Method and device for calibrating coaxiality between multiple infrared light paths based on machine vision

Abstract

This invention discloses a machine vision-based method for calibrating the coaxiality between multiple infrared optical paths, comprising the following steps: the object being calibrated emits multiple infrared optical paths; an infrared lens receives the infrared light waves from the multiple infrared optical paths; a detector senses the infrared light waves and performs photoelectric conversion to form an electrical signal; an image acquisition card acquires the electrical signal, outputs a digital image signal, and sends it to a computer; the computer performs image processing, calculates image sharpness, controls the infrared lens to focus, obtains the spot image of the infrared optical path, calculates the distance between the center coordinates of the spot, and obtains the coaxiality between the infrared optical paths. This invention solves the problems of low calibration efficiency and mismatched calibration bands between multiple infrared optical paths, and adopts a machine vision-based method for calibrating the coaxiality between multiple infrared optical paths, realizing intelligent and automated calibration of the coaxiality of multiple infrared optical paths.

Description

Technical Field

[0001] This invention relates to the field of infrared guidance technology, and more specifically, to a method and apparatus for calibrating the coaxiality between multiple infrared optical paths. Background Technology

[0002] In the field of infrared guidance, infrared guidance is a widely used guidance method. It has advantages such as being unaffected by radio interference, high guidance accuracy, ability to operate day and night, good attack concealment, simple structure, and reliable operation. These advantages make infrared guidance a promising application in various applications.

[0003] With the continuous development of infrared guidance technology, infrared anti-jamming technology is also developing accordingly. Semi-physical simulation technology and simulation devices for infrared guidance anti-jamming are also emerging and being used to test the anti-jamming performance of infrared guidance. In particular, the simulation technology of multiple infrared sources simulating multiple infrared decoys is more in line with the infrared guidance scenario in actual combat.

[0004] Therefore, multiple infrared interference sources are used in the simulation device. To ensure the initial simulation state of multiple infrared light sources, it is necessary to calibrate the coaxiality between multiple infrared optical paths. Existing calibration techniques use a theodolite to manually aim multiple infrared optical paths to measure coaxiality, which has drawbacks such as low efficiency and the calibration results being affected by subjectivity. Therefore, there is an urgent need to develop intelligent and automated coaxiality calibration methods.

[0005] Existing technology uses prisms and theodolites to calibrate the coaxiality of a portion of the visible light band in the infrared optical path using visible light. The calibration method is as follows: First, parallax is measured using a pentaprism and theodolite. Since different light sources have different radiation directions, the horizontal and vertical angles of the aperture centers of each light source are measured using the theodolite, and the relative angles between the centers of different apertures are calculated.

[0006] The existing technology has the following main drawbacks: the wavelength measured during the calibration process is the visible light band, which differs from the infrared band used in actual applications, resulting in inaccurate calibration results; the calibration process involves optical path debugging and alignment, which takes a long time; the calibration results are greatly affected by human factors, and the calibration process is highly subjective. Summary of the Invention

[0007] The purpose of this invention is to solve the problems of low coaxiality calibration efficiency and mismatch of calibration bands between multiple infrared optical paths, and to provide a calibration method and device for coaxiality between multiple infrared optical paths based on machine vision, thereby realizing intelligent and automated calibration of coaxiality of multiple infrared optical paths.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0009] A machine vision-based method for calibrating coaxiality among multiple infrared optical paths includes the following steps:

[0010] The object being studied emits multiple infrared light paths, and the infrared lens receives the infrared light waves from these multiple infrared light paths;

[0011] The detector senses infrared light waves and converts them into electrical signals through photoelectric conversion.

[0012] The image acquisition card acquires electrical signals, outputs digital image signals, and sends them to the computer;

[0013] The computer performs image processing, calculates image sharpness, controls the infrared lens to focus, obtains the spot image of the infrared light path, calculates the distance between the center coordinates of the spot, and obtains the coaxiality between the infrared light paths.

[0014] Furthermore, the formula for calculating the distance between the center coordinates of the light spot is:

[0015]

[0016] In the formula, x i x represents the x-coordinate of the centroid pixel of each auxiliary light source; x0 represents the x-coordinate of the centroid pixel of the main light source; y i y0 represents the centroid pixel coordinate of each auxiliary light source; y0 represents the centroid pixel coordinate of the main light source.

[0017] Furthermore, the formula for calculating the coaxiality between the optical paths of each light source is:

[0018]

[0019] In the formula, δ represents the coaxiality of the multiple infrared optical paths; f represents the focal length of the infrared lens.

[0020] Furthermore, by constructing an image sharpness evaluation function, image sharpness is calculated. The sharpness evaluation function is expressed as:

[0021]

[0022] In the formula, W is the effective area of ​​the image spot; g(p,q) is the gray value at pixel (p,q); and (p0,q0) is the coordinate of the pixel where the center of the image spot is located.

[0023] Furthermore, focusing is achieved by controlling the movement of the infrared lens through an autofocus mechanism. The autofocus mechanism includes a motor and a linear displacement stage. The linear displacement stage uses a ball screw, on which a slider is mounted. The ball screw and the slider are connected by a thread. The motor drives the ball screw to rotate, and the rotation of the ball screw causes the slider to move linearly. The focusing lens of the infrared lens is mounted on the slider, and the movement of the slider causes the focusing lens to move synchronously.

[0024] A machine vision-based calibration device for coaxiality between multiple infrared optical paths, employing any one of the aforementioned machine vision-based calibration methods for coaxiality between multiple infrared optical paths, includes:

[0025] Infrared lens, detector, image acquisition card, computer and autofocus mechanism are used to calibrate the coaxiality of multiple infrared light paths emitted by the object being calibrated, and the multiple infrared light paths have at least 3 light paths.

[0026] The object being calibrated includes a multi-infrared light path generating device, which includes multiple light sources with a main light source and auxiliary light sources. Each light source can be independently controlled in terms of its on / off state and luminous intensity.

[0027] An infrared lens is used to receive infrared light waves from an infrared optical path, a detector is used to sense infrared light waves and perform photoelectric conversion to form an electrical signal, and an image acquisition card is used to acquire the electrical signal and output digital image signals to a computer.

[0028] The computer is used for image processing, calculating image sharpness, controlling the autofocus mechanism to drive the infrared lens to focus, obtaining the spot image of the infrared light path, calculating the distance between the center coordinates of the spot, and obtaining the coaxiality between the infrared light paths.

[0029] Furthermore, the infrared lens is a telephoto infrared lens, employing a folding optical path to increase the focal length.

[0030] Furthermore, the light source is a blackbody radiation source, and the detector uses a precision radiation thermometer with laser aiming function to measure the blackbody radiation temperature.

[0031] Furthermore, the light source is a bulb radiation source, and the detector uses a brightness-type color temperature meter to measure the infrared radiation color temperature of the bulb radiation source.

[0032] Furthermore, the automatic focusing mechanism includes a motor and a linear displacement stage, which work together to drive the infrared lens to move for focusing.

[0033] Compared with existing technologies, this invention adopts a machine vision-based method for calibrating the coaxiality between multiple infrared optical paths, thereby calibrating the coaxiality between multiple light source infrared optical paths and realizing intelligent and automated calibration of the coaxiality of multiple light source infrared optical paths. Attached Figure Description

[0034] Figure 1 This is a flowchart illustrating a machine vision-based method for calibrating the coaxiality between multiple infrared optical paths.

[0035] Figure 2 This is a schematic diagram showing the distance between the center coordinates of the light spot image.

[0036] Figure 3 This is a schematic diagram of a calibration device for the coaxiality between multiple infrared optical paths based on machine vision.

[0037] Figure 4 This is a schematic diagram of the optical path of an infrared lens.

[0038] Figure 5 This is a schematic diagram of the automatic focusing mechanism.

[0039] Figure 6 This is a schematic diagram of the digital image processing workflow. Detailed Implementation

[0040] The following description, in conjunction with the accompanying drawings and specific embodiments, further illustrates the calibration method and apparatus for coaxiality between multiple infrared optical paths based on machine vision.

[0041] Please see Figure 1 This invention discloses a method for calibrating the coaxiality between multiple infrared optical paths based on machine vision, comprising the following steps:

[0042] The object being studied emits multiple infrared light paths, and the infrared lens receives the infrared light waves from these multiple infrared light paths;

[0043] The detector senses infrared light waves and converts them into electrical signals through photoelectric conversion.

[0044] The image acquisition card acquires electrical signals, outputs digital image signals, and sends them to the computer;

[0045] The computer performs image processing, calculates image sharpness, controls the infrared lens to focus, obtains the spot image of the infrared light path, calculates the distance between the center coordinates of the spot, and obtains the coaxiality between the infrared light paths.

[0046] Please see Figure 2 Using the optical paths of three infrared light sources as the measurement objects, three infrared light spots are obtained. The distance between the centers of the light spots of infrared light source 2 and infrared light source 3 is L. The coaxiality between the optical paths is obtained from the relationship between pixel coordinates and world coordinates.

[0047] The formula for calculating the distance between the center coordinates of the light spot is:

[0048]

[0049] In the formula, x i x represents the x-coordinate of the centroid pixel of each auxiliary light source; x0 represents the x-coordinate of the centroid pixel of the main light source; y i y0 represents the centroid pixel coordinate of each auxiliary light source; y0 represents the centroid pixel coordinate of the main light source.

[0050] The formula for calculating the coaxiality between the optical paths of each light source is:

[0051]

[0052] In the formula, δ represents the coaxiality of the multiple infrared optical paths; f represents the focal length of the infrared lens.

[0053] Image sharpness is calculated by constructing an image sharpness evaluation function, which is expressed as follows:

[0054]

[0055] In the formula, W is the effective area of ​​the image spot; g(p,q) is the gray value at pixel (p,q); and (p0,q0) is the coordinate of the pixel where the center of the image spot is located.

[0056] The sharpness evaluation function uses the squared distance from each point to the center of the image spot as a weighting term, multiplies it by the gray value of that point, sums them up as the numerator, and the denominator is the sum of all gray values ​​in the region.

[0057] This invention achieves focusing by controlling an automatic focusing mechanism to move an infrared lens. The automatic focusing mechanism includes a motor and a linear displacement stage. The linear displacement stage uses a ball screw, and a slider is mounted on the ball screw. The ball screw and the slider are connected by a thread. The motor drives the ball screw to rotate, and the rotation of the ball screw causes the slider to move linearly. The focusing lens of the infrared lens is mounted on the slider, and the movement of the slider causes the focusing lens to move synchronously.

[0058] Please see Figure 3 The present invention also discloses a calibration device for coaxiality between multiple infrared optical paths based on machine vision, which applies the calibration method for coaxiality between multiple infrared optical paths based on machine vision described in any of the above claims, including:

[0059] An infrared lens, an infrared focal plane detector, an image acquisition card, a computer, and an autofocus mechanism are used to calibrate the coaxiality of multiple infrared optical paths emitted by the object being calibrated. The multiple infrared optical paths have at least three optical paths.

[0060] The objects to be calibrated include a multi-infrared light path generator and an output lens. The multi-infrared light path generator includes multiple light sources with a main light source and an auxiliary light source. Each light source can be independently controlled in terms of its on / off state and luminous intensity.

[0061] An infrared lens is used to receive infrared light waves from an infrared optical path, an infrared focal plane detector is used to sense infrared light waves and perform photoelectric conversion to form an electrical signal, and an image acquisition card is used to acquire the electrical signal and output digital image signals to a computer.

[0062] The computer is used for image processing, calculating image sharpness, controlling the autofocus mechanism to drive the infrared lens to focus, obtaining the spot image of the infrared light path, calculating the distance between the center coordinates of the spot, and obtaining the coaxiality between the infrared light paths.

[0063] As a device that receives infrared light waves from an infrared optical path, the infrared lens plays a role in the propagation and focusing of infrared light. Key parameters of the infrared lens, such as focal length, imaging resolution, and optical transfer function (MTF), have a significant impact on the final imaging effect of the optical path. Based on the relationship between the imaging resolution, focal length, and image pixel size of the optical path, the relationship between angular resolution θ, focal length f, and image pixel size d is obtained as follows:

[0064]

[0065] As shown in the above formula, when the pixel size is constant, the larger the focal length, the higher the angular resolution. To achieve high-precision coaxiality measurement, this invention employs a long-focal-length infrared lens and a folding optical path to increase the focal length and improve resolution. A schematic diagram of the folding optical path is shown below. Figure 4 As shown.

[0066] Infrared focal plane detectors, as sensing components, detect infrared light waves and convert them into electrical signals through photoelectric conversion. Different types or models of infrared focal plane detectors are sensitive to different infrared bands, thus requiring selection of the appropriate detector. Multi-infrared light path generators generally use two types of light sources: blackbody and bulb.

[0067] The light source is a blackbody radiation source. According to Planck's law of blackbody radiation, the radiant exitance (radiant power per unit area) of a blackbody at a certain temperature at a specific wavelength, also called spectral radiant exitance, is expressed by the formula:

[0068]

[0069] In the formula, I is the spectral radiative exitance; v is the frequency; h is Planck's constant; c is the speed of light; k is Boltzmann's constant; T is the blackbody temperature; and e is the base of the natural logarithm.

[0070] Substituting the blackbody temperature into the above equation, we obtain the relationship between spectral radiative exitance and wavelength. Plotting the graph, we find the wavelength corresponding to the peak of spectral radiative exitance and take that wavelength as the center wavelength.

[0071] Blackbody temperature is measured using a precision radiation thermometer with laser aiming capability. The thermometer is aimed at the bottom of the blackbody radiation source cavity, and the blackbody temperature is measured. Three consecutive measurements are taken, and the average value is used as the blackbody temperature test result. The blackbody temperature measurement result is then substituted into Planck's formula to find the corresponding center wavelength, which is used as the basis for selecting an infrared focal plane array detector.

[0072] The light source is a bulb radiation source, and a luminance-type color temperature meter is used to measure the infrared color temperature of the bulb radiation source.

[0073] Measurement method: Adjust the filament plane, aperture diameter, and standard white board of the bulb radiation source to be on the same horizontal axis. The photometric axis of the brightness-type color temperature meter should be on the same plane and at a 45° angle to the axis of the light track. Turn on the bulb radiation source and preheat the color temperature meter. After preheating, read the value displayed on the color temperature meter.

[0074] Subtract 273.15 from the color temperature measurement result to obtain the corresponding temperature value. Substitute this value into Planck's formula to obtain the relationship between spectral radiant exitance and wavelength. Plot the relationship and find the wavelength corresponding to the peak spectral radiant exitance. Use this wavelength as the center wavelength. Select an infrared focal plane detector based on the center wavelength.

[0075] like Figure 5 As shown, the autofocus mechanism includes a motor and a linear displacement stage. The motor and the linear displacement stage work together to drive the infrared lens to move and focus. The linear displacement stage uses a ball screw, on which a slider is mounted. The ball screw and the slider are connected by a thread. The motor drives the ball screw to rotate, and the rotation of the ball screw causes the slider to move linearly. The focusing lens of the infrared lens is mounted on the slider, and the movement of the slider causes the focusing lens to move synchronously.

[0076] The motor receives control signals from the computer and drives the focusing lens in the infrared lens to move horizontally, thereby adjusting the focal length of the infrared lens. The computer displays the digital image signal output from the image acquisition card, drives the motor to move the infrared lens for focusing via the RS232 interface, calculates a sharpness evaluation function on the image signal, and obtains the focus position from the calculation result.

[0077] The focusing process is about finding the sharpest image. This is achieved by constructing an image sharpness evaluation function to calculate image sharpness. A series of images are acquired during the motor's movement, and the grayscale value of each image is input into the sharpness evaluation function to calculate its sharpness. The extreme value that minimizes the sharpness evaluation function is then found, and the position corresponding to this minimum value is the ideal position after focusing.

[0078] Align the infrared lens of the calibration device with the exit lens of the object being calibrated, and adjust the installation height and pitch angle of the calibration device so that the center of the infrared lens of the calibration device is aligned with the center of the exit lens of the object being calibrated, the pitch angle is 0, and the calibration device is horizontal.

[0079] Connect the cables to the calibration device, turn on the power switches of the light sources of the multi-infrared light path generator in sequence, acquire and display the light spot images on the computer, perform circle fitting operation, obtain the centroid coordinates of the light spot images of each light source through the centroid algorithm, and calculate the coaxiality between the light paths of each light source.

[0080] The purpose of digital image processing on the acquired infrared images is to remove stray signals that interfere with the measurement results. The coordinates of the image spot center are found using a data processing algorithm based on image grayscale values. The digital image processing flow is as follows: Figure 6 As shown, the infrared spot image undergoes preprocessing, image segmentation, and circle fitting to obtain the centroid coordinates of the spot. The coaxiality between multiple infrared optical paths is calculated by determining the center distance of the spot. This invention solves the problems of low efficiency and mismatched calibration bands in coaxiality calibration between multi-source infrared optical paths, achieving intelligent and automated calibration of the coaxiality of multi-source infrared optical paths.

[0081] The above description is a detailed description of the preferred embodiments of the present invention. However, the embodiments are not intended to limit the scope of the patent application of the present invention. All equivalent changes or modifications made under the technical spirit disclosed in the present invention should fall within the patent scope covered by the present invention.

Claims

1. A method for calibrating coaxiality among multiple infrared optical paths based on machine vision, characterized in that, Includes the following steps: The object being studied emits multiple infrared light paths, and the infrared lens receives the infrared light waves from these multiple infrared light paths; The detector senses infrared light waves and converts them into electrical signals through photoelectric conversion. The image acquisition card acquires electrical signals, outputs digital image signals, and sends them to the computer; The computer performs image processing, calculates image sharpness, controls the infrared lens to focus, obtains the spot image of the infrared light path, calculates the distance between the center coordinates of the spot, and obtains the coaxiality between the infrared light paths.

2. The calibration method for coaxiality between multiple infrared optical paths based on machine vision according to claim 1, characterized in that, The formula for calculating the distance between the center coordinates of the light spot is: In the formula, x i x represents the x-coordinate of the centroid pixel of each auxiliary light source; x0 represents the x-coordinate of the centroid pixel of the main light source; y i The ordinate of the centroid pixel for each auxiliary light source; y0 represents the centroid pixel coordinate of the main light source.

3. The calibration method for coaxiality between multiple infrared optical paths based on machine vision according to claim 1, characterized in that, The formula for calculating the coaxiality between the optical paths of each light source is: In the formula, δ represents the coaxiality of the multiple infrared optical paths; f represents the focal length of the infrared lens.

4. The calibration method for coaxiality between multiple infrared optical paths based on machine vision according to claim 1, characterized in that, Image sharpness is calculated by constructing an image sharpness evaluation function, which is expressed as follows: In the formula, W is the effective area of ​​the image spot; g(p,q) is the gray value at pixel (p,q); and (p0,q0) is the coordinate of the pixel where the center of the image spot is located.

5. The calibration method for coaxiality between multiple infrared optical paths based on machine vision according to claim 1, characterized in that, Focusing is achieved by controlling the movement of the infrared lens through an autofocus mechanism. The autofocus mechanism includes a motor and a linear displacement stage. The linear displacement stage uses a ball screw, and a slider is installed on the ball screw. The ball screw and the slider are connected by a thread. The motor drives the ball screw to rotate, and the rotation of the ball screw causes the slider to move linearly. The focusing lens of the infrared lens is installed on the slider, and the movement of the slider causes the focusing lens to move synchronously.

6. A calibration device for coaxiality between multiple infrared optical paths based on machine vision, employing the calibration method for coaxiality between multiple infrared optical paths based on machine vision as described in any one of claims 1 to 5, characterized in that, include: Infrared lens, detector, image acquisition card, computer and autofocus mechanism are used to calibrate the coaxiality of multiple infrared light paths emitted by the object being calibrated, and the multiple infrared light paths have at least 3 light paths. The object being calibrated includes a multi-infrared light path generating device, which includes multiple light sources with a main light source and auxiliary light sources. Each light source can be independently controlled in terms of its on / off state and luminous intensity. An infrared lens is used to receive infrared light waves from an infrared optical path, a detector is used to sense infrared light waves and perform photoelectric conversion to form an electrical signal, and an image acquisition card is used to acquire the electrical signal and output digital image signals to a computer. The computer is used for image processing, calculating image sharpness, controlling the autofocus mechanism to drive the infrared lens to focus, obtaining the spot image of the infrared light path, calculating the distance between the center coordinates of the spot, and obtaining the coaxiality between the infrared light paths.

7. The calibration device for coaxiality between multiple infrared optical paths based on machine vision according to claim 6, characterized in that, The infrared lens is a telephoto infrared lens, which uses a folding optical path to increase the focal length.

8. The calibration device for coaxiality between multiple infrared optical paths based on machine vision according to claim 6, characterized in that, The light source is a blackbody radiation source, and the detector uses a precision radiation thermometer with laser aiming function to measure the blackbody radiation temperature.

9. The calibration device for coaxiality between multiple infrared optical paths based on machine vision according to claim 6, characterized in that, The light source is a bulb radiation source, and the detector uses a brightness-type color temperature meter to measure the infrared radiation color temperature of the bulb radiation source.

10. The calibration device for coaxiality between multiple infrared optical paths based on machine vision according to claim 6, characterized in that, The autofocus mechanism includes a motor and a linear displacement stage. The motor and the linear displacement stage work together to drive the infrared lens to move and focus.

Images