Special issue—Single-pixel imaging
2021, 50(12): 20211061.
doi: 10.3788/IRLA20211061
Single-pixel imaging is a new imaging technique which can obtain image information through a single-pixel detector without the ability of spatial resolution. Compared with the traditional "what you see is what you get" imaging technique, single-pixel imaging has a series of advantages, such as high sensitivity, anti-interference and high resolution. It has shown broad application prospects in remote sensing detection, national defense military, biomedicine and other fields. In recent years, with the improvement of computing capability and the development of optoelectronic devices, single-pixel imaging has attracted more and more attention from researchers. The history of single-pixel imaging was briefly reviewed and the research progress of single-pixel imaging in basic principles, modulation matrix design and image reconstruction algorithms was introduced in detail. In addition, some application scenarios and prospect development trends in the future were summarized.
2021, 50(12): 20210819.
doi: 10.3788/IRLA20210819
Possessing high resolution and rich information, optical imaging is one of the most important techniques for people to obtain information. Photons are information carriers in optical imaging systems. The high-quality reconstruction of optical image depends on the efficient coupling of signal photons and the accurate decoupling of optical information. However, in important application scenarios such as remote sensing or biological imaging, due to the long operating distance or low radiation power, the number of signal photons from the object to the detection plane is small, and the signal-to-noise ratio is low, thus bringing great difficulties to the design of optical system, the signal detection and the image reconstruction, and seriously limiting the performance of optical imaging. How to obtain high-quality images under extremely weak light conditions is not only a basic problem of photoelectric imaging system research, but also a key technology to promote the vigorous development of optical imaging with a larger field of view, longer working distance and higher information flux. In recent years, with the support of light field modulation and quantum detection technology, and based on the high-order classical/quantum correlation of light field, ghost imaging has brought new opportunities for the development of optical imaging technology under extremely weak light conditions, due to its high detection sensitivity and strong ability against interference. This paper briefly reviewed the principle and mechanism of ghost imaging, and systematically introduced the schemes and methods of ghost imaging under very weak light conditions. The physical essence of these methods from the level of photon dynamics was introduced, the capability limits of these methods were discussed, and the applicable scenarios of these methods were compared.
Compared with traditional imaging that uses the first-order correlation of light fields to realize the one-to-one correspondence between the object and image spaces, ghost imaging realizes the correspondence based on the second-order correlation of light fields and thereby obtains object’s image information. By introducing light field fluctuation modulation and computational reconstruction, ghost imaging not only has higher information acquisition efficiency, but also improves the flexibility of image information acquisition modes, thus having imaging capabilities that traditional imaging does not have. The further developments of ghost imaging in system optimization and technical applications bring new demands and challenges to ghost imaging theory. In this paper, the recent research progress of ghost imaging theory in Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences was introduced from three aspects: the physical nature, the image information acquisition theory and theoretical resolution. And the future theoretical research on ghost imaging was prospected.
2021, 50(12): 20211060.
doi: 10.3788/IRLA20211060
In response to the demands of fast real-time positioning of a moving object, a single-pixel fast moving object positioning method using geometric moment detection was proposed. The key to the approach was to locate the moving object by detecting the centroid of the moving object. According to the properties of geometric moments, three geometric moment illumination lights were constructed and illuminated the moving object. A single-pixel detector was used to collect the intensities of the reflected or transmitted light after the moving object interacted with the modulated light. According to the theory of single-pixel imaging, the detected intensity values corresponded to the zero-order and first-order geometric moment values. The centroid that identified the object position parameter could be obtained by corresponding calculations from the zero-order and first-order geometric moment values. The proposed method could reach an approximately frame rate of 500 fps and 1 000 fps to position a moving object without imaging. The error of the centroid obtained by the proposed method was within 1.63 pixels, and the mean square error was 0.118 3 pixels. The proposed method provided a new way for tracking of a fast-moving object.
2021, 50(12): 20210738.
doi: 10.3788/IRLA20210738
Correlated imaging, as a novel computational imaging technology, uses a single pixel detector without spatial resolution capability and combines with spatial light modulation technology to reconstruct two-dimensional spatial information of targets by correlation algorithm. It has been a research topic of widespread concern for two decades. Single pixel detector and structured light modulator are two core elements in correlated imaging, and their performance directly determines the property of correlated imaging. Single pixel detectors often have a very high spectral response range and working bandwidth. In these respects, structured light modulators rarely match the performance of detectors. Therefore, to some extent, the renewal process of the structured light modulator determines the development of the correlated imaging technology. So far, the common structured light modulators used in correlated imaging include ground-glass, spatial light modulator, LED array and masks. Among them, masks, which have been used as a structured light modulator with a long history, are still an important choice for spatial modulation in correlated imaging and play an irreplaceable role. This paper started with the basic concept and development process of correlated imaging, analysed the working principles and application prospects of some existing correlated imaging technologies based on mask modulation, and briefly summarized the work of mask-modulated correlation imaging in non-optical wavebands.
2021, 50(12): 20210723.
doi: 10.3788/IRLA20210723
Unlike conventional digital cameras that used array detectors, single-pixel cameras used single-pixel detectors without spatial resolution to image targets. Due to its wide operating wavelength coverage and high sensitivity, single-pixel cameras are more advantageous than ordinary cameras in special scenarios such as special wavelengths and low light illumination, and are widely used in remote sensing detection, microscopic imaging, military reconnaissance, and other fields. A compact single-pixel imaging system was proposed, which used the working characteristics of the digital micro-mirror array to form a symmetrical folded double optical path to quickly complete differential measurement. The compact structure reduced the size of the camera system and allowed the system to use a standard Nikon lens as the imaging lens. The system had various modes such as dual-optical differential imaging, dual-optical averaging noise reduction, broad-spectrum band imaging, and dual-optical alternate sampling, and the application modes could be switched according to the different needs of the scene for reconstructed image signal-to-noise ratio, real-time frame rate, and imaging band. The experimental results based on the system prototype showed that it could realize the expected functions and achieve the corresponding performance. The proposed compact single-pixel imaging system was a successful engineering attempt of single-pixel imaging, which laid a good technical and engineering foundation for the subsequent practical application of single-pixel imaging technology.
2021, 50(12): 20210735.
doi: 10.3788/IRLA20210735
In traditional imaging systems, a two-dimensional pixel array detector is usually used to record the object’s image. However, in the scheme of single-pixel imaging, only a bucket detector without spatial resolution capability is needed for signal measurement. Then one can reconstruct the image with different algorithms on a computer. Compared with pixel array detector, the manufacturing cost is relatively low for a bucket detector, so the single-pixel imaging technique attracts a great deal of researchers, especially in the wavebands of X ray, near infrared imaging and terahertz imaging. Moreover, extensive attention of single-pixel imaging has also focused on the applications of fluorescence imaging, multi-spectral imaging, 3D imaging and complex wavefront imaging. Especially in the field of complex wavefront imaging, because it is important to the applications of astronomy, medical diagnosis and optical measurement. Researchers have proposed a large number of single-pixel imaging schemes to measure an unknown complex amplitude field, and these works may help the single-pixel imaging technique be used in different practical scenes. In this review, the development and basic principle of single-pixel imaging technique were introduced. Further, the application of single-pixel imaging in the field of complex amplitude measuring was reviewed and discussed detailly.
2021, 50(12): 20210734.
doi: 10.3788/IRLA20210734
Fourier-transform ghost imaging (FGI) is an imaging method which exploits the high-order correlation characteristics of optical fields to extract the Fourier information of samples. Due to its low requirement for the coherence of the light source, it provides a new technical approach for miniaturizing and high-resolution X-ray microscopy. However, in practice, limited X-ray flux is often required to reduce radiation damage to the sample, and the existence of Compton scattering will reduce the signal-to-noise ratio when X-ray photons interact with the sample. To solve these problems, X-ray FGI with limited flux was studied by simulation. The results showed that when the detection flux was 0.1 PHS/ pixel, the amplitude and phase information of the sample could still be obtained. The Geant4 Monte Carlo simulation program was adopted to analyze the influence of Compton scattering noise generated by the gold, silicon and hemoglobin samples in X-ray FGI. The results indicated that FGI could achieve better Compton scattering noise resistance than traditional X-ray diffraction imaging.
2021, 50(12): 20210790.
doi: 10.3788/IRLA20210790
Originated from the optical intensity ghost imaging, microwave coincidence imaging breaks through the limitation of antenna aperture on imaging resolution by space and time-varying radiation mode through the modulation of electromagnetic waves. It has the advantages of forward-looking, staring and fast-shooting imaging, and has broad application prospects in the fields of staring observation in key areas, autonomous sensing of unmanned systems, security inspection and security protection and so on. The technical origin of microwave coincidence imaging was briefly described, and its current research status and main progress from three aspects including the imaging principle, imaging methods and imaging systems were summarized. Through the analysis of the imaging principle, the basic conditions for coincidence imaging and the influencing factors of imaging resolution were clarified. Through the review of imaging methods, the differences and relationships among microwave coincidence imaging, optical ghost imaging and conventional microwave imaging methods were analyzed. Through the introduction of imaging systems, the features and differences among various systems such as random radiation, wavefront modulation and aperture encoding were compared, which made the development of microwave coincidence imaging easier to perceive. Finally, the future development trend of microwave coincidence imaging was summarized and prospected.
2021, 50(12): 20210717.
doi: 10.3788/IRLA20210717
Terahertz imaging technology has the advantages of perspective, safety and high spectral resolution, and has a broad applications prospect. Since terahertz detector arrays have low technological maturity and are of high cost, terahertz imaging techniques based on single-point scanning have been prevailing for many years, and have problems such as complex systems and long imaging time. Recently, computational imaging techniques have emerged as exciting approaches for achieving efficient terahertz single-pixel imaging. The basic principles, the experimental realizations, and application prospects of terahertz single-pixel computational imaging techniques were described. Some key challenges and potential solutions were discussed.
2021, 50(12): 20210657.
doi: 10.3788/IRLA20210657
Correlated imaging is a kind of novel imaging technology which can reconstruct the target information on the optical path through intensity correlation calculation. In complex environments, turbulence, scattering and other factors will affect the measurement of signal light in the form of noise. Since the signal light in correlated imaging is measured by bucket detector, it is necessary to systematically study the processing of noise-including signal light and its influence on the process of correlated imaging. Under this background, this paper systematically reviewed the related research of our research group. Firstly, the correlated imaging model in complex environment was briefly introduced. Then, the influence of noise-including signal light on correlated imaging of binary target and gray target was analyzed. Then, the signal noise caused by turbulence in the atmospheric channel was measured and analyzed. Finally, the application of the correlated imaging was prospected.
2021, 50(12): 20211058.
doi: 10.3788/IRLA20211058
As a typical computational imaging modality, single-pixel imaging uses a single-pixel detector to measure the light intensities reflected or transmitted from the target after its interaction with a series of patterns. By calculating the correlation of the measured intensities and relevant patterns with different reconstruction algorithms, the target image can be recovered. Compared with multi-pixel detector (i.e. CCD or CMOS), single-pixel imaging overcomes hardware limitations and the detection efficiency is higher, and the response is faster in some special wavebands. Metasurfaces are a kind of artificial two-dimensional materials consisting of an array of subwavelength metallic or dielectric unit cells. In the optical wavelength regime, the metasurface can display various holograms by adjusting different degrees of freedom of incident light. In the microwave regime, the metasurface can couple with the waveguide and emit various radiating modes as patterns. The research background, imaging principle, reconstruction algorithms of single-pixel imaging, and the research background of metasurface imaging were reviewed. The discussion of relevant works was mainly focused on the combination of single-pixel imaging and metasurface imaging in optical and microwave regimes, and finally a perspective on the potential development in future was proposed.
2021, 50(12): 20210856.
doi: 10.3788/IRLA20210856
Successive classification of fast-moving objects is significant in various fields. However, due to the limited data transmission bandwidth and data storage space, it is challenging to perform fast-moving object classification based on scene photography for a long duration. Inspired by single-pixel imaging and combined with deep learning, a single-pixel fast-moving object classification method based on optical-electronic hybrid neural network was proposed. The proposed method had no need to acquire the images of objects, but obtained the feature information for classification directly by using spatial light modulating and single-pixel detecting. Thus, the massive image data produced by the image-based classification for a long duration was avoided. As part of the neural network, the single-pixel detecting connected optical computing and electronic computing seamlessly, an optical-electronic hybrid neural network for object classification was constructed. The proposed method in classifying fast-moving handwritten digits on a rotating disk was experimentally demonstrated, which passed through the field of view successively. The experiment confirmed that the classification ability of the proposed method had exceeded human vision.
2021, 50(12): 20210724.
doi: 10.3788/IRLA20210724
Combining photon counting technology with single-pixel imaging can achieve highly sensitive and low cost photon counting imaging, but there are problems of long sampling time and long reconstruction time. The compressed sampling and reconstruction network, which is based on deep learning, uses the fully connected layer without the offset and activation function as the measurement matrix, achieves faster and higher quality image reconstruction by learning efficient measurement matrices from the data and avoids the huge amount of calculation caused by traditional iterative algorithms. However, when the fully connected layer is used for block compression sensing of high-resolution images, the reconstructed image will produce block artifact. In response to this problem, overlapping block sampling network (Os_net), nested sampling network (Ns_net), and convolution sampling network (Cs_net) were proposed: to replace fully connected layer sampling. In the design of the reconstructed network, the images were reconstructed by using a linear mapping network. The design experiment shows that Cs_net has the best deblocking effect. After Cs_net binarization is applied to a photon counting single-pixel imaging system, the experiment results show that Cs_net de-blocking effect is significantly better than the traditional algorithm TVAL3, and Cs_net has also achieved good results on the reconstruction quality.
