Holographic three-dimensional imaging method and system

Abstract

The application provides a holographic three-dimensional imaging method, which comprises the following steps: performing multiple holographic image sampling on a three-dimensional object, and performing compressed sensing reconstruction on the holographic images obtained through the multiple sampling to form a three-dimensional holographic image; wherein each time of holographic image sampling on the three-dimensional object comprises the following steps: loading a random hologram in a spatial light modulator; modulating a laser beam by using the spatial light modulator loaded with the random hologram to generate a random pattern; irradiating the three-dimensional object with the random pattern, and sampling the three-dimensional object by using an imaging device to obtain a holographic image; wherein when performing the compressed sensing reconstruction, an observation matrix is constructed according to the random patterns used in the multiple holographic sampling. The application can realize three-dimensional holographic imaging by collecting less data.

Description

Technical Field

[0001] The present invention relates to the field of holographic imaging technology, and in particular to a holographic three-dimensional imaging method and system. Background Art

[0002] Traditional optical imaging technology is based on the principle of lens imaging, using sensors to capture two-dimensional light intensity information and then generating images through digital signal processing. Although this method is mature and widely used, it relies on high-resolution sensors and large data storage, resulting in high hardware costs and bulky systems. In addition, traditional imaging requires complete sampling of the entire scene, which is prone to information loss or noise interference in low-light, high-speed motion, or high dynamic range scenes. It is precisely because of the aforementioned shortcomings of the lens imaging principle that lens imaging is extremely inconvenient to apply in the field of three-dimensional holographic imaging. Summary of the Invention

[0003] The holographic three-dimensional imaging method and system provided by the present invention can realize three-dimensional holographic imaging by collecting less data.

[0004] In a first aspect, the present invention provides a holographic three-dimensional imaging method, the method comprising:

[0005] Performing multiple holographic image sampling on a three-dimensional object, and performing compressed sensing reconstruction on the holographic images obtained by the multiple sampling to form a three-dimensional holographic image; wherein each holographic image sampling on the three-dimensional object includes:

[0006] loading a random hologram in a spatial light modulator;

[0007] A random hologram-loaded spatial light modulator is used to modulate the laser beam to generate a random pattern.

[0008] irradiating the three-dimensional object with the random pattern and sampling the three-dimensional object with an imaging device to obtain a holographic image;

[0009] Wherein, when performing compressed sensing reconstruction, an observation matrix is ​​constructed according to the random pattern used in multiple hologram samplings.

[0010] Optionally, performing compressed sensing reconstruction on the holographic image obtained by multiple samplings includes:

[0011] The holographic images obtained by sampling the holographic images of the three-dimensional object M times are used as vector elements to construct an observation vector with a dimension of M;

[0012] The random patterns used in each holographic image sampling of the three-dimensional object are arranged into a sampling vector of length N, and the sampling vectors corresponding to the M times of holographic image sampling of the three-dimensional object are arranged into a measurement matrix of dimension M*N;

[0013] The three-dimensional holographic image corresponding to the three-dimensional object is reconstructed based on the observation vector, the observation matrix and the compressed sensing reconstruction algorithm.

[0014] Optionally, reconstructing the three-dimensional holographic image corresponding to the three-dimensional object based on the observation vector, the observation matrix and the compressed sensing reconstruction algorithm includes:

[0015] The sparse domain vector of the three-dimensional holographic image is determined based on the observation vector, the observation matrix and the compressed sensing reconstruction algorithm; wherein each element of the observation vector is an image, and each element of the sparse domain vector is a matrix.

[0016] Optionally, determining the sparse domain vector of the three-dimensional holographic image based on the measurement vector, the measurement matrix and the compressed sensing reconstruction algorithm includes:

[0017] The sparse domain vector is determined according to the following formula:

[0018]

[0019] Wherein, x1 represents the sparse domain vector of the three-dimensional holographic image, Indicates taking the minimum value of the function, express norm, This means that the solution to x1 is limited to , where x is the three-dimensional object information, Φ is the observation matrix, and y is the observation vector.

[0020] Optionally, after the step of determining the sparse domain vector of the three-dimensional holographic image based on the measurement vector, the measurement matrix and the compressed sensing reconstruction algorithm, the method further includes:

[0021] An inverse transformation of the sparse transformation is performed on the sparse domain vector to obtain the three-dimensional holographic image.

[0022] Optionally, the modulating the laser beam using a spatial light modulator loaded with a random hologram includes:

[0023] The phase of the laser beam is modulated using a reflective or transmissive spatial light modulator loaded with a random hologram.

[0024] Optionally, the sampling of the three-dimensional object by using an imaging device includes:

[0025] A charge-coupled device (CCD) or a complementary metal oxide semiconductor (CMOS) imaging device is used to sample three-dimensional objects.

[0026] Optionally, the sampling of the three-dimensional object by using an imaging device includes:

[0027] The three-dimensional object is imaged by using a spherical lens or an aspherical lens and projected onto an imaging device.

[0028] Optionally, the modulating the laser beam using a spatial light modulator loaded with a random hologram includes:

[0029] A spatial light modulator loaded with a random hologram is used to modulate the laser beam emitted by a diode laser or a solid-state laser.

[0030] In a second aspect, the present invention provides a holographic three-dimensional imaging system, comprising:

[0031] A spatial light modulator, used to sequentially load multiple random holograms and modulate the laser beam before irradiating it onto a three-dimensional object;

[0032] An image sensor, configured to sample a holographic image of the three-dimensional object;

[0033] A processor is configured to obtain a random pattern formed each time the spatial light modulator loads a random hologram and modulates the laser beam, and to construct an observation matrix based on the random patterns formed multiple times; the processor is further configured to obtain a holographic image obtained each time the image sensor performs holographic image sampling, and to construct an observation vector based on the multiple holographic images; the processor is further configured to reconstruct a three-dimensional holographic image corresponding to the three-dimensional object based on the observation vector, the observation matrix, and a perception compression reconstruction algorithm.

[0034] In the technical solution provided by the present invention, a laser beam is first modulated by a spatial light modulator preloaded with a random hologram. The generated random pattern is then irradiated onto a three-dimensional object. An imaging device is then used to capture holographic images with varying depth information, thereby capturing the object's three-dimensional information. This 3D holographic image is then reconstructed using 3D compressed sensing reconstruction technology. This imaging method can simultaneously obtain information about the object in both the depth and horizontal directions, and the collected data can be processed using a compressed sensing algorithm to achieve efficient 3D holographic imaging. Because the 3D compressed sensing reconstruction algorithm can fully utilize the object's holographic information, it can reconstruct 3D holographic images with greater accuracy, clarity, and resolution, offering broader application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 FIG. 4 is a flow chart of a holographic 3D imaging method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0037] The embodiment of the present invention provides a holographic three-dimensional imaging method, such as Figure 1 As shown, the method includes:

[0038] Performing multiple holographic image sampling on a three-dimensional object, and performing compressed sensing reconstruction on the holographic images obtained by the multiple sampling to form a three-dimensional holographic image; wherein each holographic image sampling on the three-dimensional object includes:

[0039] loading a random hologram in a spatial light modulator;

[0040] A random hologram-loaded spatial light modulator is used to modulate the laser beam to generate a random pattern.

[0041] irradiating the three-dimensional object with the random pattern and sampling the three-dimensional object with an imaging device to obtain a holographic image;

[0042] Wherein, when performing compressed sensing reconstruction, an observation matrix is ​​constructed according to the random pattern used in multiple hologram samplings.

[0043] In some embodiments, after the laser beam is processed by a spatial light modulator loaded with a hologram, each part of the laser beam will have different phase information. By irradiating the laser beam on a three-dimensional object and then reflecting it from the three-dimensional object, the imaging device collects the laser reflected from the three-dimensional object to obtain a holographic image carrying different depth information. During multiple holographic image sampling processes, the random hologram loaded each time is different, so that the depth information carried by the holographic image collected by each holographic image sampling is also different. In the process of compressed sensing reconstruction, since the algorithm is essentially a reconstruction of high-dimensional data using low-dimensional data, multiple holographic images can be constructed to form an observation vector, and the image formed by modulating the laser beam after loading the hologram is constructed to form an observation matrix. After solving it according to the compressed sensing reconstruction algorithm, a three-dimensional holographic image can be obtained.

[0044] In the technical solution provided by an embodiment of the present invention, a laser beam is first modulated by a spatial light modulator preloaded with a random hologram. The generated random pattern is then irradiated onto a three-dimensional object. An imaging device is then used to capture holographic images with varying depth information, thereby capturing the object's three-dimensional information. This 3D holographic image is then reconstructed using 3D compressed sensing reconstruction technology. This imaging method can simultaneously obtain information about the object in both the depth and horizontal directions, and the collected data can be processed using a compressed sensing algorithm to achieve efficient 3D holographic imaging. Because the 3D compressed sensing reconstruction algorithm can fully utilize the object's holographic information, it can reconstruct 3D holographic images with greater accuracy, clarity, and resolution, offering broader application prospects.

[0045] As an optional implementation, the performing compressed sensing reconstruction on the holographic image obtained by multiple samplings includes:

[0046] The holographic images obtained by sampling the holographic images of the three-dimensional object M times are used as vector elements to construct an observation vector with a dimension of M;

[0047] The random patterns used in each holographic image sampling of the three-dimensional object are arranged into a sampling vector of length N, and the sampling vectors corresponding to the M times of holographic image sampling of the three-dimensional object are arranged into a measurement matrix of dimension M*N;

[0048] The three-dimensional holographic image corresponding to the three-dimensional object is reconstructed based on the observation vector, the observation matrix and the compressed sensing reconstruction algorithm.

[0049] In some embodiments, the holographic image is a vector element, which causes each matrix element itself to be in matrix form. At the same time, the dimension of the observation vector is the same as the number of rows of the observation matrix, and the number of columns of the observation matrix should be the same as the dimension of the sparse domain vector of the three-dimensional holographic image. Therefore, the multiplication of the observation matrix and the sparse domain vector should be understood as the multiplication of each row of the observation matrix and the sparse domain vector. At this time, since the elements of the observation matrix are numerical values, each element of the sparse domain vector should be in matrix form and the matrix size should be equal to the size of the holographic image. Only then can the observation matrix and the sparse domain vector be multiplied to obtain the observation vector. In some embodiments, each row vector of the observation matrix corresponds to the random pattern used when sampling a holographic image, and each element in the row vector corresponds to a pixel in the random pattern.

[0050] As an optional implementation, reconstructing the three-dimensional holographic image corresponding to the three-dimensional object based on the observation vector, the observation matrix, and the compressed sensing reconstruction algorithm includes:

[0051] The sparse domain vector of the three-dimensional holographic image is determined based on the observation vector, the observation matrix and the compressed sensing reconstruction algorithm; wherein each element of the observation vector is an image, and each element of the sparse domain vector is a matrix.

[0052] In some embodiments, the sparse domain vector is described as having at least k zero elements, where k is not less than 1. In this embodiment, each element of the sparse domain vector is a matrix. Therefore, in this embodiment, the sparse domain matrix should be described as having at least k zero matrices, and a zero matrix should be understood as a matrix in which all elements are 0. In some preferred embodiments, the selection of the sparse domain can be determined based on, for example, a priori results.

[0053] As an optional implementation manner, determining the sparse domain vector of the three-dimensional holographic image based on the observation vector, the observation matrix and the compressed sensing reconstruction algorithm includes:

[0054] The sparse domain vector is determined according to the following formula:

[0055]

[0056] Wherein, x1 represents the sparse domain vector of the three-dimensional holographic image, Indicates taking the minimum value of the function, express norm, This means that the solution to x1 is limited to , where x is the three-dimensional object information, Φ is the observation matrix, and y is the observation vector.

[0057] As an optional implementation, after the step of determining the sparse domain vector of the three-dimensional holographic image based on the measurement vector, the measurement matrix, and the compressed sensing reconstruction algorithm, the method further includes:

[0058] An inverse transformation of the sparse transformation is performed on the sparse domain vector to obtain the three-dimensional holographic image.

[0059] As an optional implementation, the modulating the laser beam using a spatial light modulator loaded with a random hologram includes:

[0060] The phase of the laser beam is modulated using a reflective or transmissive spatial light modulator loaded with a random hologram.

[0061] In some embodiments, the pixel pitch of the spatial light modulator may be, for example, 10 micrometers, and the resolution may be, for example, 1980*1080.

[0062] As an optional implementation manner, the sampling of the three-dimensional object by using an imaging device includes:

[0063] A charge-coupled device (CCD) or a complementary metal oxide semiconductor (CMOS) imaging device is used to sample three-dimensional objects.

[0064] In some embodiments, the focal length of the imaging device may be, for example, 2.5 mm to 100 mm, the pixel size may be, for example, 5.4 um, and the resolution may be, for example, 4K.

[0065] As an optional implementation manner, the sampling of the three-dimensional object by using an imaging device includes:

[0066] The three-dimensional object is imaged by using a spherical lens or an aspherical lens and projected onto an imaging device.

[0067] As an optional implementation, the modulating the laser beam using a spatial light modulator loaded with a random hologram includes:

[0068] A spatial light modulator loaded with a random hologram is used to modulate the laser beam emitted by a diode laser or a solid-state laser.

[0069] In some embodiments, the wavelength of the laser can be selected to be 532 nm, for example.

[0070] An embodiment of the present invention further provides a holographic three-dimensional imaging system, comprising:

[0071] A spatial light modulator, used to sequentially load multiple random holograms and modulate the laser beam before irradiating it onto a three-dimensional object;

[0072] An image sensor, configured to sample a holographic image of the three-dimensional object;

[0073] A processor is configured to obtain a random pattern formed each time the spatial light modulator loads a random hologram and modulates the laser beam, and to construct an observation matrix based on the random patterns formed multiple times; the processor is further configured to obtain a holographic image obtained each time the image sensor performs holographic image sampling, and to construct an observation vector based on the multiple holographic images; the processor is further configured to reconstruct a three-dimensional holographic image corresponding to the three-dimensional object based on the observation vector, the observation matrix, and a perception compression reconstruction algorithm.

[0074] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A holographic three-dimensional imaging method, characterized in that: The method comprises: Performing multiple holographic image sampling on a three-dimensional object, and performing compressed sensing reconstruction on the holographic images obtained by the multiple sampling to form a three-dimensional holographic image; wherein each holographic image sampling on the three-dimensional object includes: loading a random hologram in a spatial light modulator; A random hologram-loaded spatial light modulator is used to modulate the laser beam to generate a random pattern. irradiating the three-dimensional object with the random pattern and sampling the three-dimensional object with an imaging device to obtain a holographic image; Wherein, when performing compressed sensing reconstruction, an observation matrix is ​​constructed according to the random pattern used in multiple hologram samplings; The step of reconstructing the holographic image obtained by multiple samplings through compressed sensing includes: The holographic images obtained by sampling the holographic images of the three-dimensional object M times are used as vector elements to construct an observation vector with a dimension of M; The random patterns used in each holographic image sampling of the three-dimensional object are arranged into a sampling vector of length N, and the sampling vectors corresponding to the M times of holographic image sampling of the three-dimensional object are arranged into a measurement matrix of dimension M*N; Reconstructing a three-dimensional holographic image corresponding to the three-dimensional object based on the observation vector, the observation matrix and the compressed sensing reconstruction algorithm; The reconstructing of the three-dimensional holographic image corresponding to the three-dimensional object based on the observation vector, the observation matrix and the compressed sensing reconstruction algorithm includes: Determine a sparse domain vector of the three-dimensional holographic image based on an observation vector, an observation matrix, and a compressed sensing reconstruction algorithm; wherein each element of the observation vector is an image, and each element of the sparse domain vector is a matrix; The step of determining the sparse domain vector of the three-dimensional holographic image based on the observation vector, the observation matrix, and the compressed sensing reconstruction algorithm includes: The sparse domain vector is determined according to the following formula: Wherein, x1 represents the sparse domain vector of the three-dimensional holographic image, Indicates taking the minimum value of the function, express norm, This means that the solution to x1 is limited to , where x is the three-dimensional object information, Φ is the observation matrix, and y is the observation vector.

2. The method according to claim 1, characterized in that After the step of determining the sparse domain vector of the three-dimensional holographic image based on the measurement vector, the measurement matrix and the compressed sensing reconstruction algorithm, the method further includes: An inverse transformation of the sparse transformation is performed on the sparse domain vector to obtain the three-dimensional holographic image.

3. The method according to claim 1, characterized in that The method of modulating the laser beam using a spatial light modulator loaded with a random hologram comprises: The phase of the laser beam is modulated using a reflective or transmissive spatial light modulator loaded with a random hologram.

4. The method according to claim 1, wherein The sampling of the three-dimensional object by using an imaging device includes: A charge-coupled device (CCD) or a complementary metal oxide semiconductor (CMOS) imaging device is used to sample three-dimensional objects.

5. The method according to claim 1, wherein The sampling of the three-dimensional object by using an imaging device includes: The three-dimensional object is imaged by using a spherical lens or an aspherical lens and projected onto an imaging device.

6. The method according to claim 1, characterized in that The method of modulating the laser beam using a spatial light modulator loaded with a random hologram comprises: A spatial light modulator loaded with a random hologram is used to modulate the laser beam emitted by a diode laser or a solid-state laser.

7. A holographic three-dimensional imaging system, characterized in that: include: A spatial light modulator, used to sequentially load multiple random holograms and modulate the laser beam before irradiating it onto a three-dimensional object; An image sensor, configured to sample a holographic image of the three-dimensional object; a processor configured to obtain a random pattern formed by the spatial light modulator each time it loads a random hologram and modulates the laser beam, and to construct an observation matrix based on the random patterns formed multiple times; the processor further configured to obtain a holographic image obtained by the image sensor each time it performs holographic image sampling, and to construct an observation vector based on the multiple holographic images; and the processor further configured to reconstruct a three-dimensional holographic image corresponding to the three-dimensional object based on the observation vector, the observation matrix, and a perception compression reconstruction algorithm; The reconstructing of the three-dimensional holographic image corresponding to the three-dimensional object according to the observation vector, the observation matrix and the perceptual compression reconstruction algorithm includes: The holographic images obtained by sampling the holographic images of the three-dimensional object M times are used as vector elements to construct an observation vector with a dimension of M; The random patterns used in each holographic image sampling of the three-dimensional object are arranged into a sampling vector of length N, and the sampling vectors corresponding to the M times of holographic image sampling of the three-dimensional object are arranged into a measurement matrix of dimension M*N; Reconstructing a three-dimensional holographic image corresponding to the three-dimensional object based on the observation vector, the observation matrix and the compressed sensing reconstruction algorithm; The reconstructing of the three-dimensional holographic image corresponding to the three-dimensional object based on the observation vector, the observation matrix and the compressed sensing reconstruction algorithm includes: Determine a sparse domain vector of the three-dimensional holographic image based on an observation vector, an observation matrix, and a compressed sensing reconstruction algorithm; wherein each element of the observation vector is an image, and each element of the sparse domain vector is a matrix; The step of determining the sparse domain vector of the three-dimensional holographic image based on the observation vector, the observation matrix, and the compressed sensing reconstruction algorithm includes: The sparse domain vector is determined according to the following formula: Wherein, x1 represents the sparse domain vector of the three-dimensional holographic image, Indicates taking the minimum value of the function, express norm, This means that the solution to x1 is limited to , where x is the three-dimensional object information, Φ is the observation matrix, and y is the observation vector.

Images