CN-122016048-A - Optical fiber hyperspectral imaging system and method based on single-pixel computational imaging
Abstract
The invention discloses an optical fiber hyperspectral imaging system and an imaging method based on single-pixel computing imaging, the system comprises a structured light projection module, an imaging object, a compound eye type optical fiber probe, an optical fiber spectrum detection module and a data acquisition and processing module. According to the invention, single-pixel calculation imaging and optical fiber spectrum detection are combined, and the combined acquisition of space information and spectrum information is realized on the premise of not changing the internal structure of the optical fiber spectrum detection module, so that the traditional optical fiber spectrometer has high spectrum imaging capability. By introducing a compound eye type optical fiber probe structure, a plurality of optical fiber subunits are combined according to a preset space layout, so that different optical fiber subunits have differences in space orientations, target optical signals from different space directions are respectively received, and multi-directional optical signals are converged and received under a single-point spectrum detection condition, so that the equivalent optical signal receiving field of view of the system is expanded. Meanwhile, by combining a single-pixel calculation imaging mechanism, the spectral region and the space reconstruction are realized under the sampling rate lower than the full space sampling condition, so that the data acquisition amount is reduced and the data redundancy is reduced.
Inventors
- JI ZHONG
- Xing Jingyang
- LIU YUJIN
- CHEN XUELI
Assignees
- 西安电子科技大学广州研究院
Dates
- Publication Date
- 20260512
- Application Date
- 20260203
Claims (9)
- 1. A single pixel computational imaging-based fiber optic hyperspectral imaging system, comprising: The structured light projection module (1) is used for projecting a structured light field changing with time to the imaging object (2) and carrying out spatial modulation on a target scene; The compound eye type optical fiber probe (3) comprises a plurality of optical fiber subunits (4), wherein each optical fiber subunit (4) is combined according to a preset space layout to form a compound eye-like structure, and different optical fiber subunits have differences in space orientation, so that the different optical fiber subunits respectively correspond to different incident angle receiving directions and are used for carrying out non-imaging convergent receiving on target optical signals from different space directions under the single-point spectrum detection condition of keeping a single spectrum detection channel unchanged, so that the effective optical signal receiving view field of the system is expanded from the receiving view angle of a single optical fiber to the combined range of the receiving view angles of a plurality of optical fibers; The optical fiber spectrum detection module (5) is connected with the compound eye type optical fiber probe (3) and is used for carrying out spectrum integration and measurement on the received target optical signal and outputting spectrum intensity information as a single-point spectrum detection unit; The data acquisition and processing module (6) is used for controlling the structured light projection module (1) to load and output a preset spatial modulation sequence, synchronously processing spectrum intensity information corresponding to different spatial modulation states at a sampling rate lower than a full-space sampling condition, and realizing spectral region and spatial reconstruction of an imaging object (2) through a single-pixel calculation imaging algorithm by utilizing a corresponding relation between a multi-directional light signal superposition result and the spatial modulation sequence under a single spectrum intensity output condition formed by the compound eye type optical fiber probe, so that data redundancy generated in a hyperspectral imaging process is reduced.
- 2. The optical fiber hyperspectral imaging system based on single-pixel computing imaging as claimed in claim 1, wherein the receiving visual angles of different optical fiber subunits (4) in the compound eye type optical fiber probe (3) are at least partially different or partially overlapped, so that each optical fiber subunit corresponds to different space receiving directions respectively, and the convergence receiving of target optical signals from a plurality of space directions is realized.
- 3. The optical fiber hyperspectral imaging system based on single-pixel computing imaging according to claim 1 or 2, wherein the structural light projection module (1) is a projection type structural light projection device, such as a digital micromirror array or a liquid crystal spatial light modulator, and the spatial modulation sequence loaded by the structural light projection module (1) is a random matrix, a Hadamard matrix or a Fourier matrix.
- 4. The optical fiber hyperspectral imaging system based on single-pixel computing imaging as claimed in claim 1, wherein the optical fiber spectrum detection module (5) is a commercial optical fiber spectrometer, and the internal optical structure of the optical fiber spectrum detection module is not required to be changed.
- 5. The optical fiber hyperspectral imaging system based on single-pixel computing imaging as set forth in claim 1, wherein the optical fiber subunit (4) is a multimode optical fiber and is combined in a concentric distribution, an array distribution or an irregular distribution manner so as to realize the convergence reception of multi-angle optical signals under the condition of keeping a single-point spectrum detection structure unchanged.
- 6. The optical fiber hyperspectral imaging system based on single-pixel computing imaging as claimed in claim 1, wherein the optical fiber spectral detection module (5) is adjustable in spectral resolution to meet hyperspectral imaging requirements of different imaging objects.
- 7. The optical fiber hyperspectral imaging system based on single-pixel computing imaging as claimed in claim 1, wherein the system can cover a wide spectral range from ultraviolet to near infrared, and is suitable for hyperspectral imaging scenes such as plant leaf classification, medical tissue detection, industrial material analysis and the like.
- 8. A fiber hyperspectral imaging method based on single-pixel computational imaging is characterized in that: the method comprises the steps of 1, projecting a structured light field changing with time to an imaging object (2) by using a structured light projection module (1), and carrying out spatial modulation on the imaging object (2); The optical fiber hyperspectral imaging method based on single-pixel computing imaging of an imaging object (2) is received through a compound eye type optical fiber probe (3), and is characterized in that the low sampling rate is set according to the number of colors or the number of spectrum areas of the imaging object (2), so that the data acquisition amount required in the hyperspectral imaging process is reduced, the data redundancy is reduced, and the imaging efficiency is improved. Reflected light or transmitted light generated under the irradiation of the light field, and transmitting the received light signal to an optical fiber spectrum detection module (5); Step 3, setting a sampling rate according to the number of colors or the number of spectrum areas of the imaging object (2); And 4, combining the acquired spectrum intensity information with a corresponding spatial modulation sequence, and carrying out spectrum distinguishing and spatial reconstruction on the imaging object (2) through a single-pixel calculation imaging algorithm.
- 9. The method for optical fiber hyperspectral imaging based on single pixel computing imaging as set forth in claim 8, wherein the low sampling rate is set according to the number of colors or the number of spectral regions of the imaged object (2) to reduce the amount of data acquisition required in the hyperspectral imaging process, thereby reducing data redundancy and improving imaging efficiency.
Description
Optical fiber hyperspectral imaging system and method based on single-pixel computational imaging Technical Field The invention belongs to the technical field of hyperspectral imaging, and particularly relates to an optical fiber hyperspectral imaging system and method based on single-pixel computing imaging. Background Hyperspectral imaging is an imaging technology combining two modules of imaging and spectroscopy, can acquire spectral features and space images at the same time, and has the advantage of map integration. The hyperspectral imaging technology can reveal information such as chemical components, physiological states and the like of a target object by capturing reflection or radiation characteristics of the target object under different wavelengths, so that the hyperspectral imaging technology is widely applied to the fields such as remote sensing, agriculture, medical imaging and the like. The optical fiber spectrometer has the advantages of small volume, high spectral resolution, wide band coverage range and the like, and the optical signal to be detected is generally introduced into the spectral analysis module through the optical fiber to detect the spectral information of the target. However, the optical fiber spectrometer can only acquire single-point spectrum information, lacks spatial resolution capability, and is difficult to meet the imaging requirement of simultaneously acquiring the spatial distribution and the spectral characteristics of a target. In order to achieve the joint acquisition of spatial information and spectral information, conventional hyperspectral imaging systems generally employ push-broom, point-broom or snapshot imaging structures, and hyperspectral data cubes are constructed by an area array detector, a scanning mechanism or a spectral encoding device. However, the system has the problems of complex structure, large volume, high manufacturing and maintenance cost and the like, needs to collect a large amount of spectrum data in the imaging process, is easy to generate data redundancy, has low imaging efficiency, and has certain limitations on imaging efficiency and signal to noise ratio under the condition of weak light or specific wave bands. In recent years, a computational imaging technique based on a single point detector has received attention. The technology can realize imaging by combining a structured modulation and calculation reconstruction algorithm without relying on an area array detector with spatial resolution capability, and has potential advantages in the aspects of system simplification and dim light imaging. However, the related art approaches focus on the construction of independent imaging systems, and it is difficult to integrate directly with the mature commercial fiber-optic spectrometers. Meanwhile, when a single optical fiber is adopted as an optical signal receiving unit, the effective imaging view field range is still limited due to the limitation of the numerical aperture and the receiving view angle of the optical fiber, and the application requirement of large-view-field hyperspectral imaging is difficult to meet. Disclosure of Invention In order to solve the problems that an optical fiber spectrometer can only acquire single-point spectrum information, an imaging view field is limited, and a traditional hyperspectral imaging system is complex in structure, high in cost and large in data redundancy in the prior art, the invention provides an optical fiber hyperspectral imaging system and an imaging method based on single-pixel computing imaging. According to the invention, by utilizing a single-pixel calculation imaging principle and combining a non-planar array single-point spectrum detection mode, the combined acquisition of space information and spectrum information is realized on the premise of not changing the internal optical structure of the optical fiber spectrum detection module, so that the traditional optical fiber spectrometer has high spectrum imaging capability. By adopting a mode of combining single-point detection and calculation reconstruction, the invention avoids dependence on a high-density array detector and a complex optical light splitting component, effectively reduces the complexity of a system structure and the realization cost, and simultaneously improves the light signal utilization efficiency under the condition of weak light by utilizing a spectrum integration mode, thereby being beneficial to improving the imaging signal to noise ratio. In addition, by combining the sparse characteristic of the target spectrum information, the spectrum region and the space reconstruction meeting the imaging requirement can be realized under the sampling rate lower than the sampling rate of the full space, so that the data acquisition amount and the data redundancy generated in the hyperspectral imaging process are reduced. Furthermore, the invention introduces a compound eye type optical fiber probe structure, and comb