A high signal-to-noise ratio ghost imaging method based on orthogonal modulation of light spot of light source

Abstract

A high signal-to-noise ratio ghost imaging method based on orthogonal modulation of light spot of light source, a special light spot pattern is used to irradiate a sample to obtain a high signal-to-noise ratio image. The special light spot pattern makes the functions of the intensity of the light source irradiated on different positions of the sample changing with time orthogonal to each other, for example, the light intensity fluctuations at different positions have different frequencies. By detecting the light signal intensity fluctuations through the object and analyzing the frequency components of the signal, the shape of the object can be restored. The signal-to-noise ratio of the image is greatly improved. Correspondingly, the number of light spot patterns required for clear imaging is greatly reduced, thereby improving the imaging speed.

Description

Technical Field

[0001] This invention relates to the field of ghost imaging technology, and to a high signal-to-noise ratio ghost imaging method based on orthogonal modulation of light source spot. Background Technology

[0002] Ghost imaging is a novel imaging technique that utilizes relevant measurements to detect the projectivity or reflectivity of an object. It possesses the ability to image around obstacles or resist interference from the propagation medium, making it a research hotspot in the field of optical imaging in recent years. In ghost imaging, when detecting light passing through a sample, only a power detector is needed to measure the change in light intensity behind the object as the light spot pattern of the light source changes. An algorithm can then be used to obtain an image of the sample. Because it can be lensless, flexibly positioned, and unaffected by obstructions, it can be used to detect objects hidden behind occlusions. Performance improvements in ghost imaging mainly focus on two aspects: the signal-to-noise ratio (SNR) and the imaging speed. Generally, the SNR of ghost imaging is often low; obtaining a single image requires tens of thousands of measurements to reconstruct the object's outline. To achieve a clearer image, the number of measurements needs to be further increased. Especially for low-contrast objects, the effect of ghost imaging is often unsatisfactory. Since the SNR of ghost imaging is affected by the number of measurements, and the number of measurements affects the imaging speed, improvements in imaging speed and SNR in ghost imaging are often mutually reinforcing. Therefore, how to improve the imaging signal-to-noise ratio and imaging speed is one of the core research contents that has received widespread attention in the field of imaging, and it has also been a bottleneck problem that has been troubling people for a long time. Summary of the Invention

[0003] To address the issues of low signal-to-noise ratio and excessive measurement count in ghost imaging, the present invention aims to provide a high signal-to-noise ratio ghost imaging method based on orthogonal modulation of light source spot.

[0004] A high signal-to-noise ratio ghost imaging method based on orthogonal modulation of light source spot uses orthogonal spot pattern to illuminate the sample to obtain a high signal-to-noise ratio image. The overall optical path includes an area array light source, a lens group, a sample object, an obstacle, a power meter, and a computer in sequence.

[0005] A speckle pattern light source emits light through a lens array, which is then focused and projected onto a sample object. The light passing through the sample object is blocked by an obstacle and diffusely reflected. A power meter is used to detect the total light intensity after diffuse reflection through the obstacle. As the speckle pattern changes, the power of the light passing through the obstacle, as detected by the power meter, also changes. The power meter transmits the obtained light intensity signal to a computer, which uses a correlation algorithm between the speckle pattern and the final detected power to reconstruct the outline of the object.

[0006] Furthermore, the speckle pattern requires 2XY frames. The light with the speckle pattern is obtained by playing a certain number of speckle patterns at a certain frame rate, introducing XY different frequencies to obtain an image of X*Y pixels.

[0007] Furthermore, in a complete measurement process, the XY frequencies go through 1 to XY complete cycles, requiring a total of 2XY sampling points.

[0008] Furthermore, for speckle patterns, a set of trigonometric functions is chosen as the intensity modulation function; for a sample image of X*Y pixels, the intensity of light illuminating a spatial position (x,y) is expressed as:

[0009] I(x,y,n)=cos[2π(Xy+xX)n]

[0010] Furthermore, the total light intensity received by the power meter is expressed as follows using Fourier transform:

[0011]

[0012] Where t(x,y) is the transmittance of the sample at the spatial location (x,y);

[0013] To obtain the transmittance information of the sample, a Fourier transform is performed on the detected light intensity signal:

[0014]

[0015] That is, the transmittance information of the sample at position (x,y) will correspond to the data after calculating the Fourier transform p = Xy + xX, where ∝ means "proportional to", p represents the one-dimensional relative position of point (x,y) in the sample image X*Y, and δ represents the impulse function.

[0016] Furthermore, a lens group is placed between the obstacle and the power meter to converge the light transmitted through the object, allowing the power meter to receive more light information.

[0017] Furthermore, the sample object is patterned using a binary representation.

[0018] Furthermore, the sample objects are rendered using grayscale patterns.

[0019] Furthermore, the obstacle is made of a semi-transparent material. When light passing through the sample object encounters the obstacle, it undergoes diffuse reflection. The diffusely reflected light passes through the obstacle, and the power meter is positioned behind the obstacle to receive the signal.

[0020] Furthermore, the obstacle is made of an opaque material. When light passing through the sample object encounters the obstacle, it undergoes diffuse reflection. The diffusely reflected light is reflected in various directions in front of the obstacle, and the power meter is placed in front of the obstacle to receive the signal.

[0021] Beneficial effects: (1) By using the light spot pattern modulation method, the light intensity change of each point in space is controlled, and combined with the corresponding image restoration algorithm, a high signal-to-noise ratio ghost imaging system is developed; (2) The improvement of the signal-to-noise ratio can also reduce the number of speckle patterns required for imaging and reduce the detection time; (3) It has important inspirational significance and potential application value for the future development of a high signal-to-noise ratio and high speed ghost imaging system. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the optical path in an embodiment of the present invention.

[0023] Figure 2 This is a comparison chart of the numerical simulation effects of ordinary ghost imaging and frequency domain ghost imaging in an embodiment of the present invention.

[0024] Figure 3 This is a comparison diagram of the experimental results of ordinary ghost imaging and frequency domain ghost imaging in the embodiments of the present invention.

[0025] In the diagram, 1-area array light source, 2-lens group, 3-sample object, 4-obstacle, 5-power meter, 6-power meter head, 7-computer. Detailed Implementation

[0026] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings.

[0027] A light source projects light with a specific pattern onto the sample. Due to the continuous change in the light source pattern, each location on the object receives a series of light intensity signals. The power meter receives this signal modulated by the object's transmittance. Then, the correlation between the signal received by the power meter and the intensity signal at each location is calculated. The light spot pattern in classic ghost imaging is a random number sequence. Because the random number sequences illuminating different spatial locations are not orthogonal, the signal-to-noise ratio is too low. To solve this problem, this invention uses a trigonometric function array as an orthogonal function set to modulate the intensity of light illuminating various locations on the sample. For an X*Y pixel sample image, the intensity of light illuminating the spatial location (x,y) is expressed as:

[0028] I(x,y,n)=cos[2π(Xy+xX)n]

[0029] The signal received by the detector can be written as:

[0030]

[0031] Where t(x,y) is the transmittance of the sample at point (x,y), and n represents the sequence number of the light spot image. Since the attenuation of light by obstacles and the influence of disturbances on the light converged at various locations are not significantly different, this part of the influence is ignored in the above formula. To obtain the transmittance information of the sample, it is only necessary to perform a Fourier transform on the detected light intensity signal, find the first XY points of the low-frequency part of the Fourier frequency domain, and put them into the X*Y matrix in sequence. This corresponds to the transmittance of each point on the sample object, thus recovering the outline of the object.

[0032] Based on the above principles, such as Figure 1 As shown, the ghost imaging system includes: an area array light source 1, a lens group 2, a sample object 3, an obstacle 4, a power meter 5, a power meter head 6, and a computer 7. The area array light source 1 emits light with a certain speckle pattern, which is converged by the lens group 2 and projected onto the sample object 3. The light passing through the sample object 3 is blocked by the obstacle 4 and diffused. The power meter 5 detects the total light intensity passing through the obstacle 4. As the speckle pattern changes, the power of the light passing through the obstacle 4 detected by the power meter 5 also changes accordingly. The power meter 5 converts the light intensity signal collected by the power meter head 6 into an electrical signal and transmits it to the computer 7. An algorithm that correlates the speckle pattern with the finally detected power is used to reconstruct the image, thus obtaining the outline of the object.

[0033] Image restoration can be performed using Fourier transform, and the total light intensity received by the power meter can be written as:

[0034]

[0035] Where t(x,y) is the transmittance of the sample at the spatial location (x,y);

[0036] To obtain the transmittance information of the sample, a Fourier transform is performed on the detected light intensity signal:

[0037]

[0038] That is, the transmittance information of the sample at position (x,y) will correspond to the data after calculating the Fourier transform p = Xy + xX, where ∝ represents proportional to, p represents the one-dimensional relative position of point (x,y) in the sample image X*Y, and δ represents the impulse function.

[0039] The area array light source 1 uses an LED array. Alternatively, the area array light source can be replaced by a projector, screen, light source, and light intensity modulation device. When the area array light source is replaced by a light source and light intensity modulation device, the light emitted by the light source is incident on the light intensity modulator and DMD (digital micromirror device), and after modulation, it is incident on the lens group.

[0040] A single large-aperture convex lens or a group of large-aperture convex lenses can be added between the light source and the sample to change the imaging range and resolution. In this embodiment, reference... Figure 1 It can consist of two convex lenses. A lens can also be added between the obstacle and the power meter to converge the light transmitted through the object, allowing the power meter to receive more light information and resulting in better imaging.

[0041] The orthogonal function set uses trigonometric functions, but other orthogonal function sets can also be used instead, thus eliminating the need for a Fourier transform. Since the Fourier transform converts a time-domain signal to the frequency domain, the ghost imaging method that uses a trigonometric function set as the orthogonal function set can be called frequency-domain ghost imaging.

[0042] Sample object 3 uses a binary pattern. Alternatively, sample objects can be replaced with grayscale patterns of common objects in daily life.

[0043] Obstacle 4 is made of frosted glass. Alternatively, the obstacle can be replaced by a semi-transparent or opaque object such as a curtain or tracing paper. When the obstacle is replaced by another semi-transparent object, the light passing through the sample object is diffusely reflected when it encounters the obstacle. The diffusely reflected light passes through the obstacle, and the power meter is therefore placed behind the obstacle to receive the signal. When the obstacle is replaced by an opaque object, the light passing through the sample object is diffusely reflected when it encounters the obstacle. The diffusely reflected light is reflected in various directions in front of the obstacle, and the power meter needs to be placed in front of the obstacle to receive the signal.

[0044] like Figure 2 and Figure 3 As shown, the images recovered by the ordinary ghost imaging algorithm and the frequency domain ghost imaging algorithm were calculated in computer simulation and experiments, respectively, thus proving that the ghost imaging method based on orthogonal modulation of light source spots described in this invention can indeed improve the signal-to-noise ratio of ghost imaging. Figure 2 As shown, using computer numerical simulation, taking a 100*100 image as an example, both ordinary ghost imaging and frequency domain ghost imaging use 20,000 light spot patterns to recover sample information. Figure 2 (a) is the original image. Figure 2 (b) is a reconstruction of a normal ghost image from 20,000 light spot patterns. Figure 2 (c) is a reconstruction of a normal ghost image from 200,000 light spot patterns. Figure 2 (d) shows the frequency domain ghost imaging reconstruction image of 20,000 light spot patterns. In addition, actual samples were used for comparison in the experiment, with a 32*32 pixel LED dot matrix screen as the area light source. Both ordinary ghost imaging and frequency domain ghost imaging used 2048 light spot patterns to reconstruct sample information. The imaging effects are compared as follows: Figure 3 As shown, Figure 3 (a) is the original image. Figure 3 (b) is a reconstruction of a normal ghost image. Figure 3(c) is the frequency domain ghost imaging reconstruction.

[0045] The comparison of the results shows that frequency-domain ghost imaging can completely restore the original image using only a small number of light spot patterns. In numerical simulations, the image quality restored by classical ghost imaging is already poor when the number of samples is 20,000, but frequency-domain ghost imaging can still restore a high-quality image. Even with ten times the number of measurements, the classical ghost imaging algorithm still exhibits significant noise. This is because the algorithm of classical ghost imaging inherently introduces considerable noise, while the algorithm of frequency-domain ghost imaging does not introduce noise in principle. In actual systems, the noise in frequency-domain ghost imaging only comes from the optical path and detection system, as well as noise generated by computer numerical calculations; the latter noise is generally negligible. Furthermore, the experimental image quality is lower than the numerical simulation results because of factors such as fluctuations in light source intensity in the experiment.

[0046] This invention focuses on developing a high signal-to-noise ratio (SNR) ghost imaging method based on orthogonal modulation of light source spots. Specifically, it demonstrates superior performance compared to classical ghost imaging methods for a two-dimensional single-path ghost imaging system. Compared to existing ghost imaging SNR improvement methods, the orthogonal modulation ghost imaging technique not only retains the advantages of the original system, such as imaging through obstacles and being unaffected by medium disturbances, but also boasts a high SNR. It significantly improves the SNR without increasing the number of measurements, providing a novel approach to high SNR ghost imaging.

[0047] The above description is only a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. Any equivalent modifications or changes made by those skilled in the art based on the content disclosed in the present invention should be included within the scope of protection set forth in the claims.

Claims

1. A high signal-to-noise ratio ghost imaging method based on orthogonal modulation of light source spot, characterized in that: High signal-to-noise ratio images are obtained by illuminating the sample using orthogonal spot patterns. The overall optical path includes, in sequence, an area array light source, a lens group, the sample object, an obstacle, a power meter, and a computer. The area array light source emits light with a speckle pattern, which is focused and projected onto the sample object by the lens group. The light passing through the sample object is blocked by the obstacle and forms diffuse reflection. The power meter is used to detect the total light intensity diffusely reflected by the obstacle. As the light spot pattern changes, the power of the light passing through the obstacle detected by the power meter also changes accordingly; the power meter transmits the obtained light intensity signal to the computer, and the outline of the object is recovered through the correlation algorithm between the speckle and the final detected power. For a sample image of X*Y pixels, the total light intensity received by the power meter is expressed as follows using Fourier transform: Where t(x,y) is the transmittance of the sample at point (x,y), and n represents the sequence number of the light spot image of the light source; To obtain the transmittance information of the sample, a Fourier transform is performed on the detected light intensity signal: That is, the transmittance information of the sample at position (x,y) will correspond to the data after calculating the Fourier transform p = Xy + xX, where ∝ represents proportional to, p represents the one-dimensional relative position of point (x,y) in the sample image X*Y, and δ represents the impulse function.

2. The high signal-to-noise ratio ghost imaging method based on orthogonal modulation of light source spot according to claim 1, characterized in that: The speckle pattern requires 2XY frames. The light with the speckle pattern is obtained by playing a certain number of speckle patterns at a certain frame rate, introducing XY different frequencies to obtain an image of X*Y pixels.

3. The high signal-to-noise ratio ghost imaging method based on orthogonal modulation of light source spot according to claim 2, characterized in that: In a complete measurement process, the XY frequencies go through 1 to XY complete cycles, requiring a total of 2XY sampling points.

4. The high signal-to-noise ratio ghost imaging method based on orthogonal modulation of light source spot according to claim 1, characterized in that: For speckle patterns, a set of trigonometric functions is chosen as the intensity modulation function; for a sample image of X*Y pixels, the light illuminating the spatial position (x,y), where n represents the index of the light spot image, is expressed as: I(x,y,n)=cos[2π(Xy+xX)n].

5. The high signal-to-noise ratio ghost imaging method based on orthogonal modulation of light source spot according to claim 1, characterized in that: A lens group is placed between the obstacle and the power meter to converge the light transmitted through the object, so that the power meter can receive more light information.

6. The high signal-to-noise ratio ghost imaging method based on orthogonal modulation of light source spot according to claim 1, characterized in that: The sample object is represented by a binary pattern.

7. The high signal-to-noise ratio ghost imaging method based on orthogonal modulation of light source spot according to claim 1, characterized in that: The sample objects are presented in grayscale patterns.

8. The high signal-to-noise ratio ghost imaging method based on orthogonal modulation of light source spot according to claim 1, characterized in that: The obstacle is made of a semi-transparent material. When light passing through the sample object encounters the obstacle, it undergoes diffuse reflection. The diffusely reflected light passes through the obstacle, and the power meter is located behind the obstacle to receive the signal.

9. The high signal-to-noise ratio ghost imaging method based on orthogonal modulation of light source spot according to claim 1, characterized in that: The obstacle is made of an opaque material. When light passing through the sample object encounters the obstacle, it undergoes diffuse reflection. The diffusely reflected light is reflected in various directions in front of the obstacle. The power meter is placed in front of the obstacle to receive the signal.

Images