[0072] Then the distance measurement point is selected. According to the location and distribution characteristics of the intersection points, the optimal distance measurement point is selected. The intersection point at the center of the distance measurement strip is selected as the distance measurement point p0 to ensure the accuracy of the calculation. This selection process is combined with geometric constraints and point stability for comprehensive evaluation to ensure that the final selected distance measurement point is not affected by external interference during the measurement process.

[0073] Similar to the method of determining the distance measurement point, a straight line is used to intersect the masks of the two positioning strips to calculate the central axis positioning points p1 and p2, and then the distance of the positioning strip on the central axis is solved.

[0074] The step 6 comprises:

[0075] Step 1): Deformation compensation algorithm, when deformation is detected, uses perspective transformation technology to perform geometric correction on the image to restore the deformed image and ensure the accuracy of measurement;

[0076] Step 2): Calculate the roll angle of the cylinder using the similar triangle principle through the height difference between the distance measuring point and the central axis;

[0077] Set the vertical distance of the measuring point on the central axis to The distance between the positioning bars on the central axis is

[0078] In summary, the cylinder roll angle visual measurement algorithm based on the fully surrounded target in the embodiment of the present invention adopts a non-contact sensor combined with image processing technology, and can capture and analyze the dynamic motion state of the cylinder in real time.

[0079] The basic idea of ​​the present invention is:

[0080] Through high-frequency image acquisition and analysis, high-precision measurement of cylinders under different environmental conditions can be achieved. The system uses computer vision algorithms to process the captured images, extract key feature points, and then calculate the roll angle.

[0081] The overall technical effects brought by this solution include: significantly improving measurement accuracy and real-time performance, reducing dependence on the external environment, and simplifying the equipment installation and maintenance process. In addition, the flexibility of the system enables it to be widely used in many fields such as automated production lines, scientific research experiments and quality monitoring, meeting the high standards of modern industry for measurement technology.

[0082] In order to more clearly demonstrate the technical solution and technical effects provided by the present invention, the embodiments of the present invention are described in detail with specific embodiments below.

[0083] Example 1

[0084] A method for measuring the roll angle of a cylinder based on a fully surrounded target is commonly used in the fields of industrial automation and mechanical monitoring. The specific implementation steps of the present invention will be described in detail below with reference to the accompanying drawings.

[0085] like Figure 1 As shown, in this method, a camera is used to capture the target pattern, and the roll angle of the cylinder is accurately analyzed through image processing technology.

[0086] Step 1: Use a uniformly distributed target design method to ensure that the target features are obvious and have anti-interference capabilities.

[0087] Step 2: Determine the exact position of the positioning bar and the ranging bar through image feature enhancement technology and appropriate thresholds.

[0088] Step 3: Perform closing operation and denoising on the positioning bar, and use polygon fitting and vector rotation method to determine the position of the positioning point.

[0089] Step 4: Analyze the positioning points of the positioning bar, connect the two midpoints to form the central axis, and solve the relevant geometric constraints and axis equations.

[0090] Step 5: Combine the distance measuring strip and the central axis to calculate the distance measuring points, and select stable feature points as distance measuring references.

[0091] Step 6: Apply the deformation compensation algorithm and the principle of similar triangles to accurately calculate the roll angle.

[0092] Examples of specific implementation methods:

[0093] 1. First, the target needs to be designed, which is the basis of the entire measurement process to ensure that it can be accurately captured at any roll angle. The design must ensure uniform distribution, obvious color or geometric features, and a certain degree of anti-interference ability. The specific requirements are as follows:

[0094] Ⅰ Uniform distribution: Arranging fully surrounded targets on the surface of the cylinder can capture characteristic patterns that can be captured by the camera at any roll angle. The distribution of these patterns forms a surrounding structure on the surface of the cylinder, ensuring that the characteristic patterns can continuously appear in the field of view of the camera when the roll angle changes.

[0095] Ⅱ Obvious color or geometric features: Choose patterns with high color contrast and clear edges. They should be easily identifiable in the image through edge detection and color segmentation algorithms.

[0096] Ⅲ Ability to resist environmental interference: Target design also needs to take into account interference factors such as external lighting changes and image noise, so that the target pattern can still be stably captured even under complex lighting conditions.

[0097] Therefore, the fully enclosed target is designed as a combination of two red positioning bars and a yellow distance measuring bar on the side of the cylinder in a "Z" shape. The specific target design is as follows: Figure 2 Figure 3 shown.

[0098] 2. After the target is designed, it is necessary to determine the positioning bar and the ranging bar to enhance the image features and lay the foundation for subsequent calculations.

[0099] The feature enhancement method of RGB(R) minus Gray is used to make the red positioning bar more prominent in the image. This process can significantly improve the accuracy of positioning point extraction and provide a clear positioning point basis for subsequent image processing. The positioning bar is then processed using the threshold segmentation method, and the RGB(R)-Gray mode part in the image is extracted by setting an appropriate threshold.

[0100] In addition, in the CMYK color space, the Y channel represents yellow, the M channel represents magenta, and the K channel represents black. The feature enhancement method of CMYK (YKM) is used to further enhance the recognizability of the yellow rangefinder. Through this color space conversion, the edge of the rangefinder is ensured to be clear, which improves the accuracy of the subsequent rangefinder algorithm. Then, a suitable threshold is set to extract the corresponding part of the image to ensure that the features of the rangefinder are clearly visible.

[0101] 3. After determining the specific positions of the positioning bar and the distance measuring bar, the positioning point is solved to accurately locate the positioning point on the target. First, the closed operation is performed to remove noise, and the morphological closed operation is performed on the image after threshold segmentation. Through the expansion and corrosion operations, small noise is eliminated and the connectivity of the features is enhanced, making the edge of the target smoother and continuous, thereby improving the stability of subsequent feature point extraction. The polygon fitting algorithm is used to fit the feature points of the positioning bar after the closing operation. This method simplifies the representation of feature points, extracts the geometric shape of the positioning bar, and obtains a more accurate determination of the position of the feature points.

[0102] 4. Finally, the vector rotation method is used to determine the positioning point. The vector angle is calculated through the known joint centroid coordinates of the two positioning bars and the polygon fitting feature points of each positioning bar to form a fan-shaped area. The required two positioning points are on both sides of the area, and the positions of the positioning points are calculated. Figure 4 shown.

[0103] 5. After completing the positioning point solution, the central axis needs to be determined, which is a necessary condition for the distance measurement point solution. First, perform positioning point analysis, extract the characteristic points on the positioning bar, and prepare the basic data for calculating the central axis by analyzing the spatial distribution of the points to avoid calculating the line perpendicular to the central axis. Then calculate the midpoint of each positioning bar, and form the central axis by connecting the two midpoints. This process can use the coordinate averaging method to calculate the midpoint; by calculating the midpoint coordinates of each positioning point, a central axis can be determined and expressed by a mathematical equation.

[0104]

[0105] 6. After the center axis is determined, the distance measurement point can be solved to calculate the accurate roll angle, such as Figure 4 As shown. First, calculate the intersection of the central axis and the ranging strip, and use the geometric algorithm to detect the intersection between the central axis and the ranging strip. By intersecting the straight line and the ranging strip mask, all the intersection points are obtained, and the calculated intersection points are collected and sorted to form a point set. The coordinates of each intersection point will be used for subsequent ranging point selection. Then the ranging point selection is carried out, and the optimal ranging point is selected according to the position and distribution characteristics of the intersection point. The intersection point at the center of the ranging strip is selected as the ranging point p0 to ensure the accuracy of the calculation. This selection process can be combined with geometric constraints and point stability for comprehensive evaluation to ensure that the final selected ranging point is not affected by external interference during the measurement process.

[0106] Similar to the method of determining the distance measurement point, the center axis positioning points p1 and p2 are calculated by using a straight line to intersect the masks of the two positioning bands, as shown in Figure 4 As shown. Then the distance of the positioning bar on the central axis is calculated.

[0107] 7. Finally, the roll angle is calculated to obtain the roll angle of the cylinder, such as Figure 5 , 6 This step includes:

[0108] Step 1: Deformation compensation algorithm, when deformation is detected, the image is geometrically corrected using perspective transformation technology to restore the deformed image and ensure the accuracy of the measurement.

[0109] Step 2: Principle of similar triangles: Calculate the roll angle of the cylinder by using the height difference between the distance measuring point and the central axis. Set the vertical distance of the distance measuring point on the central axis to The distance between the positioning bars on the central axis is

[0110]

[0111] Beneficial effects brought by the technical solution of the present invention:

[0112] High-precision measurement: The fully enclosed target design ensures that sufficient image information can be obtained at various viewing angles, overcoming the recognition failure problem caused by viewing angle changes in traditional methods.

[0113] Highly cost-effective: It does not rely on expensive hardware equipment, and only requires a single camera and a simple image processing system to achieve high-precision measurement of the roll angle.

[0114] Improved real-time performance: The present invention reduces the computational burden of three-dimensional modeling and data fusion through simplified image processing and angle calculation algorithms, and significantly improves the real-time response capability of the system.

[0115] Enhanced environmental adaptability: The present invention can adapt to the measurement of cylinder roll angles in different scenarios.

[0116] In specific implementation, the following solution can be used instead

[0117] Target design: The positioning bar and the distance measuring bar can be replaced by other colors with obvious characteristics to improve the recognition accuracy and the ability to adapt to different lighting environments.

[0118] Feature enhancement and segmentation of positioning bars and ranging bars: In addition to using the RGB color space, you can also consider HSV, Lab and other color spaces to enhance the significance of positioning bars and ranging bars. In addition to using the Otsu algorithm to automatically determine the threshold, you can also use a histogram analysis-based method or an adaptive threshold segmentation technique.

[0119] Position point calculation: The positioning point calculation method can be achieved through different image processing algorithms, such as corner point detection algorithms (such as Harris corner point detection), or feature point extraction through machine learning methods (such as convolutional neural networks) to improve positioning accuracy. The calculation results can also be further optimized through sub-pixel precision optimization algorithms (such as iterative closest point ICP algorithm).

[0120] Distance measurement point solution: A minimum energy-based method can be introduced to select the point with the smallest measurement error among multiple intersection points to further optimize the selection of distance measurement points.

[0121] Key technical points of the present invention:

[0122] Fully enclosed target design: fully enclosed target is used to improve the viewing angle range and accuracy of measurement. Multiple visual feature points are set on the target, and these feature points are used for image processing and analysis.

[0123] Image processing algorithm: We have developed an efficient image processing algorithm that can quickly identify and extract target feature points. Combined with image distortion compensation technology, it ensures stable measurement under different environmental conditions.

[0124] Dynamic calculation of roll angle: Through real-time image acquisition and feature point tracking, the roll angle of the cylinder is dynamically calculated, and the geometric model is established using the principle of similar triangles to improve the accuracy and real-time performance of the calculation.

[0125] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed in the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims. The information disclosed in the background technology section of this article is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as an admission or in any form that the information constitutes prior art known to those skilled in the art.

Claims

1. A cylindrical roll angle visual measurement algorithm based on a fully surrounded target, characterized in that: The target pattern is captured by a camera, and the roll angle of the cylinder is accurately analyzed through image processing technology, including the following steps: Step 1: Use a uniformly distributed target design method to ensure that the target features are obvious and have anti-interference capabilities; Step 2: Determine the exact position of the positioning bar and the ranging bar through image feature enhancement technology and appropriate thresholds; Step 3: Perform closing operation and denoising on the positioning bar, and use polygon fitting and vector rotation method to determine the position of the positioning point; Step 4: By analyzing the positioning points of the positioning bar, connecting the two midpoints to form the central axis, and solving the relevant geometric constraints and axis equations; Step 5: Calculate the distance measurement points by combining the distance measurement strip and the central axis, and select stable feature points as distance measurement references; Step 6: Apply the deformation compensation algorithm and the principle of similar triangles to accurately calculate the roll angle.

2. The cylinder roll angle visual measurement algorithm based on a fully surrounded target according to claim 1 is characterized in that: In the step 1, the fully enclosed target is designed to be a "Z" shape composed of two red positioning bars and a yellow distance measuring bar on the unfolded rectangle on the side of the cylinder.

3. The cylinder roll angle visual measurement algorithm based on a fully surrounded target according to claim 2 is characterized in that: The step 2, determining the positions of the positioning bar and the distance measuring bar, includes any one or both of the following: Method 1: Use the RGB(R) minus Gray feature enhancement method to make the red positioning bar more prominent in the image, and then use the threshold segmentation method to process the positioning bar. By setting an appropriate threshold, the RGB(R)-Gray mode part in the image is extracted. Method 2: In the CMYK color space, the Y channel represents yellow, the M channel represents magenta, and the K channel represents black; the feature enhancement method of CMYK (YKM) is used to further enhance the recognizability of the yellow ranging bar; Through this color space conversion, the edges of the ranging bars are ensured to be clear, which improves the accuracy of the subsequent ranging algorithm. Then, a suitable threshold is set to extract the corresponding part of the image, ensuring that the features of the ranging bars are clearly visible.

4. The cylinder roll angle visual measurement algorithm based on a fully surrounded target according to claim 3 is characterized in that: The step 3, solving the positioning point to accurately locate the positioning point on the target, includes: First, the closing operation is performed to denoise the image after threshold segmentation, and the morphological closing operation is performed on the image; Through dilation and corrosion operations, small noise is eliminated and the connectivity of features is enhanced, making the edge of the target smoother and more continuous, thereby improving the stability of subsequent feature point extraction; A polygon fitting algorithm is used to fit the characteristic points of the positioning strip after the closing operation; This method simplifies the representation of feature points, extracts the geometric shape of the positioning bar, and obtains a more accurate determination of the position of the feature points; Finally, the positioning point is determined by using the vector rotation method, and the vector angle is calculated through the known joint centroid coordinates of the two positioning strips and the polygon fitting feature points of each positioning strip to form a fan-shaped area; The two required positioning points are on both sides of the area, and the positions of the positioning points are calculated.

5. The cylinder roll angle visual measurement algorithm based on a fully surrounded target according to claim 4 is characterized in that: The step 4, determining the central axis, includes: First, the positioning point analysis is performed to extract the characteristic points on the positioning strips. By analyzing the spatial distribution of the points, the basic data for calculating the central axis is prepared to avoid calculating the line perpendicular to the central axis. Then, the midpoint of each positioning strip is calculated, and the central axis is formed by connecting the two midpoints. This process can use the coordinate average method to calculate the midpoint. By calculating the midpoint coordinates of each positioning point, a central axis is determined and expressed as a mathematical equation:

6. The cylinder roll angle visual measurement algorithm based on a fully surrounded target according to claim 5 is characterized in that: The step 5, performing the calculation of the ranging points to accurately calculate the rolling angle, includes: Firstly, the intersection of the central axis and the distance measuring strip is calculated, and the intersection point between the central axis and the distance measuring strip is detected by using a geometric algorithm; By intersecting the straight line with the ranging strip mask, all the intersection points are obtained, and the calculated intersection points are collected to form a point set. The coordinates of each intersection point will be used for subsequent ranging point selection; Then the distance measurement point is selected. According to the location and distribution characteristics of the intersection points, the optimal distance measurement point is selected. The intersection point at the center of the distance measurement strip is selected as the distance measurement point p0 to ensure the accuracy of the calculation. This selection process is combined with geometric constraints and point stability for comprehensive evaluation to ensure that the final selected distance measurement point is not affected by external interference during the measurement process. Similar to the method of determining the distance measurement point, a straight line is used to intersect the masks of the two positioning strips to calculate the central axis positioning points p1 and p2, and then the distance of the positioning strip on the central axis is solved.

7. The cylinder roll angle visual measurement algorithm based on a fully surrounded target according to claim 6 is characterized in that: The step 6 comprises: Step 1): Deformation compensation algorithm, when deformation is detected, uses perspective transformation technology to perform geometric correction on the image to restore the deformed image and ensure the accuracy of measurement; Step 2): Calculate the roll angle of the cylinder using the similar triangle principle through the height difference between the distance measuring point and the central axis; Set the vertical distance of the measuring point on the central axis to The distance between the positioning bars on the central axis is

Images