Single photon counting compressive imaging using a generative model optimized via sampling and transfer learning.

Single photon counting compressive imaging, a combination of single-pixel-imaging and single-photon-counting technology, is provided with low cost and ultra-high sensitivity. However, it requires a long imaging time when applying traditional compressed sensing (CS) reconstruction algorithms. A deep-learning-based compressed reconstruction network refrains iterative computation while achieving efficient reconstruction. This paper proposes a compressed reconstruction network (OGTM) based on a generative model, adding sampling sub-network to achieve joint-optimization of sampling and generation for better reconstruction. To avoid the slow convergence caused by alternating training, initial weights of the sampling and generation sub-network are transferred from an autoencoder. The results indicate that the convergence speed and imaging quality are significantly improved. The OGTM validated on a single-photon compressive imaging system performs imaging experiments on specific and generalized targets. For specific targets, the results demonstrate that OGTM can quickly generate images from few measurements, and its reconstruction is better than the existing compressed sensing recovery algorithms, compensating defects of the generative models in compressed sensing.

Deep-learning-based single-photon-counting compressive imaging via jointly trained subpixel convolution sampling.

The combination of single-pixel-imaging and single-photon-counting technology can achieve ultrahigh-sensitivity photon-counting imaging. However, its applications in high-resolution and real-time scenarios are limited by the long sampling and reconstruction time. Deep-learning-based compressive sensing provides an effective solution due to its ability to achieve fast and high-quality reconstruction. This paper proposes a sampling and reconstruction integrated neural network for single-photon-counting compressive imaging. To effectively remove the blocking artefact, a subpixel convolutional layer is jointly trained with a deep reconstruction network to imitate compressed sampling. By modifying the forward and backward propagation of the network, the first layer is trained into a binary matrix, which can be applied to the imaging system. An improved deep-reconstruction network based on the traditional Inception network is proposed, and the experimental results show that its reconstruction quality is better than existing deep-learning-based compressive sensing reconstruction algorithms.

A binary sampling Res2net reconstruction network for single-pixel imaging.

The traditional algorithm for compressive reconstruction has high computational complexity. In order to reduce the reconstruction time of compressive sensing, deep learning networks have proven to be an effective solution. In this paper, we have developed a single-pixel imaging system based on deep learning and designed the binary sampling Res2Net reconstruction network (Bsr2-Net) model suitable for binary matrix sampling. In the experiments, we compared the structural similarity, peak signal-to-noise ratio, and reconstruction time using different reconstruction methods. Experimental results show that the Bsr2-Net is superior to several deep learning networks recently reported and closes to the most advanced reconstruction algorithms.

Large-area single photon compressive imaging based on multiple micro-mirrors combination imaging method.

We demonstrate a single photon compressive imaging system with the image plane up to the entire digital micro-mirror device (DMD) work area. A parallel light source is designed to reduce the influence of light scattering on imaging resolution and a photon counting photomultiplier tube (PMT) with a large photosensitive area is used to effectively collect light reflected from the full screen of DMD. A control and counting circuit, based on Field-Programmable Gate Array (FPGA), is developed to load binary random matrix into the DMD controller for each measurement, and to count single-photon pulse output from PMT simultaneously. To reduce imaging time and huge memory occupation for image reconstruction, a multiple micro-mirrors combination imaging method is proposed. The signal-to-noise ratio and detection limit of the imaging system is theoretically deduced. Theoretical analysis and experimental results show that micro-mirrors combination imaging method is more suitable for faster imaging in a weaker-light-level environment. In order to achieve high imaging quality, the size of the combined pixels and the average time of each measurement should be moderate, so that the impact of Poisson shot noise is minimized.