CN117288684A - Cone-shaped optical fiber single-pixel imaging system and method based on compressed sensing - Google Patents

Abstract

The invention discloses a tapered optical fiber single-pixel imaging system and a tapered optical fiber single-pixel imaging method based on compressed sensing, wherein the tapered optical fiber single-pixel imaging system and the tapered optical fiber single-pixel imaging method based on compressed sensing comprise the following steps: the system comprises a conical multimode fiber, a two-dimensional micrometer mobile station, an imaging module, a signal collection module and a point detector; the conical multimode optical fiber is arranged on the two-dimensional micrometer mobile station, the position of the conical multimode optical fiber is adjusted through the two-dimensional micrometer mobile station, and the laser light source emits multimode interference speckles after being transmitted by the conical multimode optical fiber and is used for irradiating and modulating sample information. The tail end core diameter of the conical multimode fiber can be controlled at a micron level, and the speckle illumination field of view is reduced by reducing the tail end core diameter so as to realize imaging of a sample with smaller size and realize calculation super-resolution imaging by combining a compressed sensing technology. The imaging limit of the method is far higher than that of a common multimode optical fiber.

Description

Tapered optical fiber single-pixel imaging system and method based on compressed sensing

Technical field

The invention belongs to the field of imaging technology, and in particular relates to a tapered optical fiber single-pixel imaging system and method based on compressed sensing.

Background technique

For centuries, optical microscopy has been an irreplaceable and important means of characterization of chemical, biological and other materials. With the development of technology, optical imaging has been breaking through to smaller sizes, and the Abelian diffraction limit shows that the resolution of far-field imaging systems has always been limited by the wavelength of light and the numerical aperture of the imaging system. In order to overcome the limitations caused by the far-field diffraction limit, a series of super-resolution imaging technologies have been broken through and developed, including near-field optical microscopy, which bypasses the limitations of the far-field and uses optical fiber probes to change the light source into a nanoscale light source and It is detected deep into the near-field region of the sample, that is, within a distance of less than one optical wavelength, and super-resolution imaging of a certain area is achieved by scanning. This technology is often used for imaging molecules or chemical materials. In addition, super-resolution microscopy technology based on fluorescence characteristics has good advantages in far-field observation. Photoactivated localization microscopy (PALM), stochastic optical reconstruction microscopy (STORM), and stimulated emission depletion (STED) microscopy have successively appeared. , super-resolution imaging based on optical scintillation (SOFI), super-resolution imaging based on scintillation radius (SRRF). Super-resolution microscopes based on fluorescent molecules are more suitable for analyzing biological slices, while optical fiber-based imaging systems have shown great advantages in deep tissue imaging. Their tiny size and good toughness can flexibly penetrate deep into living organisms for endoscopic observation. Imaging.

At present, endoscopic imaging technology based on optical fiber bundles, namely endoscopes, has been relatively maturely used in the medical industry. The fiber bundle itself is composed of thousands of optical fibers, which are not miniaturized in size. The imaging resolution depends on the number of optical fibers. The gaps between optical fibers can easily cause imaging artifacts. In recent years, imaging technology based on a single optical fiber has gradually been widely studied, and it has more advantages than fiber bundles in terms of miniaturization. The imaging method mainly uses the complex wavefront of wavefront shaping to generate laser focus points at different positions at the output end of the multimode fiber, thereby performing raster scanning imaging. In order to simplify the complexity of wavefront shaping and improve the imaging speed, the optical fiber speckle illumination imaging method based on the compressed sensing (CS) reconstruction algorithm has also been further developed. The essence of this imaging method is the same as that of a single-pixel camera. The differences are The modulated light spot does not simply rely on the digital micromirror (DMD) or the spatial light modulator (SLM), but uses the modal interference effect of multi-mode fiber or the speckle generated by the fiber end-face coating as the modulated light spot of the illumination, and finally uses barrel detection The detector collects one-dimensional intensity signals. Moreover, the imaging resolution of single multimode fiber imaging technology based on traditional wavefront shaping is limited by the diffraction limit of the fiber itself, and cannot further improve the imaging resolution. On the contrary, CS technology is based on the sparseness of the original signal and can reconstruct far more data than the compressed signal. Similarly, for sparse image signals, it can also reconstruct images with higher resolution than far-field images. Recently, simulations have discussed in detail the resolution limit of single-pixel imaging with speckle illumination based on CS technology. The essential factor limiting the imaging resolution is the sparsity of the sample. In addition, the number of data measurements, the selection of measurement speckles, and the performance of the reconstruction algorithm will also greatly affect the imaging quality and thus the imaging resolution. At the experimental level, recent research has used laser excitation to produce multiple interference speckles at different positions of the incident end core of multi-mode optical fibers to modulate illumination and detect fluorescence signals on fluorescent beads, and finally reconstruct the fluorescence image through a compressed sensing reconstruction algorithm. Imaging of submicron dimensions was achieved, which successfully exceeded the diffraction limit of multimode optical fiber. However, it is of great significance to further improve the resolution of fiber-optic endoscopic imaging, which will help to observe more microscopic activities in organisms or tissues.

Contents of the invention

The technical problem to be solved by the present invention is to provide a tapered optical fiber single-pixel imaging system and method based on compressed sensing in order to further improve the resolution of single-fiber imaging in the prior art. By combining multi-mode The optical fiber is prepared into a tapered multi-mode optical fiber to reduce the imaging field of view and thereby effectively improve the imaging resolution.

The present invention is realized through the following technical solutions:

On the one hand, the present invention provides a tapered optical fiber single-pixel imaging system and method based on compressed sensing, including: tapered multi-mode optical fiber, two-dimensional micron mobile stage, imaging module, signal collection module and point detector; tapered The multimode fiber is installed on a two-dimensional micron moving stage. The position of the tapered multimode fiber is adjusted through the two-dimensional micron moving stage. After the laser light source is transmitted through the tapered multimode fiber, it emits multimode interference speckles that change with the spatial position. to irradiate and modulate sample information.

Tapered multimode fiber is made by transversely tapering a multimode fiber and cutting it from the waist region. Multimode fiber transmits multiple modes at a given operating wavelength. The two-dimensional micron mobile stage changes the laser incident position by moving the optical fiber fixed on it, thereby generating spatially modulated random multi-mode interference speckles. The minimum moving distance of the two-dimensional micron mobile stage is not greater than 0.1μm, and the movable direction is at least x-y directions.

The imaging module includes an objective lens and a camera, which is used to image the sample and observe the illumination of the laser speckle on the sample.

The signal collection module includes an objective lens and a beam splitter point detector. The signal collection module uses the objective lens to focus the light intensity information after speckle modulation of the sample and uses a beam splitter to separate a path of light and collect it with a point detector.

The point detector is used to detect the one-dimensional light intensity information after the modulated speckle is applied to the sample, and reconstruct the two-dimensional image of the undersampled one-dimensional light intensity through the compressed sensing reconstruction algorithm;

A tapered multi-mode optical fiber super-resolution imaging method based on compressed sensing and single-pixel imaging, which is characterized by including the following steps: S1 generates multi-mode interference speckles, S2 generates spatially modulated speckle illumination sources, and S3 adjusts the speckle illumination. The field of view of the sample, S4 represents the modulated speckle sequence information, and S5 collects the light intensity information and reconstructs the imaging.

The specific steps are as follows

S1: The laser light source is incident on the input end face of the tapered multi-mode fiber and is transmitted over a distance to produce laser speckles of multiple modes of interference;

S2: Use a two-dimensional micron platform to control the movement of the tapered multimode optical fiber input surface to generate spatially modulated random laser speckles;

S3: The output end of the tapered multimode optical fiber is directly close to the sample illumination. By controlling the relative distance between the output end of the tapered multimode optical fiber and the sample, the change in the illumination field of view is controlled, thereby achieving imaging with different magnifications;

S4: Use the imaging module composed of the objective lens and the camera to accurately control the illumination position and field of view size of the modulated speckle, and use the camera to pre-characterize the speckle information as an input parameter for subsequent image reconstruction;

S5: The signal collection module focuses the light intensity information after the modulated speckle acts on the sample and collects it through the point detector. The sample image is reconstructed based on the detected light intensity and speckle information through the compressed sensing reconstruction algorithm.

The invention fixes the end face of the tapered multi-mode optical fiber on a two-dimensional micron platform; controls the movement of the micron platform to change the excitation position of the laser on the end face of the tapered multi-mode optical fiber; generates spatially modulated random multi-mode interference speckles; and passes the point detector The light intensity after the modulated speckle acts on the sample is collected. Based on the speckle and the detected light intensity, the sample image information is reconstructed using the compressed sensing reconstruction algorithm. Compared with existing technology, it has the following advantages:

(1) Due to its fixed end core diameter, the traditional multi-mode optical fiber imaging technology based on compressed sensing and single-pixel imaging cannot further reduce the illumination field of view when irradiating the sample. However, the present invention proposes to prepare the multi-mode optical fiber into a tapered multi-mode Optical fiber can flexibly control the minimum illumination field of view by preparing tapered multi-mode optical fibers with different end core diameters.

(2) Traditional multi-mode optical fiber imaging technology based on compressed sensing and single-pixel imaging has limited improvement in imaging resolution. However, the tapered multi-mode optical fiber used in the present invention reduces the illumination field of view while retaining the multi-mode speckle characteristics. , showing greater advantages than ordinary multimode optical fiber in CS-based computational super-resolution imaging.

Description of drawings

Figure 1 is a schematic diagram of the optical path of the tapered multi-mode optical fiber imaging system of the present invention;

Figure 2 is an imaging principle diagram of the present invention combining compressed sensing and single-pixel imaging technology;

Figure 3 is an optical diagram of the tapered multimode optical fiber prepared by the present invention.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below through specific implementation and in conjunction with the accompanying drawings.

Embodiment 1

Traditional single multimode fiber imaging based on wavefront shaping has slow imaging speed and imaging resolution is limited by the diffraction limit of the fiber due to its scanning imaging characteristics. Although single multi-mode fiber imaging based on compressed sensing single-pixel imaging improves imaging efficiency and achieves sub-micron size imaging, its imaging field of view cannot be further reduced and the imaging resolution still needs to be further improved.

This embodiment discloses a tapered multi-mode optical fiber imaging system based on compressed sensing and single-pixel imaging. The multi-mode optical fiber is prepared so that the end core diameter is smaller than the original size and the core diameter size is flexibly controlled to the micron level. The tapered multimode optical fiber is combined with a two-dimensional micron moving stage to generate spatially modulated multimode interference speckles as modulated light. The tapered multimode fiber retains the modulated illumination speckle while reducing the minimum imaging field of view, effectively achieving imaging of smaller size samples and combining with CS technology to achieve computational super-resolution imaging.

The specific system module diagram of this embodiment is shown in Figure 1, including:

Narrowband laser (Laser); in this embodiment, a 532nm laser is used as the light source and focused on the input end face of the tapered multi-mode optical fiber through the focusing optical path. Lasers of different bands can be selected according to the needs of the imaging band;

Tapered multimode optical fiber (Tapered MMF) is made by transversely tapering ordinary multimode optical fiber and cutting it from the waist area. The end of the optical fiber is conical, and the end core diameter ranges from a few microns to tens of microns, Figure 3 The optical diagrams of three different sizes of tapered multimode optical fibers are given; in this embodiment, the laser speckle generated by the tapered multimode optical fiber is used as the modulated light source for sample imaging;

Two-dimensional micron moving stage (XY stage); a programmable micron stage with two moving directions of x-y, the minimum moving accuracy can reach 0.1μm; in this embodiment, it is used to move the incident end face of the tapered multi-mode fiber to change the laser excitation Position produces spatially modulated multimode interference speckles;

Imaging module; located behind the sample, consisting of an objective lens (OL) and a camera (CCD); in this embodiment, it is used to image the sample to observe the irradiation of the laser speckle on the sample;

Signal collection module; located behind the sample, consists of an objective lens (OL), a beam splitter (BS) and a point detector (PD); in this embodiment, it is used to focus the light intensity signal after the speckle acts on the sample and use the point detector to detector collection.

According to the compressed sensing-based tapered optical fiber single-pixel imaging system and method proposed by embodiments of the present invention, the multi-mode optical fiber is prepared so that the end core diameter is smaller than the original size and the core diameter size is flexibly controlled to the micron level. The tapered multimode optical fiber is combined with a two-dimensional micron moving stage to generate spatially modulated multimode interference speckles as modulated light. The tapered multimode fiber retains the modulated illumination speckle while reducing the minimum imaging field of view, effectively achieving imaging of smaller size samples and combining with CS technology to achieve computational super-resolution imaging.

Embodiment 2

This embodiment discloses a tapered multi-mode optical fiber super-resolution imaging method based on compressed sensing and single-pixel imaging. This embodiment is an imaging method implemented based on the imaging system in Embodiment 1. The specific principle is shown in Figure 2 shown, the specific steps are as follows:

S1: The laser light source is incident on the input end face of the tapered multi-mode optical fiber and then propagates over a certain distance to produce laser speckles of multiple modes of interference;

S2: Use a two-dimensional micron platform to control the movement of the tapered multimode optical fiber input surface to generate spatially modulated random laser speckles;

A two-dimensional path planning scan is performed on the two-dimensional micron stage. The tapered multi-mode optical fiber end face fixed on the micron stage is scanned with the corresponding two-dimensional path, so that the laser incident point is relatively scanned; the total range of the scanning path must not exceed the fiber core. The range of sizes, speckles required for the series of illumination generated with the scan in the example. All illumination speckles correspond to the measurement matrix Ф in CS theory, as shown in Figure 2.

S3: The output end of the tapered multimode optical fiber is directly close to the sample illumination. By controlling the relative distance between the output end of the tapered multimode optical fiber and the sample, the change in the illumination field of view is controlled, thereby achieving imaging with different magnifications;

S4: Use the imaging module composed of the objective lens and the camera to accurately control the illumination position and field of view size of the modulated speckle, and use the camera to pre-characterize the speckle information as an input parameter for subsequent image reconstruction;

S5: The signal collection module focuses the light intensity information after the modulated speckle acts on the sample and collects it through the point detector, and reconstructs the sample image based on the detected light intensity and speckle information through the compressed sensing reconstruction algorithm;

The characterized speckle information forms the measurement matrix Ф in CS theory. Specifically, each speckle is expanded into a one-dimensional row vector, and the length of the row vector is defined as N, where N also corresponds to the signal length of the sample x to be measured, that is, the resolution. rate, the number of speckles is defined as M; therefore the measurement matrix Ф is an M×N two-dimensional matrix. Since the number of speckle measurements M is usually much smaller than the signal length N of the sample x to be measured, this constitutes an undersampling model in CS theory and lays a theoretical foundation for realizing computational super-resolution imaging. Defining the one-dimensional light intensity signal measured by the point detector as y, the entire single-pixel detection process can be described as the following mathematical process:

y=Φx (1)

Among them, y is the light intensity signal obtained by the point detector, Ф is the measurement matrix composed of illumination speckles, and x is the sample to be imaged. In CS theory, the solution to the underdetermined equation of equation (1) is based on the sparsity constraint of the sample. Here, the sample x itself does not need to be sparse, but only needs to have sparse characteristics in a certain transformation domain. Most samples in nature are sparse in certain transformation domains, such as cosine transformation domain, gradient transformation domain, etc. Based on the sparse constraint of the sample, the CS reconstruction algorithm is used to solve equation (1), and finally the sample signal x is obtained.

In short, the present invention uses a tapered multi-mode fiber as a modulated light generating device for single-pixel imaging. It generates modulated speckles and can flexibly adjust the minimum imaging field of view to achieve imaging of smaller samples. The speckles are then characterized by a camera. information and collects the light intensity information after speckle modulation of the sample through a point detector. Finally, the image is reconstructed based on CS technology and computational super-resolution imaging is completed.

It is necessary to point out here that the above examples and test examples are only for further elaboration and understanding of the technical solution of the present invention, and cannot be understood as further limitations of the technical solution of the present invention. Those skilled in the art will not make outstanding achievements. Inventions and creations with substantial features and significant advancements still fall within the scope of protection of the present invention.

Claims (10)

1. A tapered optical fiber single-pixel imaging system and method based on compressed sensing are characterized by comprising the following steps: the system comprises a conical multimode fiber, a two-dimensional micrometer mobile station, an imaging module, a signal collection module and a point detector; the conical multimode optical fiber is arranged on the two-dimensional micrometer mobile station, the movement of the conical multimode optical fiber is controlled by the two-dimensional micrometer mobile station, the position of the corresponding laser incident on the end face of the conical multimode optical fiber changes, and the laser source emits multimode interference speckles modulated by the spatial position after being transmitted by the conical multimode optical fiber and is used for irradiating and modulating sample information.

2. The compressed sensing based tapered fiber single-pixel imaging system and method as claimed in claim 1, wherein the tapered multimode fiber is made from multimode fiber that is tapered laterally and cut from the waist region.

3. The compressed sensing-based tapered fiber optic single-pixel imaging system and method as claimed in claim 2, wherein: multimode optical fibers transmit multiple modes at a given operating wavelength.

4. The compressed sensing-based tapered optical fiber single-pixel imaging system and method according to claim 3, wherein the two-dimensional micrometer movable stage changes the incident position of the laser by moving the optical fiber fixed on the two-dimensional micrometer movable stage, thereby generating spatially modulated random multimode interference speckle, the minimum moving distance of the two-dimensional micrometer movable stage is not more than 0.1 μm, and the movable direction is at least in the x-y directions.

5. The compressed sensing-based tapered fiber single-pixel imaging system and method as claimed in claim 4, wherein the imaging module comprises an objective lens and a camera for imaging the sample to observe the illumination of the sample by the laser speckle.

6. The compressed sensing based tapered fiber single-pixel imaging system and method of claim 5, wherein the signal collection module comprises an objective lens, a beam splitter, and a point detector.

7. The system and method for tapered fiber single-pixel imaging based on compressed sensing as claimed in claim 6, wherein the signal collection module is an objective lens for focusing the light intensity information of the speckle-modulated sample and a beam splitter for splitting a path of light for collection by the point detector.

8. The compressed sensing-based tapered optical fiber single-pixel imaging system and method according to claim 6 or 7, wherein the point detector is used for detecting one-dimensional light intensity information after modulation speckle is applied to the sample, and performing two-dimensional image reconstruction on undersampled one-dimensional light intensity through a compressed sensing reconstruction algorithm.

9. A conical multimode optical fiber super-resolution imaging method based on compressed sensing and single-pixel imaging is characterized by comprising the following steps of: s1 generates multimode interference speckle, S2 generates a space modulation speckle illumination source, S3 adjusts the field of view of a speckle illumination sample, S4 characterizes modulated speckle sequence information, and S5 collects light intensity information and reconstructs an image.

10. The method for tapered multimode fiber super-resolution imaging based on compressed sensing and single-pixel imaging as claimed in claim 9 comprises the following specific steps of

S1: the laser source enters the tapered multimode fiber input end face to be transmitted for a certain distance to generate laser speckles with multiple mode interference;

s2: controlling the movement of the input surface of the tapered multimode fiber to generate spatially modulated random laser speckles by using a two-dimensional micrometer stage;

s3: the conical multimode fiber output end is directly close to the sample for illumination, and the change of the illumination field range is controlled by controlling the relative distance between the conical multimode fiber output end and the sample, so that imaging with different magnification is realized;

s4: an imaging module formed by an objective lens and a camera is utilized to precisely control the irradiation position and the view field size of the modulated speckle, and the camera is utilized to pre-characterize the speckle information as an input parameter of the subsequent image reconstruction;

s5: the signal collection module focuses the light intensity information after the modulated speckles act on the sample and collects the light intensity information through the point detector, and the image of the sample is reconstructed according to the detected light intensity and speckle information through the compressed sensing reconstruction algorithm.