Fourier single-pixel imaging is one of the main single-pixel imaging techniques. To improve the imaging efficiency, some of the recent method typically select the low-frequency and discard the high-frequency information to reduce the number of acquired samples. However, sampling only a small amount of low-frequency components will lead to the loss of object details and will reduce the imaging resolution. At the same time, the ringing effect of the restored image due to frequency truncation is significant. In this paper, a new sparse Fourier single-pixel imaging method is proposed that reduces the number of samples explorations while maintaining increased image quality. The proposed method makes a special use of the characteristics of the Fourier spectrum distribution based on which the power of image information decreases gradually from low to high frequencies in the Fourier space. A variable density random sampling matrix is employed to achieve random sampling with Fourier single-pixel imaging technology, followed by the processing of the sparse Fourier spectra using compressive sensing algorithms to recover the high-quality information of the object. The new algorithm can effectively improve the quality of object restoration comparing with the existing Fourier single-pixel imaging methods which only acquire the low-frequency parts. Additionally, considering that the resolution of the system is diffraction limited, super-resolution imaging can also be achieved. Experimental results demonstrate the mainly correctness but also effectiveness of the proposed method.

Single-pixel imaging has the ability to generate images at nonvisible wavelengths and under low light conditions and thus has received increasing attention in recent years. Fourier single-pixel imaging (FSI) utilizes deterministic basis patterns for illumination to greatly improve the quality of image reconstruction. However, the original FSI based on grayscale Fourier basis illumination patterns is limited by the imaging speed as the digital micro-mirror devices (DMD) used to generate grayscale patterns operate at a low refresh rate. In this paper, a new approach is proposed to increase the imaging speed of DMD-based FSI without reducing the imaging spatial resolution. In this strategy, the grayscale Fourier basis patterns are split into a pair of grayscale patterns based on positive/negative pixel values, which are then decomposed into a cluster of binary basis patterns based on the principle of decimalization to binary. These binary patterns are used to illuminate the imaged object. The resulting detected light intensities multiply the corresponding weighted decomposed coefficients and are summed, and the results can be used to generate the Fourier spectrum for the imaged object. Finally, an inverse Fourier transform is applied to the Fourier spectrum to obtain the object image. The proposed technique is verified by a computational simulation and laboratory experiments. Both static and dynamic imaging experiments are carried out to demonstrate the proposed strategy. 128 × 128 pixels dynamic scenes at a speed of ~9 frames-per-second are captured under 22 KHz projection rate using a DMD. The reported technique accelerates the imaging speed for DMD-based FSI and provides an alternative approach to improve FSI efficiency.

Optics InfoBase is the Optical Society's online library for flagship journals, partnered and copublished journals, and recent proceedings from OSA conferences.

Computational ghost imaging has attracted wide attention from researchers in many fields over the last two decades. Multispectral imaging as one application of computational ghost imaging possesses spatial and spectral resolving abilities, and is very useful for surveying scenes and extracting detailed information. Existing multispectral imagers mostly utilize narrow band filters or dispersive optical devices to separate light of different wavelengths, and then use multiple bucket detectors or an array detector to record them separately. Here, we propose a novel multispectral ghost imaging method that uses one single bucket detector with multiplexed illumination to produce a colored image. The multiplexed illumination patterns are produced by three binary encoded matrices (corresponding to the red, green and blue colored information, respectively) and random patterns. The results of the simulation and experiment have verified that our method can be effective in recovering the colored object. Multispectral images are produced simultaneously by one single-pixel detector, which significantly reduces the amount of data acquisition.

Target tracking has found important applications in particle tracking, vehicle navigation, aircraft monitoring, etc. However, employing single-pixel imaging techniques to track a fast-moving object with a high frame rate is still a challenge, due to the limitation of the modulation frequency of the spatial light modulator and the number of required patterns. Here we report a complementary single-pixel object tracking approach which requires only two geometric moment patterns to modulate the reflected light from a moving object in one frame. Using the complementary nature of a digital micromirror device (DMD), two identical single-pixel detectors are used to measure four intensities which can be used to acquire the values of zero-order and first-order geometric moments to track the centroid of a fast-moving object. We experimentally demonstrate that the proposed method successfully tracks a fast-moving object with a frame rate of up to 11.1 kHz in the first two experiments. In the third experiment, we compare previous works and find that the method can also accurately track a fast-moving object with a changing size and moving speed of 41.8 kilopixel/s on the image plane. The root mean squared errors in the transverse and axial directions are 0.3636 and 0.3640 pixels, respectively. The proposed method could be suitable for ultrafast target tracking.

Single-pixel imaging (SPI), a novel computational imaging technique that has emerged in the past decades, can effectively capture the image of a static object by consecutively measuring light intensities from it. However, when SPI is applied to imaging the dynamic object, severe motion blur in the restored image tends to appear. In this Letter, a new SPI scheme is proposed to largely alleviate such a problem by leveraging a calibrated radon spectrum. Such a spectrum is obtained by translating the acquired one-dimensional projection functions (1DPFs) according to the positional relationship among the 1DPFs. Simulation and experimental results demonstrate that, without prior knowledge, our approach can effectively reduce motion blur and restore high-quality images of the fast-moving object. In addition, the proposed scheme can also be used for fast object tracking.