A method for precise measurement of geometric features of inner surface of small hole with high aspect ratio

Abstract

The present invention belongs to the technical field of geometric quantity precision measurement based on machine vision, and discloses a method for precision measurement of geometric features of the inner surface of a small hole with a high aspect ratio. The method of the present invention is based on a measurement system of a ring laser, and realizes non-destructive measurement of the inner surface of a deep hole. Due to the high collimation of the ring laser, the integrated measurement of geometric features such as the size, roundness, straightness, surface waviness of the deep hole is realized; the image is processed during the measurement process, and the accuracy of obtaining the inner surface information of the deep hole is improved. In the process of restoring the image, a complete measurement model is established, and the image is restored quickly and accurately.

Description

Technical Field

[0001] The present invention belongs to the technical field of geometric quantity precision measurement based on machine vision, and in particular provides a method for precisely measuring the geometric features of the inner surface of a small hole with a high aspect ratio. Background Art

[0002] Core equipment in major projects such as nuclear power, energy and power generation, and aerospace features high technical content, significant manufacturing complexity, and demanding reliability. Its comprehensive manufacturing capabilities are a concentrated reflection of my country's overall industrial and scientific strength. Deep-hole machining technology plays a crucial role in the equipment manufacturing industry due to its multidisciplinary nature, complex machining environment, high difficulty, and high cost. This is particularly true for typical small-diameter deep-hole components, such as nuclear power steam generator tubesheet ducts, steam generator heat exchange metal molybdenum tubes, and internal combustion engine crankshaft oil holes. These components operate in environments characterized by high temperatures, high pressures, and severe impact loads. Excessive errors in bore diameter, roundness, and straightness during deep-hole machining can directly impact component performance and, in severe cases, lead to component failure and scrapping, significantly impacting workpiece performance. Consequently, extremely high demands are placed on the machining accuracy and quality of these small-diameter deep-hole components. This, in turn, increases the accuracy requirements for key geometric parameters such as bore diameter, roundness, and straightness. While external and planar dimension measurements have now reached resolutions of 1nm or even 0.1nm, the relative accuracy of deep hole internal dimension measurements is approximately 0.2% to 0.5%. Currently, there is no mature and stable solution internationally for measuring arbitrary cross-sectional dimensional and shape errors in deep, small holes with diameters less than 20mm and aspect ratios greater than 40. However, internal surface inspection of deep-hole components plays a crucial role in product quality control and fault diagnosis, making the development of integrated methods for multi-geometric measurement of small-diameter, deep holes of great significance.

[0003] Currently, the commonly used normal vector detection methods are divided into two categories, including contact measurement methods and non-contact measurement methods:

[0004] Contact measurement methods include oscillation scanning and internal diameter dial indicator measurement. In 1993, Masuzawa of the University of Tokyo proposed a scanning method using probe oscillation for testing workpieces with a sample hole depth of 700 μm and a diameter of 200 to 300 μm in his paper "Vibroscanning Method for Nondestructive Measurement of Small Holes." This method uses an oscillator to drive a stylus at a specific frequency, generating small amplitude vibrations perpendicular to the surface being measured. When the stylus makes critical contact with the surface, the circuit between the two is connected, the sensitive circuit closes, and an electrical signal is detected. Because the stylus oscillates continuously, the circuit becomes intermittent as it approaches the object. The duration of the signal detection period during one oscillation cycle is determined by the distance between the stylus and the object being measured: the shorter the distance, the longer the contact period.

[0005] Non-contact measurement methods include laser triangulation and the halo cross-section method. In the paper "ALaser Triangulation-Based 3D Measurement System for Inner Surface of Deep Holes," Zhai Chao of the University of Science and Technology of China (USTC) investigated the effects of assembly errors and refractive distortion based on the principle of laser triangulation and proposed a flexible laser plane calibration technique based on binocular vision to accurately capture the cross-sectional profile features of deep holes. A halo cross-section method: Su Jie and others from Tianjin University developed an inspection system suitable for large and medium-sized pipelines. The authors used a light ring modulated by an optical system to project a light band onto the inner wall of the pipeline. The inner surface information was then captured using a CCD. Finally, the grayscale image of the ring was analyzed on a PC to detect the inner wall's topography and defects. However, this method requires that the visual sensor and the measured aperture must be on the same axis, resulting in low applicability. In view of this, the current measurement of the geometric characteristics of the inner surface of deep holes has the problems of single measurement parameters and discontinuous measurement data, which requires the inner surface parameters to be measured multiple times in steps. There are problems such as complex measurement process, low measurement efficiency, and poor accuracy. In order to solve the above problems, it is necessary to find a method for precise measurement of the inner surface geometry of small holes with a high aspect ratio. Summary of the Invention

[0006] Based on the above problems, the present invention establishes a method for precisely measuring the geometric features of the inner surface of a small hole with a high aspect ratio.

[0007] The technical solution of the present invention:

[0008] A method for rapid and high-precision measurement of the geometric features of the inner surface of a deep hole, the steps are as follows:

[0009] Step 1: Build an endoscopic measurement system to acquire images

[0010] The endoscopic measurement system includes a light source emission module, an inner surface imaging module and an image acquisition module;

[0011] The light source emission module is composed of a point laser and an optical diffraction element, which emits a collimated ring laser;

[0012] The inner surface imaging module consists of a frustum-shaped reflector with a slope of 50° and a conical reflector with an upper angle of 40°;

[0013] The image acquisition module consists of a geometric reflecting prism with an angle of 45° and a CCD camera vertically below it;

[0014] The inner surface imaging module, image acquisition module, and light source emission module are arranged from left to right and encapsulated by a glass tube, ensuring that the truncated cone reflector, conical reflector, point laser, and the axis of the glass tube coincide.

[0015] During measurement, the endoscopic measurement system is placed inside the deep hole, ensuring that the axis of the glass tube is completely aligned with the axis of the deep hole. The light source emission module emits a collimated ring laser beam, which passes through the image acquisition module and enters the inner surface imaging module. The ring laser beam is first incident on the surface of a 50° frustum-shaped reflector, and is totally reflected by the reflector to the inner surface of the deep hole. There, the reflector is totally reflected again on the inner surface of the deep hole and enters the 40° conical reflector. Finally, the reflector is totally reflected again on the surface of the conical reflector and emitted in parallel, entering the geometric reflection prism in the image acquisition module, where it is vertically redirected and enters the CCD camera below the geometric reflection prism to obtain an image.

[0016] Step 2: Image processing

[0017] After acquiring the image of the deep hole surface from the CCD camera, due to the problem of the annular laser spot width and the influence of the deep hole inner surface roughness on the laser reflectivity, there will be halos and dark areas in the image obtained in the previous step. These halos and dark areas will thicken the laser spot, resulting in the final image being a blurred strip-shaped circle. This image will reduce the accuracy of this measurement scheme, so the obtained image needs to be processed; after using the image processing algorithm to extract the edge of the obtained image, the least squares method is used to fit its edge to obtain a linear circle to improve the accuracy of the information it carries, providing more accurate initial data for subsequent image restoration;

[0018] Step 3: Image restoration

[0019] In this measurement method, the distance from the inner surface of the deep hole to the deep hole axis is used to represent the geometric characteristics of the deep hole inner surface; the distance from a point on the image obtained by the CCD camera to the center of the image is used to represent the distance from the inner surface of the deep hole to the deep hole axis. However, in the endoscopic measurement system, due to multiple reflections of light, the distance from the inner surface of the deep hole to the deep hole axis is not directly equal to the distance from a point on the image obtained by the CCD camera to the center of the image. Therefore, it is necessary to substitute the distance from a point on the image obtained by the CCD camera to the center of the image into the mathematical model established based on the optical path model to restore the distance from the inner surface of the deep hole to the deep hole axis, thereby representing the geometric characteristics of the inner surface of the deep hole.

[0020] The geometric features of the deep hole inner surface are represented by the distance R from the deep hole inner surface to the deep hole axis, and the distance r from a point on the image obtained by the CCD camera to the image center is represented. In the process of restoring the distance from the deep hole inner surface to the deep hole axis, a polar coordinate system is established on the image obtained by the CCD camera with the image center as the polar coordinate center. According to the value of the independent variable θ, as the independent variable θ gradually increases from 0 to 2π, it is easy to obtain the value of the distance r from a point on the image obtained by the CCD camera to the image center in the interval [0, 2π]. The mathematical model is expressed as follows:

[0021]

[0022] Substitute the distance r from a point on the image obtained by the CCD camera to the center of the image into the above mathematical model to obtain the value of the distance R from the inner surface of the deep hole to the axis of the deep hole, and finally complete the measurement.

[0023] Beneficial effects of the present invention: The present invention is based on a ring laser measurement system, which realizes non-destructive measurement of the inner surface of a deep hole. Due to the high collimation of the ring laser, the integrated measurement of geometric characteristics such as deep hole size, roundness, straightness, surface waviness, etc. is realized; the image is processed during the measurement process, which improves the accuracy of obtaining deep hole inner surface information; in the process of restoring the image, a complete measurement model is established, and the image is restored quickly and accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 Diagram of the measuring device.

[0025] Figure 2 It is a two-dimensional schematic diagram of the optical path model.

[0026] Figure 3 Schematic diagram of deep hole inner surface measurement, where (a) is the measurement diagram when the inner surface is intact, and (b) is the measurement diagram when the inner surface is damaged.

[0027] Figure 4 Figure 2 is an image processing process diagram, where (a) is the image taken by the CCD camera and (b) is the image after processing.

[0028] Figure 5 Restore the flow chart for the image.

[0029] Figure 6 Diagram for optical modeling.

[0030] Figure 7 This is the simulation result of the captured image.

[0031] Figure 8 The simulation result diagram is restored for the image. DETAILED DESCRIPTION

[0032] The specific implementation of the present invention is further described below in conjunction with the accompanying drawings and technical solutions.

[0033] A method for rapid and high-precision measurement of cross-sectional profile features in deep holes, the steps are as follows:

[0034] Step 1: Build an endoscopic measurement system to acquire images

[0035] The endoscopic measurement system includes a light source emission module, an inner surface imaging module and an image acquisition module;

[0036] The light source emission module is composed of a point laser and an optical diffraction element, which emits a collimated ring laser;

[0037] The inner surface imaging module consists of a frustum-shaped reflector with a slope of 50° and a conical reflector with an upper angle of 40°;

[0038] The image acquisition module consists of a geometric reflecting prism with an angle of 45° and a CCD camera vertically below it;

[0039] The inner surface imaging module, image acquisition module, and light source emission module are arranged from left to right and encapsulated by a glass tube, ensuring that the truncated cone reflector, conical reflector, point laser, and the axis of the glass tube coincide.

[0040] During measurement, the endoscopic measurement system is placed inside the deep hole, ensuring that the axis of the glass tube is completely aligned with the axis of the deep hole. The light source emission module emits a collimated ring laser beam, which passes through the image acquisition module and enters the inner surface imaging module. The ring laser beam is first incident on the surface of a 50° frustum-shaped reflector, and is totally reflected by the reflector to the inner surface of the deep hole. There, the reflector is totally reflected again on the inner surface of the deep hole and enters the 40° conical reflector. Finally, the reflector is totally reflected again on the surface of the conical reflector and emitted in parallel, entering the geometric reflection prism in the image acquisition module, where it is vertically redirected and enters the CCD camera below the geometric reflection prism to obtain an image.

[0041] Step 2: Image processing

[0042] Because of the problem of annular laser spot width and the influence of the geometric characteristics of the inner surface of the measurement target on the laser reflectivity, there will be halos, dark areas and nonlinear fluctuations in the image obtained in the previous step. In the CCD camera, halos and dark areas will make the laser spot thicker, such as Figure 4 (a) shows that the image needs to be processed. In this solution, the image processing algorithm is used to extract its edges, and then the least squares method is used to fit its edges to obtain a more accurate image to improve the accuracy of the information it wants to express. The processed image is shown in Figure Figure 4 (b) shown.

[0043] After obtaining the image processed in the previous step, it is necessary to restore and express the information carried by the image by establishing a measurement model to obtain the true geometric features of the deep hole inner surface.

[0044] Combine Figure 2 and Figure 5 Explanation and explanation of step 3:

[0045] The measurement model is shown in the figure. The geometric features of the inner surface of the deep hole mainly include roundness and surface waviness. The maximum change in the radius of the inner surface of the deep hole represents its roundness, and surface waviness is a unit of measurement that is microscopic and greater than the surface roughness. The measurement of roundness and surface waviness requires measuring the radius of each point on the inner surface of the deep hole, and the final measurement result is obtained by calculating the difference in radius. In the measurement model constructed this time, the distance R from a point on the inner surface of the deep hole to the axis of the deep hole is used to represent the radius. How to restore the distance R from a point on the inner surface of the deep hole to the axis of the deep hole based on the distance r obtained in measurement step 2 is the focus of this step 3.

[0046] according to Figure 2 The relationship between r and R is deduced from the two-dimensional schematic diagram of the measurement model. It is known that: PT is the radius of the halo IV, and UVEU, whose lengths are both known and ∠PIT=40°∠LEA1=50°; IH is extended to intersect with PL (light); from the geometric relationship, it is easy to obtain the isosceles triangle ZPN; NS=sin40°IN; IN=IP-NP=PT / sin40°-NP (PT is the halo radius); ∵ isosceles triangle ZPN; ∴NP=2cos40°ZP; ZP=ZL+LP LP=VT+(UV-EA1)∵∠LEA1=50°LA1=PT-EU∴EA1=LA1 / tan50°; LA1=PT-EU ZL=2MQ / tan80°; VT=PT / tan40°-IV

[0047] So in the end

[0048]

[0049] We know that MQ = R - PT, NS = r, and we can get:

[0050]

[0051] Put forward R and substitute r measured from the image to get:

[0052]

[0053] The above calculation model is for the restoration and expression between single points. Because the deep hole is a measurement target that can be approximately regarded as a cylinder, the restoration of the entire image can be completed by rotating a single point one circle. This positive characteristic is suitable for restoration in polar coordinates. As shown in the restoration process from point to circle, after obtaining r from step 2, arbitrarily determine an angle as the initial angle, calculate the corresponding r at this angle, substitute it into the above calculation model, obtain the corresponding R value at this angle and mark it on the image, select the next angle, and repeat the above steps until the image is complete.

[0054] Example

[0055] Use the optical software tracepro to simulate. First, establish the optical measurement unit according to step 1 and simulate the light path operation. Figure 6 shown.

[0056] By using the ray tracing function of TracePro, the coordinates of all the light rays obtained on the CCD camera can be accurately simulated, thereby restoring the coordinate data inside the measurement target according to the established mathematical model during the simulation process.

[0057] During the simulation, a well-established deep hole is measured, and the established measuring unit is placed into the cylinder. Ray tracing is performed in TracePro, and finally the coordinates of the light rays obtained from the CCD camera image in step 2 are accurately simulated. The coordinate data obtained from the simulated CCD camera in TracePro is converted into the coordinate data according to the formula (r = √α 2 +β 2 ) to get r, and plot the data r in combination with the angle, the effect is as follows Figure 7 shown

[0058] Substitute the obtained data into the established mathematical model to restore the data R that can represent the geometric characteristics of the deep hole inner surface. According to the restoration process, the complete image is output. The result is as follows: Figure 8 shown.

[0059] The descriptions presented in the above exemplary embodiments are only intended to illustrate the technical solutions of the present invention and are not intended to be exhaustive or to limit the present invention to the precise forms described. Obviously, it is possible for a person of ordinary skill in the art to make many changes and variations based on the above teachings. The exemplary embodiments are selected and described in order to explain the specific principles of the present invention and its practical applications, so that other persons skilled in the art can easily understand, implement and utilize the various exemplary embodiments of the present invention and its various selected forms and modified forms. The scope of protection of the present invention is intended to be defined by the appended claims and their equivalents.

Claims

1. A method for rapid and high-precision measurement of the geometric features of the inner surface of a deep hole, characterized in that: Here are the steps: Step 1: Build an endoscopic measurement system to obtain images The endoscopic measurement system includes a light source emission module, an inner surface imaging module and an image acquisition module; The light source emission module is composed of a point laser and an optical diffraction element, so that it emits a collimated ring laser; The inner surface imaging module consists of a truncated cone reflector with a slope of 50° and a conical reflector with an upper angle of 40°; The image acquisition module consists of a geometric reflection prism with an angle of 45° and a CCD camera vertically below it; The inner surface imaging module, image acquisition module and light source emission module are arranged from left to right and encapsulated by a glass tube, ensuring that the truncated cone reflector, conical reflector, point laser and the axis of the glass tube coincide; During measurement, the endoscopic measurement system is made to enter the deep hole and ensure that the axis of the glass tube is completely coincident with the axis of the deep hole. The light source emission module emits a beam of collimated annular laser. After the annular laser passes through the image acquisition module and enters the inner surface imaging module, it is first incident on the surface of the 50° frustum-type reflector and is totally reflected to the inner surface of the deep hole. It is totally reflected again on the inner surface of the deep hole and enters the 40° conical reflector. Finally, it is totally reflected again on the surface of the conical reflector and emitted in parallel, entering the geometric reflection prism in the image acquisition module, and is vertically redirected to enter the CCD camera below the geometric reflection prism to obtain an image. Step 2: Image processing After the edge of the image is extracted using an image processing algorithm, the least squares method is used to fit the edge to obtain a linear circle to improve the accuracy of the information it carries, providing more accurate initial data for subsequent image restoration; Step 3: Image restoration In this measurement method, the distance from the inner surface of the deep hole to the axis of the deep hole is used to represent the geometric features of the inner surface of the deep hole; the distance from a point on the image obtained by the CCD camera to the center of the image is used to represent the distance from the inner surface of the deep hole to the axis of the deep hole. However, in the endoscopic measurement system, with multiple reflections of light, the distance from the inner surface of the deep hole to the axis of the deep hole is not directly equal to the distance from a point on the image obtained by the CCD camera to the center of the image; therefore, the distance from a point on the image obtained by the CCD camera to the center of the image needs to be substituted into the mathematical model established according to the optical path model to restore the distance from the inner surface of the deep hole to the axis of the deep hole, thereby representing the geometric features of the inner surface of the deep hole; The geometric features of the inner surface of the known deep hole are represented by the distance R from the inner surface of the deep hole to the axis of the deep hole, and the distance r from a point on the image obtained by the CCD camera to the center of the image is represented; in the process of restoring the distance from the inner surface of the deep hole to the axis of the deep hole, a polar coordinate system is established on the image obtained by the CCD camera with the center of the image as the polar coordinate center. According to the value of the independent variable θ, in the process of the independent variable θ gradually increasing from 0 to 2π, it is easy to obtain the value of the distance r from a point on the image obtained by the CCD camera to the center of the image in the interval [0,2π]. The mathematical model is expressed as: In this mathematical model, the independent variable r is the value obtained from the CCD camera, PT is the emission radius of the ring laser, IV is the sum of the widths of the imaging cone and the reflective cone, UV is the width of the reflective cone, and EU is the top surface radius of the reflective cone. The distance r from a point on the image obtained by the CCD camera to the center of the image is substituted into the above mathematical model, and the image is restored according to the designed drawing process, and finally the value of the distance R from the inner surface of the deep hole to the axis of the deep hole is obtained, and the measurement is finally completed.

Images