A Dual-Line-of-Sight Spot Centroid Calculation Method Based on Center-of-Gravity Shift

Abstract

The present invention relates to a method for calculating the centroid of double-line-of-sight spots based on centroid offset. By utilizing the information characteristics of the spot images of a double-line-of-sight Hartmann wavefront sensor and adopting an iterative centroid calculation with a variable gate, the centroid positions of the sub-spots corresponding to the two lines of sight within the sub-apertures can be respectively located. Different from the commonly used correlation algorithms or region division methods in the existing double-line-of-sight centroid calculation methods, the present invention can achieve the separate positioning of two groups of sub-spots by using a simple centroid calculation method, which has an obvious advantage in terms of computational amount and can provide an efficient technical route for wavefront reconstruction of a double-line-of-sight Hartmann wavefront sensor under prior conditions.

Description

Technical Field

[0001] The present invention belongs to the technical field of optical information measurement, and relates to a method for reconstructing the wavefronts of two beacon beams simultaneously measured by a Hartmann wavefront sensor, and particularly relates to a method for calculating the centroid of double-line-of-sight light spots based on centroid offset. Background Art

[0002] As a simple and efficient device for measuring the wavefront phase of a light beam, the Hartmann wavefront sensor has been widely applied in the fields of optical detection, laser beam purification, laser communication, astronomical observation, etc. The classical Hartmann wavefront sensor mainly consists of a microlens array and a photodetector. In recent years, in the face of continuously expanding application fields and scenarios, the detection object of the Hartmann wavefront sensor is no longer limited to the wavefront distortion of traditional single-point source targets. To achieve a breakthrough in the wavefront detection mechanism, the reconstruction of wide-field multi-line-of-sight wavefront information for multiple beacons and multiple targets has become a research hotspot in the field of wavefront sensing.

[0003] In 2010, Anne Costille et al. published an article on the closed-loop tomographic control of the MCAO system. In the article, a wide-field HS wavefront sensor with a 7×7 aperture segmentation was used to simultaneously detect three guide stars and one observation target, and a multi-object beam co-diffraction array imaging scheme was described [Anne Costille et al., Widefield adaptive optics laboratory demonstration with closed-loop tomographic control, J. Opt. Sci. Am. A 27(3):469-483, 2010]. The wavefront reconstruction strategy in the article re-segmented the sub-aperture images according to different lines of sight, segmented the multi-spot images into single-spot images of three small regions according to prior conditions, and then used the classical centroid algorithm and wavefront reconstruction algorithm to obtain wavefront information. In 2015, Lebao Yang et al. introduced a multi-object HS wavefront sensor for wide-field observation of the retina [Lebao Yang et al., Multiple-object Shack–Hartmann wavefront sensor design for a wide field of view on the retina, Chinese Opt. Lett. 13(12):120801-1-5, 2015], which also adopted the scheme of multi-line-of-sight beams entering a single wavefront sensor for imaging. This method limited multiple sub-spots within a sub-aperture to fall in different fixed regions, took the regular arrangement of multi-spots as the premise of wavefront reconstruction, and the wavefront reconstruction algorithm performed sub-region segmentation processing on the sub-aperture according to the limited conditions, and then restored the wavefront with traditional algorithms. From the principles of the above two schemes, it can be seen that the existing multi-line-of-sight wavefront reconstruction methods rely on artificially set limiting conditions and prior information on the spot landing positions, and the algorithm cannot achieve autonomous multi-spot centroid positioning, which has serious limitations in engineering implementation and application. Therefore, it is necessary to develop a multi-spot centroid automatic positioning method that can truly adapt to the multi-line-of-sight wavefront detection scenario. Summary of the Invention

[0004] The technical problem to be solved by the present invention is: to overcome the defect of the need for human intervention or prior information in the prior art solution, and without using complex image matching methods, based on the premise assumption that it is a small probability event that the center of gravity of the double-line-of-sight spot array image falls exactly in the middle of the two, and on the premise of ensuring the calculation efficiency of the spot centroid, a simple centroid calculation method is adopted to provide a novel double-line-of-sight spot centroid calculation method based on the center-of-gravity offset. According to the characteristic information of the spot image of the double-line-of-sight Hartmann wavefront sensor itself, following a simple iterative and variable-gate centroid calculation process, the centroid positions of the sub-spots corresponding to the two lines of sight within the sub-aperture can be calculated, that is, the fast, accurate, and automatic positioning calculation of the double-line-of-sight spots can be realized.

[0005] The technical solution of the present invention is as follows:

[0006] A method for calculating the centroid of double-line-of-sight spot based on centroid offset, which is applied to the sub-aperture spot image collected by a Hartmann wavefront sensor. This method uses a variable window and iterative centroid operation to locate the centroid of the double-line-of-sight spot, and includes the following steps:

[0007] Step 1: Extract the current sub-aperture spot image, which has a sub-spot image of double-line-of-sight beam; use the centroid calculation formula to calculate the data centroid position in the current sub-aperture spot image, that is, the overall centroid of the double-line-of-sight sub-spot image;

[0008] Step 2: Take the overall centroid of the double-line-of-sight sub-spot image as the center, intercept a new image area from the sub-spot image for centroid calculation to obtain new centroid data, and the width of the new image area is less than that of the sub-spot image;

[0009] Step 3: Take the new centroid as the center, further reduce the width of the centroid calculation image area, intercept a smaller area for centroid calculation, and update the centroid position data;

[0010] Step 4: Repeat Step 3 until the width of the reduced centroid calculation image area is equal to the first-line-of-sight minimum box selection width set in advance, then calculate the centroid position data at the minimum box selection width as the centroid data of the first-line-of-sight sub-spot image in the current sub-aperture spot image;

[0011] Step 5: Set the image data within the current minimum box selection width in the sub-aperture spot image to zero, that is, remove the first-line-of-sight sub-spot image information in the sub-aperture spot image;

[0012] Step 6: Calculate the new centroid position after removing the first-line-of-sight sub-spot image information, and continuously reduce the width of the centroid calculation image area with this centroid position as the center, perform iterative centroid calculation, and update the centroid data;

[0013] Step 7: When the width of the centroid calculation image area in Step 6 is equal to the second-line-of-sight minimum box selection width set in advance, output the current centroid data as the centroid data of the second-line-of-sight sub-spot image in the current sub-aperture spot image;

[0014] Step 8: Replace the current sub-aperture spot image, execute the centroid calculation process described in Steps 1 to 7 until every valid sub-aperture spot image collected by the Hartmann wavefront sensor is traversed, then the double-line-of-sight spot centroid information of each sub-aperture spot image can be obtained, and thus the centroid data of the spot array corresponding to each line of sight in the double-line-of-sight spot array can be obtained.

[0015] Further, the sub-aperture spot image is the spot image data after preprocessing to remove noise and interference signals.

[0016] Further, the centroid calculation formula adopts the method of calculating the centroid of image data, including the centroid algorithm or the weighted centroid algorithm.

[0017] Further, the preset minimum bounding width of the first line of sight in step 4 is greater than the pixel width of a single sub-spot of the first line of sight, that is, not less than the full width size of the designed micro-lens focused Airy spot.

[0018] Further, the preset minimum bounding width of the second line of sight in step 7 is greater than the pixel width of a single sub-spot of the second line of sight, that is, not less than the full width size of the designed micro-lens focused Airy spot.

[0019] The present invention breaks the conventional method that requires pre-determining the imaging regions and orientations of the double-line-of-sight spots in the sub-aperture images respectively, does not adopt complex image matching algorithms, and does not require hardware modification. By using the centroid calculation method with variable gates, the positioning of the double-line-of-sight spots can be completed. It has the characteristics of simple implementation and strong versatility, providing an effective centroid positioning algorithm for the Hartmann wavefront sensor to detect the wavefront of the double-line-of-sight beam, and can be directly applied to various wide-field double-line-of-sight Hartmann wavefront sensors.

[0020] Compared with the prior art, the present invention has the following advantages: Different from the commonly used correlation algorithms or region division methods in the existing double-line-of-sight centroid calculation methods, the simple centroid calculation method can be used to realize the separate positioning of the two groups of sub-spots, with obvious speed and efficiency advantages. It can provide an efficient technical route for the centroid calculation of the double-line-of-sight spots for the wavefront reconstruction of the double-line-of-sight Hartmann wavefront sensor under prior conditions, and has obvious application value for improving the wavefront measurement efficiency of the wide-field double-line-of-sight Hartmann wavefront sensor. Description of the Drawings

[0021] Figure 1 is the principle flow chart of the method implementation of the present invention;

[0022] Figure 2 is the detection schematic diagram of the wide-field double-line-of-sight Hartmann wavefront sensor in the embodiment of the present invention;

[0023] Figure 3 is the spot array image of the double-line-of-sight Hartmann wavefront sensor in the embodiment of the present invention;

[0024] Figure 4 is the centroid positioning result of the double-line-of-sight spot array image by the method of the present invention. Detailed Embodiments

[0025] The present invention will be further described below with reference to the drawings and embodiments.

[0026] Figure 1 This is the principle flow chart of the dual-line-of-sight spot centroid calculation method based on centroid offset according to the present invention. Figure 2 This is the working principle diagram of a typical wide-field Hartmann wavefront sensor for detecting the wavefront of dual-line-of-sight beams in the first embodiment of the present invention. As Figure 2 shown, the emitted beams of the beacon light source 1 incident on the sensor in the first line of sight and the beacon light source 2 incident on the sensor in the second line of sight respectively pass through the atmosphere 3 and then a part of them enters the field of view of the Hartmann wavefront sensor, and respectively form the first-line-of-sight beam 4 and the second-line-of-sight beam 5. After the dual-line-of-sight beams pass through the microlens array 6, they are simultaneously focused to form two sets of spot array focused images. The photosensitive surface of the photodetector 7 is located on the focal plane of the microlens array 6 to collect the spot array images of two different lines of sight. The number of effective microlenses in the microlens array 6 is 16×16, the shape of the incident beam pupil is a standard circle, and the size of each sub-aperture image is 32×32 pixels.

[0027] Figure 3 This is the spot array image collected by the photodetector 7 under the incidence of typical dual-line-of-sight beams. It can be seen from the figure that two light spots are formed by each microlens in the circular lighted area, corresponding to the two incident beams of sight respectively. Each square frame in the figure corresponds to a valid sub-aperture for wavefront reconstruction. Taking the first valid sub-aperture at the leftmost of the first row in the figure as the No. 1 valid sub-aperture, each valid sub-aperture is numbered in sequence from left to right and from top to bottom, up to the 188th valid sub-aperture at the rightmost of the bottom row, with a total of 188 valid sub-apertures. Next, the centroid of the dual-line-of-sight spots in each sub-aperture is calculated through the following steps in sequence.

[0028] Step 1: Extract the spot image of the No. 1 valid sub-aperture, with an image size of 32×32 pixels. Use the image threshold algorithm to remove the noise and interference signals in the sub-aperture image, and use the centroid calculation formula for the threshold-processed sub-aperture image to calculate the centroid position of the dual-line-of-sight sub-spot image data corresponding to the No. 1 valid sub-aperture, that is, the overall centroid of the dual-line-of-sight sub-spot image.

[0029] Step 2: Taking the overall centroid of the dual-line-of-sight sub-spot image obtained in Step 1 as the center, intercept a new image centroid calculation area from the sub-spot image, with the intercepted image size of 30×30, and calculate the new centroid data of the intercepted sub-spot image.

[0030] Step 3: Taking the new image centroid data as the center, reduce the width of the centroid calculation image area. The width of the image area is reduced by 2 pixels each time, and the equivalent half-width is decreased by 1 pixel each time. Intercept a smaller area for centroid calculation and update the centroid position data.

[0031] Step 4: Repeat Step 3 until the width of the reduced centroid calculation image area is equal to the pre-set minimum bounding width. In the first embodiment, the pre-set minimum width of the sub-spot image is 6 pixels, that is, the minimum centroid calculation image area for iteration stop is 6×6 pixels. At this time, the obtained centroid is marked as the centroid data of the first line-of-sight sub-spot of the first sub-aperture.

[0032] Step 5: Set the image data of 6×6 pixels within the current minimum bounding width in the sub-aperture spot image to zero, and remove the first line-of-sight sub-spot image information in the sub-aperture spot image.

[0033] Step 6: Calculate the centroid of the new 32×32 pixel sub-aperture spot image after removing the first line-of-sight sub-spot image information, and continuously reduce the width of the centroid calculation image area with the current image centroid as the center to perform centroid calculation and update the centroid data. The step size for reducing the width of the image area set in the first embodiment is 2 pixels.

[0034] Step 7: When the width of the centroid calculation image area in Step 6 is equal to the pre-set minimum bounding width, output the current centroid data as the centroid data of the second line-of-sight sub-spot of the first sub-aperture.

[0035] Step 8: Perform the centroid calculation process of Steps 1 to 7 on the sub-aperture images of the 2nd to 188th sub-apertures of the Hartmann wavefront sensor, and the centroid information of the double line-of-sight spots of each effective sub-aperture can be obtained, so as to obtain the centroid data of the spot array corresponding to each line of sight in the double line-of-sight spot array.

[0036] Typical centroid positioning results are as Figure 4 shown. In the figure, two spots are accurately positioned in each sub-aperture, and the centroid positions of the two spots are represented by circle and cross marks respectively. The centroid data marked by circles in all effective sub-apertures constitutes the centroid information of the first line-of-sight spot array, and the centroid data marked by cross marks in all effective sub-apertures constitutes the centroid information of the second line-of-sight spot array.

[0037] As described above, it is only the specific implementation manner in the present invention, but the protection scope of the present invention is not limited thereto. Any transformation or replacement that can be understood and conceived by those familiar with the technology within the technical scope disclosed by the present invention should be covered within the scope of the present invention.

Claims

1. A method for calculating the centroid of double - sight spot based on the centroid offset, which is applied to the sub - aperture spot image collected by a Hartmann wavefront sensor, is characterized in that This method uses a variable wavegate and iterative centroid operation to locate the centroid of the double-line-of-sight spot, and includes the following steps: Step 1: Extract the current sub-aperture spot image, where the sub-aperture spot image has sub-spot images of double-line-of-sight beams; use the centroid calculation formula to calculate the data centroid position in the current sub-aperture spot image, that is, the overall centroid of the double-line-of-sight sub-spot images; Step 2: Centered on the overall centroid of the double-line-of-sight sub-spot images, intercept a new image area from the sub-spot image for centroid calculation to obtain new centroid data, and the width of the new image area is smaller than that of the sub-spot image; Step 3: Centered on the new centroid, further reduce the width of the centroid calculation image area, intercept a smaller area for centroid calculation, and update the centroid position data; Step 4: Repeat Step 3 until the width of the reduced centroid calculation image area is equal to the first-line-of-sight minimum bounding width set in advance, then calculate the centroid position data at the minimum bounding width as the centroid data of the first-line-of-sight sub-spot image in the current sub-aperture spot image; Step 5: Set the image data within the current minimum bounding width in the sub-aperture spot image to zero, that is, remove the first-line-of-sight sub-spot image information in the sub-aperture spot image; Step 6: Calculate the new centroid position after removing the first-line-of-sight sub-spot image information, and continuously reduce the width of the centroid calculation image area centered on this centroid position, perform centroid calculation iteratively, and update the centroid data; Step 7: When the width of the centroid calculation image area in Step 6 is equal to the second-line-of-sight minimum bounding width set in advance, output the current centroid data as the centroid data of the second-line-of-sight sub-spot image in the current sub-aperture spot image; Step 8: Replace the current sub-aperture spot image and execute the centroid calculation process described in Steps 1 to 7 until each valid sub-aperture spot image collected by the Hartmann wavefront sensor is traversed, then the double-line-of-sight spot centroid information of each sub-aperture spot image can be obtained, and thus the centroid data of the spot array corresponding to each line of sight in the double-line-of-sight spot array can be obtained.

2. The method for calculating the centroid of a double-line-of-sight light spot based on centroid offset according to claim 1, wherein: The sub-aperture spot image is the spot image data after preprocessing to remove noise and interference signals.

3. A method for calculating the centroid of a double-line-of-sight light spot based on centroid offset according to claim 1, characterized in that: The centroid calculation formula uses the method of calculating the data centroid, including the centroid algorithm or the weighted centroid algorithm.

4. A method for calculating the centroid of a double-line-of-sight light spot based on centroid offset according to claim 1, characterized in that: The first-line-of-sight minimum bounding width set in advance in Step 4 is greater than the single sub-spot pixel width of the first line of sight, that is, not less than the full width size of the designed microlens focusing Airy spot.

5. A method for calculating the centroid of a double-line-of-sight light spot based on centroid offset according to claim 1, characterized in that: The second-line-of-sight minimum bounding width set in advance in Step 7 is greater than the single sub-spot pixel width of the second line of sight, that is, not less than the full width size of the designed microlens focusing Airy spot.

Images