Binocular passive optical positioning device

Abstract

The application relates to the technical field of optical positioning, and discloses a binocular passive optical positioning device, which comprises system hardware, camera calibration modules, light source dynamic control modules, image processing modules and light spot screening and matching modules. The system hardware comprises reflective marker points, a rack, cameras and light source assemblies. Two groups of the cameras are installed on the rack, and two groups of the light source assemblies are coaxially arranged with the shooting directions of the two groups of the cameras. The camera calibration modules are used for calibrating optical parameters and position parameters of the two groups of the cameras. The light source dynamic control modules are used for dynamically controlling the two groups of the light source assemblies. The image processing modules are used for processing images acquired by the two groups of the cameras to acquire processed images. The light spot screening and matching modules are used for screening light spots in the processed images, performing coordinate calculation on the screened light spots, performing stereoscopic matching on the screened light spots corresponding to the two groups of the cameras, and acquiring light spot matching results. The three-dimensional reconstruction modules are used for calculating accurate coordinates of each reflective marker point in a three-dimensional space according to the light spot matching results.

Description

Technical Field

[0001] This application relates to the technical field of optical positioning, and in particular to a binocular passive optical positioning device. Background Technology

[0002] Optical positioning devices mimic human binocular stereo vision by calculating the parallax of the same target point in the images of two cameras and using the principle of geometric triangulation to calculate its three-dimensional spatial coordinates. They play an important role in surgical navigation, industrial inspection, robot guidance, and motion capture. Passive optical positioning devices are characterized by their markers not requiring power, making the tracked objects lighter and more flexible.

[0003] In existing technologies, auxiliary light sources are used to adjust the illumination intensity within the field of view. The brightness adjustment range mainly relies on the prior calibration of the field of view. By adjusting the brightness of the light source, the illumination intensity within the field of view is made suitable, thereby ensuring the quality of the acquired image information and improving the accuracy of marker extraction. However, in actual optical positioning, the field of view is also affected by ambient light. At the same time, the marker has an optimal range of reflective brightness during the setting process. Therefore, it is difficult to achieve the optimal adjustment of the light source by relying solely on the brightness range calibrated by the field of view. Therefore, how to improve the accuracy of optical positioning by optimizing the brightness adjustment of the auxiliary light source is the fundamental problem that this invention aims to solve. Summary of the Invention

[0004] To improve the accuracy of optical positioning by optimizing the brightness adjustment of the auxiliary light source, this application provides a binocular passive optical positioning device, which adopts the following technical solution:

[0005] A binocular passive optical positioning device, comprising:

[0006] The system hardware includes reflective markers, a stand, cameras, and light source components. The two sets of cameras are mounted on the stand, and the two sets of light source components are respectively set coaxially with the shooting direction of the two sets of cameras.

[0007] The camera calibration module is used to calibrate the optical and positional parameters of the two sets of cameras.

[0008] The light source dynamic control module is used for dynamic control of the two sets of light source components;

[0009] The image processing module is used to process the images acquired by the two sets of cameras to obtain processed images;

[0010] The light spot filtering and matching module is used to filter light spots in the processed image, calculate the coordinates of the filtered light spots, perform stereo matching on the filtered light spots corresponding to the two sets of cameras, and obtain the light spot matching result.

[0011] The 3D reconstruction module calculates the precise coordinates of each reflective marker in 3D space based on the light spot matching results.

[0012] Optionally, the process by which the light source dynamic control module dynamically controls the two sets of light source components includes:

[0013] The optimal reflective brightness range of the reflective marker is obtained. The optimal scene brightness range is obtained by testing under the condition of turning off the ambient light. The preset brightness value is determined by combining the optimal reflective brightness range and the optimal scene brightness range.

[0014] Ambient light data is collected in real time, analyzed, and a brightness adjustment coefficient is obtained. The real-time control brightness value is determined based on the preset brightness value and the brightness adjustment coefficient.

[0015] Optionally, the process of determining the preset brightness value includes:

[0016] Determine if there is any overlap between the optimal reflective brightness range and the optimal scene brightness range:

[0017] If so, the center value of the intersection will be used as the preset brightness value;

[0018] If not, adjust the scene until the optimal reflective brightness range and the optimal scene brightness range intersect;

[0019] The process of acquiring ambient light data includes:

[0020] The illumination values ​​of each preset point outside the camera's field of view are obtained. The mean, maximum difference, and variance of the illumination values ​​of each preset point are obtained. The mean, maximum difference, and variance of the illumination values ​​are normalized and then weighted and summed to obtain the ambient light influence coefficient. The brightness adjustment coefficient is obtained according to the range of the ambient light influence coefficient. The brightness adjustment coefficient is greater than 1.

[0021] The process of determining the real-time control brightness value includes:

[0022] The real-time control brightness value to be selected is obtained by multiplying the preset brightness value and the brightness adjustment coefficient.

[0023] Determine whether the selected real-time control brightness value falls within both the optimal reflective brightness range and the optimal scene brightness range simultaneously:

[0024] If so, the real-time control brightness value to be selected will be used as the real-time control brightness value.

[0025] If not, the maximum value of the intersection of the optimal reflective brightness range and the optimal scene brightness range will be used as the real-time control brightness value.

[0026] Optionally, the image processing module processes the image by including:

[0027] The difference between the real-time control brightness value and the preset brightness value is obtained. The noise processing level is determined according to the range of the difference. The image is then denoised according to the noise processing level to enhance the contrast of the image and obtain the pre-processed image.

[0028] The preprocessed image is binarized according to a preset threshold to obtain a binary image;

[0029] The binary image is analyzed by connected component analysis, and each independent bright spot region is marked. The binary image marked with bright spot regions is used as the processed image.

[0030] Optionally, the process by which the spot filtering and matching module filters spots in the processed image includes:

[0031] Obtain the expected size range of the reflective markers, and then filter the light spots in the processed image based on the expected size range;

[0032] Calculate the circularity of each light spot, where circularity = 4 × π × spot area / spot perimeter², and perform secondary screening based on the circularity value of the light spot;

[0033] The expected brightness range of the light spot is calculated based on ambient light data, real-time control brightness value, and reflection data of reflective markers. The light spot is then filtered three times based on the expected brightness range to obtain the filtered light spot.

[0034] Optionally, the process of obtaining the expected brightness range of the light spot includes:

[0035] Obtain the refractive efficiency range of the reflective markers;

[0036] Calculate the minimum and maximum values ​​of the refractive brightness. Based on the numerical intervals where the minimum and maximum values ​​of the refractive brightness are located, obtain the expected minimum and maximum values ​​of the brightness respectively. The interval enclosed by the expected minimum and maximum values ​​of the brightness is taken as the expected brightness interval.

[0037] Wherein, the minimum refractive brightness = real-time controlled brightness value × maximum value of the refractive efficiency range + minimum illuminance value in ambient light data × minimum value of the refractive efficiency range;

[0038] Maximum refractive brightness = Real-time controlled brightness value × Maximum value of refractive efficiency range + Maximum illumination value in ambient light data × Maximum value of refractive efficiency range.

[0039] Optionally, when performing three rounds of screening, the relationship between the number of light spots falling within the expected brightness range and the number of reflective markers is determined:

[0040] If the number of light spots falling within the expected brightness range is equal to the number of reflective markers, then a stereo matching process is performed;

[0041] If the number of light spots falling within the expected brightness range is less than the number of reflective markers, then the algorithm is used to predict the position of the obscured reflective markers.

[0042] If the number of light spots falling within the expected brightness range is greater than the number of reflective markers, then the light spots are screened again based on the absolute value of the difference between the light spot brightness and the median value of the expected brightness range. The light spots are then eliminated in descending order of the absolute value of the difference between the light spot brightness and the median value of the expected brightness range until the number of light spots falling within the expected brightness range equals the number of reflective markers.

[0043] Optionally, the three-dimensional reconstruction module calculates the precise coordinates of each reflective marker point in three-dimensional space using triangulation.

[0044] In summary, this application includes at least one of the following beneficial technical effects:

[0045] This invention uses a dynamic control module to dynamically control two sets of light source components. It can select a more suitable brightness for adjustment by comprehensively considering the conditions within the field of view, the influence of ambient light data, and the state of reflective markers, thereby improving the accuracy of optical positioning. Through a three-stage screening process of the light spot, it can more accurately eliminate the influence of interfering light spots on optical positioning. Attached Figure Description

[0046] Figure 1 This is a logic diagram of the binocular passive optical positioning device in this invention.

[0047] Figure 2 This is a schematic diagram of the system hardware layout in the binocular passive optical positioning device of the present invention. Detailed Implementation

[0048] The embodiments of this application are described in detail below, and examples of the embodiments are shown in the accompanying drawings.

[0049] In the description of this specification, the references to "certain embodiments," "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples" refer to specific features, structures, materials, or characteristics described in connection with the described embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0050] This application discloses a binocular passive optical positioning device, referring to... Figure 1 The device includes system hardware, a camera calibration module, a light source dynamic control module, an image processing module, and a light spot selection and matching module, as shown in the reference. Figure 2 The system hardware describes the hardware implementation of the binocular passive optical positioning device in this embodiment, including a reflective marker, a stand, cameras, and light source components. Two sets of cameras are mounted on the stand, and two sets of light source components are set coaxially with the shooting directions of the two sets of cameras. The two sets of cameras can realize two modes: parallel shooting mode and convergent shooting mode. The light source components will adjust with the adjustment of the shooting angle. The reflective marker consists of a refractive lens and a lens coated with a high-reflectivity film. The reflected light basically coincides with the incident light, so the light entering the reflective marker will be reflected back along the original path.

[0051] In terms of the software of the binocular passive optical positioning device in this embodiment, the camera calibration module is used to calibrate the optical and positional parameters of the two sets of cameras. The optical parameters include focal length, principal point coordinates, lens distortion parameters, etc., and the positional parameters are the rotation matrix and translation vector of the right camera relative to the left camera, which the user will then use to calculate the position of the reflective marker. The light source dynamic control module is used to dynamically control the two sets of light source components. This process includes: obtaining the optimal reflective brightness range of the reflective marker; testing to obtain the optimal scene brightness range under the condition of turning off ambient light; determining the preset brightness value by combining the optimal reflective brightness range and the optimal scene brightness range; collecting ambient light data in real time; analyzing the ambient light data to obtain the brightness adjustment coefficient; and determining the real-time control brightness value based on the preset brightness value and the brightness adjustment coefficient. Through the above process, the more suitable brightness can be selected for adjustment by comprehensively considering the conditions within the field of view, the influence of ambient light data, and the state of the reflective marker, thereby improving the accuracy of optical positioning.

[0052] In addition, the image processing module is used to process the images acquired by the two sets of cameras to obtain processed images; the spot filtering and matching module is used to filter the spots in the processed images, calculate the coordinates of the filtered spots, and perform stereo matching on the filtered spots corresponding to the two sets of cameras to obtain spot matching results; the three-dimensional reconstruction module is used to calculate the precise coordinates of each reflective marker point in three-dimensional space based on the spot matching results. This process can be solved by triangulation, which will not be described in detail here.

[0053] In one embodiment, a process for determining a preset brightness value is provided, including: determining whether there is an intersection between the optimal reflective brightness range and the optimal scene brightness range. Under normal conditions, the optimal reflective brightness range and the optimal scene brightness range are relatively close, but due to differences in the state within different fields of view, there is a possibility that they do not intersect. In this case, the scene within the field of view needs to be adjusted until the optimal reflective brightness range and the optimal scene brightness range intersect, ensuring that the acquisition process of the reflective markers is not interfered with. If they do intersect, the center value of the intersection is taken as the preset brightness value. Clearly, the preset brightness value comprehensively considers the state of the field of view. Based on factors such as the state and reflective markers, this embodiment obtains the illumination values ​​of various preset points outside the camera's field of view. These preset points are adaptively selected according to the conditions of different fields of view. The mean, maximum difference, and variance of the illumination values ​​at each preset point are then obtained. When the mean, maximum difference, and variance of the illumination values ​​at the preset points are large, it indicates that there is significant ambient light interfering with the field of view. The mean, maximum difference, and variance of the illumination values ​​are normalized and then weighted and summed. The corresponding weights are set according to the test data. Therefore, the ambient light influence coefficient is obtained through this judgment. The influence of ambient light is assessed. Since ambient light interference necessitates increasing the brightness of the light source components to reduce interference, this embodiment establishes a correspondence between the ambient light influence coefficient and the brightness adjustment coefficient by testing the optimal brightness increase under different environmental conditions. The brightness adjustment coefficient is obtained based on the range of the ambient light influence coefficient. Because brightness needs to be increased, the adjustment coefficient is greater than 1. In determining the real-time control brightness value, the selected real-time control brightness value is first obtained by multiplying the preset brightness value and the brightness adjustment coefficient. It is then determined whether the selected real-time control brightness value falls within both the optimal reflection brightness range and the optimal scene brightness range. If yes, the selected real-time control brightness value is used as the real-time control brightness value. Therefore, the real-time control brightness value is determined by comprehensively considering ambient light factors, reflective marker factors, and field of view factors, ensuring optimal selection and improving the accuracy of optical positioning. If no, the maximum value of the intersection of the optimal reflection brightness range and the optimal scene brightness range is used as the real-time control brightness value. The obtained real-time control brightness value is the optimal value selected by comprehensively considering various factors, thus improving the accuracy of optical positioning.

[0054] In one embodiment, the image processing module processes the image by: obtaining the difference between the real-time controlled brightness value and the preset brightness value; determining the noise processing level based on the range of the difference; performing noise reduction processing on the image according to the noise processing level; enhancing the image contrast; and obtaining a pre-processed image. The pre-processed image is then binarized according to a preset threshold to obtain a binary image. The binary image is analyzed through connected component analysis to mark each independent bright spot region. The binary image with the marked bright spot regions is used as the processed image. During this process, the difference between the real-time controlled brightness value and the preset brightness value corresponds to different noise processing levels. This correspondence is set based on noise processing data under different brightness levels. Since higher or lower brightness levels result in more noise, and excessive or insufficient noise processing can affect the accuracy of the reflective marker positions, a suitable noise processing level is determined to adaptively improve the accuracy of the reflective marker positions. Then, by enhancing the image contrast, binarizing the pre-processed image, and performing connected component analysis, each independent bright spot region can be marked more accurately, facilitating subsequent screening.

[0055] In the process of filtering light spots in the processed image by the spot selection and matching module, the expected size range of the reflective markers is first obtained based on the test data of the reflective markers. Therefore, spots that are too large or too small do not belong to the light spots corresponding to the reflective markers. Thus, the light spots in the processed image are filtered according to the expected size range to achieve size filtering. After that, the circularity of each light spot is calculated, where circularity = 4 × π × spot area / spot perimeter². Since the light spots reflected by the reflective markers tend to be circular or are circular, and since a standard circle... The circularity is 1. Therefore, when the circularity of the light spot approaches 1, for non-circular shapes, since their perimeter is larger relative to their area, the calculated value will be less than 1. Thus, by filtering based on circularity, non-circular light spots can be filtered out, achieving a secondary filtering process based on the circularity value of the light spot. Finally, the expected brightness range of the light spot is calculated based on ambient light data, real-time control brightness values, and reflection data from reflective markers. A third filtering process is then performed based on the expected brightness range to obtain the filtered light spots. The process of obtaining the expected brightness range of the light spot includes: obtaining... The refractive efficiency range of the reflective marker is determined. The minimum and maximum refractive brightness values ​​are calculated. Based on the numerical ranges of the minimum and maximum refractive brightness values, the expected minimum and maximum brightness values ​​are obtained respectively. The range enclosed by the expected minimum and maximum brightness values ​​is taken as the expected brightness range. The correspondence between the refractive brightness range and the expected brightness is set by fitting test data under different refractive brightness values. The minimum refractive brightness value = real-time controlled brightness value × maximum value of the refractive efficiency range + minimum illumination value in ambient light data × minimum value of the refractive efficiency range; the maximum refractive brightness value = real-time controlled brightness value × maximum value of the refractive efficiency range + maximum illumination value in ambient light data × maximum value of the refractive efficiency range. Through this process, the range of reflected light from the reflective marker can be determined based on the real-time ambient light and real-time controlled brightness value. When the brightness range of the light spot is not within the expected brightness range (mainly considering light spots with brightness lower than the expected brightness range), it indicates that it is an interference point. Therefore, through three screening processes, the influence of interference light spots on optical positioning can be eliminated relatively accurately.

[0056] In addition, during the three-stage screening process, the relationship between the number of light spots falling within the expected brightness range and the number of reflective markers is determined: if the number of light spots falling within the expected brightness range equals the number of reflective markers, it means that all reflective markers have been identified, and therefore a stereo matching process is performed; if the number of light spots falling within the expected brightness range is less than the number of reflective markers, it means that some reflective markers are blocked, resulting in a lower number of light spots than reflective markers. In this case, an algorithm is used to predict the position of the blocked reflective markers, and the specific prediction algorithm is set according to the actual selection; if the number of light spots falling within the expected brightness range is greater than the number of reflective markers, since the three-stage screening process is quite rigorous, this state is less likely to occur. In this case, the absolute value of the difference between the light spot brightness and the median value of the expected brightness range is used for further screening. Light spots are eliminated in descending order of the absolute value of the difference between the light spot brightness and the median value of the expected brightness range until the number of light spots falling within the expected brightness range equals the number of reflective markers. Through this process, interfering light spots can be accurately eliminated in most cases, improving the accuracy of optical positioning.

[0057] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A binocular passive optical positioning device, characterized in that, include: The system hardware includes reflective markers, a stand, cameras, and light source components. The two sets of cameras are mounted on the stand, and the two sets of light source components are respectively set coaxially with the shooting direction of the two sets of cameras. The camera calibration module is used to calibrate the optical and positional parameters of the two sets of cameras. The light source dynamic control module is used for dynamic control of the two sets of light source components; The image processing module is used to process the images acquired by the two sets of cameras to obtain processed images; The light spot filtering and matching module is used to filter light spots in the processed image, calculate the coordinates of the filtered light spots, perform stereo matching on the filtered light spots corresponding to the two sets of cameras, and obtain the light spot matching result. The 3D reconstruction module calculates the precise coordinates of each reflective marker in 3D space based on the light spot matching results; The process by which the dynamic control module of the light source dynamically controls the two sets of light source components includes: The optimal reflective brightness range of the reflective marker is obtained. The optimal scene brightness range is obtained by testing under the condition of turning off the ambient light. The preset brightness value is determined by combining the optimal reflective brightness range and the optimal scene brightness range. Ambient light data is collected in real time, analyzed, and a brightness adjustment coefficient is obtained. The real-time control brightness value is determined based on the preset brightness value and the brightness adjustment coefficient. The process of determining the preset brightness value includes: Determine if there is any overlap between the optimal reflective brightness range and the optimal scene brightness range: If so, the center value of the intersection will be used as the preset brightness value; If not, adjust the scene until the optimal reflective brightness range and the optimal scene brightness range intersect; The process of acquiring ambient light data includes: The illumination values ​​of each preset point outside the camera's field of view are obtained. The mean, maximum difference, and variance of the illumination values ​​of each preset point are obtained. The mean, maximum difference, and variance of the illumination values ​​are normalized and then weighted and summed to obtain the ambient light influence coefficient. The brightness adjustment coefficient is obtained according to the range of the ambient light influence coefficient. The brightness adjustment coefficient is greater than 1. The process of determining the real-time control brightness value includes: The real-time control brightness value to be selected is obtained by multiplying the preset brightness value and the brightness adjustment coefficient. Determine whether the selected real-time control brightness value falls within both the optimal reflective brightness range and the optimal scene brightness range simultaneously: If so, the real-time control brightness value to be selected will be used as the real-time control brightness value. If not, the maximum value of the intersection of the optimal reflective brightness range and the optimal scene brightness range will be used as the real-time control brightness value.

2. The binocular passive optical positioning device according to claim 1, characterized in that, The image processing module processes images in the following ways: The difference between the real-time control brightness value and the preset brightness value is obtained. The noise processing level is determined according to the range of the difference. The image is then denoised according to the noise processing level to enhance the contrast of the image and obtain the pre-processed image. The preprocessed image is binarized according to a preset threshold to obtain a binary image; The binary image is analyzed by connected component analysis, and each independent bright spot region is marked. The binary image marked with bright spot regions is used as the processed image.

3. The binocular passive optical positioning device according to claim 2, characterized in that, The process by which the spot filtering and matching module filters spots in the processed image includes: Obtain the expected size range of the reflective markers, and then filter the light spots in the processed image based on the expected size range; Calculate the circularity of each light spot, where circularity = 4 × π × spot area / spot perimeter², and perform secondary screening based on the circularity value of the light spot; The expected brightness range of the light spot is calculated based on ambient light data, real-time control brightness value, and reflection data of reflective markers. The light spot is then filtered three times based on the expected brightness range to obtain the filtered light spot.

4. A binocular passive optical positioning device according to claim 3, characterized in that, The process of obtaining the expected brightness range of the light spot includes: Obtain the refractive efficiency range of the reflective markers; Calculate the minimum and maximum values ​​of the refractive brightness. Based on the numerical intervals where the minimum and maximum values ​​of the refractive brightness are located, obtain the expected minimum and maximum values ​​of the brightness respectively. The interval enclosed by the expected minimum and maximum values ​​of the brightness is taken as the expected brightness interval. Wherein, the minimum refractive brightness = real-time controlled brightness value × maximum value of the refractive efficiency range + minimum illuminance value in ambient light data × minimum value of the refractive efficiency range; Maximum refractive brightness = Real-time controlled brightness value × Maximum value of refractive efficiency range + Maximum illumination value in ambient light data × Maximum value of refractive efficiency range.

5. A binocular passive optical positioning device according to claim 3, characterized in that, When performing three rounds of screening, determine the relationship between the number of light spots falling within the expected brightness range and the number of reflective markers: If the number of light spots falling within the expected brightness range is equal to the number of reflective markers, then a stereo matching process is performed; If the number of light spots falling within the expected brightness range is less than the number of reflective markers, then the algorithm is used to predict the position of the obscured reflective markers. If the number of light spots falling within the expected brightness range is greater than the number of reflective markers, then the light spots are screened again based on the absolute value of the difference between the light spot brightness and the median value of the expected brightness range. The light spots are then eliminated in descending order of the absolute value of the difference between the light spot brightness and the median value of the expected brightness range until the number of light spots falling within the expected brightness range equals the number of reflective markers.

6. A binocular passive optical positioning device according to claim 1, characterized in that, The three-dimensional reconstruction module calculates the precise coordinates of each reflective marker point in three-dimensional space using triangulation.

Images