Optical Corner Detection Imaging Method and System Based on Angular Hilbert Transform

Abstract

This invention discloses an optical corner detection imaging method and system based on angle-to-Hilbert transform. Collimated light is used to illuminate the object under test. The beam passes through an optical imaging system and reaches an imaging sensor to obtain corner detection images of the object. A metasurface is provided at the Fourier surface of the optical imaging system. The phase distribution of the metasurface is set with a phase difference according to the azimuth angle transformation. This invention achieves phase transformation along the azimuth angle through the metasurface, realizing angle-to-Hilbert transform, and achieving corner detection imaging without relying on any digital processing algorithms. This enables a purely optical corner detection method based on photon processing, which is helpful for developing next-generation, more energy-efficient, more efficient, and faster information processing systems.

Description

Technical Field

[0001] This invention relates to the field of corner detection imaging, and in particular to an optical corner detection imaging method and system based on the angular Hilbert transform. Background Technology

[0002] Corner detection plays a crucial role in fields such as machine vision, enabling functions like image alignment, object recognition, and image alignment. Corner detection imaging can significantly compress image information, achieving rapid and efficient information extraction, and has potential advantages in motion recognition and tracking. However, existing corner detection methods rely on digital processing algorithms, such as Harris corner detection. This method digitally processes the image, moves a local window across the image, and judges whether there are significant changes in image grayscale. If the grayscale values ​​within the window change significantly, a corner is determined to exist in the area containing that window. Digital processing algorithms are often energy-intensive, and their computational speed is limited due to the slow electronic response. Photonics, due to its parallel characteristics and low energy consumption, can provide a new development path for next-generation high-speed, energy-efficient information technology; however, currently, there is no solution for corner detection imaging using optical information processing. Summary of the Invention

[0003] Purpose of the invention: The purpose of this invention is to provide an optical corner detection and imaging method and system based on the angular Hilbert transform, which can directly realize corner detection and imaging using micro-nano optical elements without any digital processing algorithms.

[0004] Technical solution: The optical corner detection imaging method based on angular Hilbert transform described in this invention includes the following steps:

[0005] The object under test is illuminated by collimated light. After passing through the optical imaging system, the light beam reaches the imaging sensor to obtain the corner detection image of the object under test.

[0006] The optical imaging system has micro-nano optical elements at its Fourier surface; the phase distribution of the micro-nano optical elements is set with a phase difference of 180° as the azimuth angle changes.

[0007] Furthermore, based on the size of the corner points in the object to be tested... Determine corner detection parameters ;

[0008] The phase distribution of the micro / nano optical element is as follows:

[0009]

[0010] Here, floor refers to the floor operation, and arctan refers to the arctangent operation. The coordinates of the plane in which the micro / nano optical element is located are two-dimensional spatial coordinates. .

[0011] Furthermore, based on the greatest common divisor of the sizes of all corner points in the object to be tested... Determine corner detection parameters ;

[0012] The phase distribution of the micro / nano optical element is as follows:

[0013]

[0014] Here, floor refers to the floor operation, and arctan refers to the arctangent operation. The coordinates of the plane in which the micro / nano optical element is located are two-dimensional spatial coordinates. .

[0015] Furthermore, a polarizing filter is provided behind the object to be tested, and a polarizing filter is provided in front of the imaging sensor.

[0016] Furthermore, the optical imaging system is a 4f imaging system or a single-lens imaging system.

[0017] Furthermore, the micro / nano optical element includes one of the following: a phase plate, a diffractive optical element (DOE), a freeform surface element, a spatial light modulator, and a metasurface;

[0018] The phase plate, diffractive optical element (DOE), or freeform surface element employs a phase modulation method based on optical path accumulation.

[0019] The spatial light modulator employs a phase modulation method based on liquid crystal modulation.

[0020] The metasurface employs one phase design method or a combination of several phase design methods, including geometric phase, propagation phase, resonance phase, and chiral phase.

[0021] The optical corner detection imaging system based on the angle-to-Hilbert transform described in this invention includes an object to be measured, an optical imaging system, micro-nano optical components, and an imaging sensor;

[0022] After collimated light illuminates the object under test, the light beam passes through the optical imaging system and reaches the imaging sensor to obtain corner detection imaging of the object under test;

[0023] The optical imaging system has micro-nano optical elements at its Fourier surface; the phase distribution of the micro-nano optical elements is set with a phase difference of 180° as the azimuth angle changes.

[0024] The electronic device of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is loaded into the processor, it implements the optical corner detection imaging method based on the angle-to-Hilbert transform.

[0025] The computer-readable storage medium of the present invention stores a computer program, which, when executed by a processor, implements the optical corner detection imaging method based on the angle-to-Hilbert transform.

[0026] The computer program product of the present invention includes a computer program that, when executed by a processor, implements the optical corner detection and imaging method based on angular Hilbert transform.

[0027] Beneficial effects: Compared with the prior art, the advantages of the present invention are as follows:

[0028] 1) This invention proposes the concept of angle-to-Hilbert transform and uses optical means to realize corner detection imaging. This method of corner detection based on optical simulation calculation fills the technical gap of optical corner detection.

[0029] 2) This invention is based on the analog computing paradigm of photons and utilizes the parallelism and ultra-low power consumption characteristics of photons to perform wavefront manipulation. Compared with traditional electronic digital signal processing, it has higher energy efficiency and processing speed.

[0030] 3) The corner detection paradigm based on angular Hilbert transform of the present invention is compatible with both amplitude and phase-type objects, and can detect multiple corners in a single shot.

[0031] 4) The corner detection method based on metasurface in this invention has wide-field and broadband imaging capabilities and is applicable to most imaging scenarios, including microscopy and photography.

[0032] 5) This invention ensures that the device is small, lightweight, and simple in structure, making it easy to integrate with existing imaging systems. Furthermore, the micro-nano optical elements in the embodiments of this invention utilize metasurfaces, which have a thinner profile, further facilitating miniaturization and large-scale fabrication, thereby reducing device costs. Attached Figure Description

[0033] Figure 1 This is a flowchart of the optical corner detection imaging method according to an embodiment of the present invention.

[0034] Figure 2 This is a schematic diagram of the phase distribution of the metasurface in an embodiment of the present invention.

[0035] Figure 3 This is a schematic diagram of the phase distribution of the actual processed metasurface according to an embodiment of the present invention.

[0036] Figure 4 This is a schematic diagram of a corner detection imaging device according to an embodiment of the present invention.

[0037] Figure 5 This is a schematic diagram of another corner detection imaging device according to an embodiment of the present invention.

[0038] Figure 6 This is a schematic diagram of the corner detection imaging results of different objects at different wavelengths according to an embodiment of the present invention.

[0039] Figure 7 This is a schematic diagram of the corner detection imaging results of amplitude and phase objects according to an embodiment of the present invention. Detailed Implementation

[0040] This invention proposes the concept of angle-to-Hilbert transform and utilizes metasurfaces to achieve optical corner detection imaging. The optical corner detection imaging method based on angle-to-Hilbert transform includes the following steps: illuminating the object under test with collimated light; the beam passes through an optical imaging system and reaches an imaging sensor to obtain a corner detection image of the object under test; micro / nano optical elements are disposed at the Fourier surface of the optical imaging system; the phase distribution of the micro / nano optical elements is adjusted according to the azimuth angle. The phase difference. This method requires no digital processing algorithms and can directly achieve corner detection imaging using light propagation, making it faster and more energy-efficient.

[0041] Specifically, micro-nano optical elements include phase plates, diffractive optical elements (DOE), freeform surface elements, spatial light modulators, metasurfaces, and other elements that can achieve the above-mentioned phase modulation requirements. In this embodiment, metasurfaces are used as an example for detailed description.

[0042] Specifically, this invention can perform independent and simultaneous detection of corner points in an object under test, with the object showing significant light intensity only at the corner points. Independent detection refers to detecting only a single corner point of a specific size in the object under test, while simultaneous detection refers to detecting multiple corner points of different sizes in the object under test simultaneously. Based on the size of the corner points, appropriate corner detection parameters are determined. , A smaller size allows for the detection of more types of corner points and offers higher angular resolution, but lower spatial resolution. Therefore, it is necessary to select an appropriate size based on the application. The following sections will introduce them separately:

[0043] (a) Independent detection: The feature corner points in the object under test are all the same size, and are all... Or, the size of the feature corner points in the object being measured may differ, but only for those with a size of [missing information]. The corner points are detected and imaged, and the corner detection parameters are set as follows: . , The range is .

[0044] (b) Mixed detection: The size of the feature corner points in the object to be tested includes , , Wait, the corner point size that this invention can detect is , It is a positive integer, therefore it needs to be based on , , Determine the appropriate size of the equals Let the corner detection parameters This allows for the detection of multiple corner points in a single shot. One possible implementation is to take... , , The greatest common divisor is For example, the size of corner points in the object being measured includes , , and ,in , and The greatest common divisor is Then let So a single shot can be used to... , and Three types of corner points are detected; in addition, and The greatest common divisor is Then let You can do it at that time and Two types of corner points are detected.

[0045] Determining corner detection parameters Then, the phase distribution of the metasurface was determined. By setting the design parameters, the prepared metasurface is placed in an optical imaging system, and a single imaging operation can obtain an image with obvious light intensity at only the corner points of a preset size.

[0046] Phase distribution as follows:

[0047]

[0048] Here, floor refers to the floor operation, and arctan refers to the arctangent operation. Let be the two-dimensional spatial coordinates of the plane in which the metasurface lies. In some alternative embodiments, , By designing the phase distribution as described above, the phase of the metalens can be made to change with the azimuth angle. Changes in and Switching between them, and having The phase difference is used to achieve the angular-to-Hilbert transform. The angular-to-Hilbert transform is defined as applying a phase difference along the angular direction to the spectrum of the signal at a ratio of 2. Periodic transformation and amplitude difference of Fang Bo.

[0049] The phase design of metasurfaces can employ one of the following phase design methods: geometric phase, propagation phase, resonant phase, and chiral phase, or a combination of several phase design methods. When selecting metacells with both geometric and chiral phase modulation capabilities, the metacells are required to have high average transmittance and polarization conversion efficiency in the target wavelength band. When selecting metacells with propagation and resonant phase modulation capabilities, the two metacells are required to exhibit near-linear phase changes and near-linear phase conversion efficiency over a wide wavelength range. The phase difference.

[0050] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0051] like Figure 1 As shown, the optical corner detection imaging method based on angular Hilbert transform includes the following steps.

[0052] Step 1: Determine the size of the feature corner points in the object to be measured, and select appropriate corner detection parameters based on this information. .

[0053] Step 2: Determine the phase distribution of the metasurface This embodiment uses , For example, the specific expression is as follows:

[0054]

[0055] Step 3: Determine the design parameters of the metasurface and metacell, including center wavelength, aperture, substrate material, cell structure material, cell dimensions (length, width, and height), and cell structure period.

[0056] In this embodiment, the visible light band is selected as the working band, with a center wavelength set at 532 nm. Accordingly, silicon nitride (SiN) is selected, which has high transmittance and high refractive index in the visible light band. x As the constituent material of the metaunit, the substrate material is chosen to be quartz (SiO2), such as... Figure 2 As shown in (a) above. The superstructure unit adopts a cuboid shape, and therefore has a height h, length l, width w, and rotation angle. Four parameters, degrees of freedom The metasurface is constructed using a periodic arrangement of metaunits. The period must satisfy subwavelength conditions to avoid energy dispersion by higher-order diffraction; a period of 360 nm is chosen. The metasurface aperture is set to 1 mm. To achieve broadband corner detection, this embodiment employs a geometric phase design method for the metasurface, namely, local phase of the metasurface. Size and the rotation angle of the superunit Related. The specific relationship is as follows: .

[0057] Figure 2 (b) shows an angle detection parameter of The phase distribution of the metasurface, with the local phase in the angular direction. For a period of 0 and The dimensions of the superstructure remain unchanged, while the element rotation angle changes along the angular direction. and Changes between For example, you can take , ,or , wait.

[0058] To achieve efficient geometric phase modulation, the metaunits are required to possess high polarization conversion efficiency and transmittance. Therefore, a simulated scanning method is used to obtain... Figure 2 (c) represents the structure library of cuboid metaatomic atoms with a unit cell period of 360 nm. Within this library, unit parameters with the highest average polarization conversion efficiency and transmittance in the visible light band were selected. The final unit cell size w×l×h was chosen to be 120×280×1200 nm. 3 The transmittance and polarization conversion efficiency of this unit are as follows: Figure 2 As shown in (d), the black curve corresponds to transmittance, and the red curve corresponds to polarization conversion efficiency.

[0059] Corner detection parameters Taking a metasurface with a radius of 30° as an example Figure 3 (a) shows the phase distribution of the actually fabricated metasurface. It can be seen that the metasurface essentially achieves the desired phase distribution along the angular direction at 0 and... The phase distribution varies periodically between these phases. A physical microscopic image of this metasurface is shown below. Figure 3 As shown in (b), the metasurface exhibits high transmittance in the visible light band. The shape and distribution of the metaunits within the metasurface are as follows: Figure 3As shown in electron microscopy image (c), the top view of the microstructure is above the dashed line, and the side view is below. It can be seen that the metastructure satisfies the design requirement of constant element size, and the element rotation angle changes between 45° and 135° along the angular direction.

[0060] Step four: Place the prepared metasurface at the Fourier surface position in the optical imaging system.

[0061] As an optional embodiment, such as Figure 4 As shown, the optical imaging system is a standard 4f imaging system, including a first lens 3 and a second lens 5. A corner detection imaging device is built based on this 4f imaging system, including a sample 1, a polarizing filter 2, a metasurface 4, an analyzing polarizing filter 6, and an imaging sensor 7. After the collimated beam illuminates the sample 1, it passes through the polarizing filter 2 to obtain the required incident polarization state of the metasurface, and then enters the first lens 3 after being transmitted by the focal length of the first lens 3. After undergoing a Fourier transform in the first lens 3, it reaches the metasurface 4 located at the rear focal plane of the first lens 3, and then enters the second lens 5 after being transmitted by the focal length of the second lens 5. The second lens 5 performs an inverse Fourier transform and then passes through the polarizing filter 6 to eliminate background stray light. Finally, the beam reaches the imaging sensor 7, obtaining the final corner detection imaging effect.

[0062] As another alternative embodiment, such as Figure 5 As shown, the optical imaging system is a standard finite-distance single-lens imaging system, including a first lens 3. The optical path length of this imaging system is less than [missing information]. Figure 4 The standard 4f system shown is suitable for scenarios requiring a certain degree of optical path compactness. A corner detection imaging device is built based on this single-lens imaging system, including a sample 1, a polarizing filter 2, a metasurface 4, an analyzing polarizing filter 6, and an imaging sensor 7. A collimated beam illuminates sample 1, and after being polarized by polarizing filter 2, the required incident polarization state for the metasurface is obtained. A first lens 3 images the object, and the metasurface 4 is placed on the back focal plane of the first lens 3, then passes through the analyzing polarizing filter 6 to eliminate stray light. Finally, the imaging sensor 7 captures an image with corner detection capabilities.

[0063] Based on the requirement for compact optical path, utilize Figure 4 or Figure 5 The imaging optical path shown in the figure realizes an optical corner detection imaging method based on the angular Hilbert transform.

[0064] To showcase different The broadband corner detection effect of metasurfaces is shown in this embodiment. , , and Four types of metasurfaces were selected as examples. , and Three test wavelengths. Figure 6 (a) through (c) in the text demonstrate , , The corner detection effects of three metasurfaces on objects of different shapes under an illumination wavelength of 638nm are shown. The first row of images shows the bright field imaging effect of the object without the metasurface, and the second row shows the corner detection image after the metasurface is added, that is, the object only has obvious light intensity at the corners.

[0065] Figure 6 Each column (d) in the diagram shows... , , Below, with Figure 6 The results of corner detection for the same object in (a) to (c) are shown at different wavelengths. It can be seen that the corner detection imaging effect is basically the same when only the illumination wavelength is changed without changing the detection metasurface, which proves that the scheme has broadband corner detection imaging capability, that is, it is suitable for collimated white light illumination. Figure 6 (e) in the middle shows The metasurface was used to detect corner points on a cube array sample. It is evident that after adding the metasurface, the corner points of the cubes were highlighted throughout the entire imaging field of view, while other areas showed no light intensity response, demonstrating that this method can achieve corner point detection imaging with a wide field of view.

[0066] Figure 7 Showing The imaging effect of the metasurface on amplitude-type and phase-type samples. An amplitude-type sample is a sample in which the light intensity changes after the light beam passes through it, while a phase-type sample is a sample in which the light intensity does not change after the light beam passes through it. Figure 7 The first column of (a) and (b) shows the bright-field images of the two samples. Both samples exhibit the same corner features in both amplitude and phase images, i.e., they both have... Figure 7 The corner features are shown in (a) above. After inserting the metasurface, the light intensity in the areas without corner features is zero, and there is an intensity change only at the corners. The red box is a magnified image of the small area in the middle. It can be seen that although the bright-field images of the two objects are different, the corner detection imaging effect is consistent, proving that the scheme has the ability to detect corners with complex amplitudes, and that different corner angles correspond to different light intensity distributions.

[0067] The optical corner detection imaging system based on the angle-to-Hilbert transform described in this invention includes an object to be measured, an optical imaging system, micro-nano optical components, and an imaging sensor;

[0068] After collimated light illuminates the object under test, the light beam passes through the optical imaging system and reaches the imaging sensor to obtain corner detection imaging of the object under test;

[0069] The optical imaging system has micro-nano optical elements at its Fourier surface; the phase distribution of the micro-nano optical elements is set with a phase difference of 180° as the azimuth angle changes.

[0070] The electronic device of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is loaded into the processor, it implements the optical corner detection imaging method based on the angle-to-Hilbert transform.

[0071] The computer-readable storage medium of the present invention stores a computer program, which, when executed by a processor, implements the optical corner detection imaging method based on the angle-to-Hilbert transform.

[0072] The computer program product of the present invention includes a computer program that, when executed by a processor, implements the optical corner detection and imaging method based on angular Hilbert transform.

[0073] The computer-readable storage medium may include RAM, ROM, EEPROM, CD-ROM or other optical disc storage devices, magnetic disk storage devices or other magnetic storage devices, flash memory, or any other media that can be used to store program code in the form of instructions or data structures and is accessible by a computer.

[0074] The processor is used to execute a computer program stored in memory to implement the various steps in the methods described in the above embodiments.

Claims

1. An optical corner detection imaging method based on angular Hilbert transform, characterized in that, Includes the following steps: The object under test is illuminated by collimated light. After passing through the optical imaging system, the light beam reaches the imaging sensor to obtain the corner detection image of the object under test. The optical imaging system has micro / nano optical elements at its Fourier surface; the phase distribution of the micro / nano optical elements is adjusted according to the azimuth angle. The phase difference.

2. The optical corner detection imaging method based on angular Hilbert transform according to claim 1, characterized in that, Based on the size of the corner points in the item to be tested Determine corner detection parameters ; The phase distribution of the micro / nano optical element is as follows: Here, floor refers to the floor operation, and arctan refers to the arctangent operation. The coordinates of the plane in which the micro / nano optical element is located are two-dimensional spatial coordinates. .

3. The optical corner detection imaging method based on angular Hilbert transform according to claim 1, characterized in that, Based on the greatest common divisor of the sizes of all corner points of the object to be tested Determine corner detection parameters ; The phase distribution of the micro / nano optical element is as follows: Here, floor refers to the floor operation, and arctan refers to the arctangent operation. The coordinates of the plane in which the micro / nano optical element is located are two-dimensional spatial coordinates. .

4. The optical corner detection imaging method based on angular Hilbert transform according to claim 1, characterized in that, An insertable polarizing filter is provided behind the object to be tested, and an insertable polarizing filter is provided in front of the imaging sensor.

5. The optical corner detection imaging method based on angular Hilbert transform according to claim 1, characterized in that, The optical imaging system is a 4f imaging system or a single-lens imaging system.

6. The optical corner detection imaging method based on angular Hilbert transform according to claim 1, characterized in that, The micro / nano optical elements include one of the following: phase plate, diffractive optical element (DOE), freeform surface element, spatial light modulator, and metasurface; The phase plate, diffractive optical element (DOE), or freeform surface element employs a phase modulation method based on optical path accumulation. The spatial light modulator employs a phase modulation method based on liquid crystal modulation. The metasurface employs one phase design method or a combination of several phase design methods, including geometric phase, propagation phase, resonance phase, and chiral phase.

7. An optical corner detection imaging system based on angular Hilbert transform, characterized in that, This includes the object under test, optical imaging systems, micro / nano optical components, and imaging sensors; After collimated light illuminates the object under test, the light beam passes through the optical imaging system and reaches the imaging sensor to obtain corner detection imaging of the object under test; The optical imaging system has micro-nano optical elements at its Fourier surface; the phase distribution of the micro-nano optical elements is set with a phase difference of 180° as the azimuth angle changes.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is loaded into the processor, it implements the optical corner detection imaging method based on the angular Hilbert transform according to any one of claims 1-6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the optical corner detection imaging method based on the angular Hilbert transform according to any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the optical corner detection imaging method based on the angular Hilbert transform according to any one of claims 1-6.

Images