Ghost imaging based on Y-net: a dynamic coding and decoding approach

Zhu, Rihong; Ye, Hong; Tan, Zhijie; Lu, Rengui; Han, Shensheng; Huang, Zengfeng; Wang, Jian

doi:10.1364/oe.395000

Cited by 28 publications

(8 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 1 shows that the performances of SP-ILC on single objects samples are better than that on the multiple object samples since the single object task is easier to process. The peak signal-to-noise ratio (PSNR) and the structural similarity function (SSIM) are widely used to measure image quality [51,52]. The PSNR is defined by:…”

Section: Concurrent Imaging Location and Classificationmentioning

confidence: 99%

SP-ILC: Concurrent Single-Pixel Imaging, Object Location, and Classification by Deep Learning

et al. 2021

View full text Add to dashboard Cite

We propose a concurrent single-pixel imaging, object location, and classification scheme based on deep learning (SP-ILC). We used multitask learning, developed a new loss function, and created a dataset suitable for this project. The dataset consists of scenes that contain different numbers of possibly overlapping objects of various sizes. The results we obtained show that SP-ILC runs concurrent processes to locate objects in a scene with a high degree of precision in order to produce high quality single-pixel images of the objects, and to accurately classify objects, all with a low sampling rate. SP-ILC has potential for effective use in remote sensing, medical diagnosis and treatment, security, and autonomous vehicle control.

show abstract

Section: Concurrent Imaging Location and Classificationmentioning

confidence: 99%

SP-ILC: Concurrent Single-Pixel Imaging, Object Location, and Classification by Deep Learning

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Interestingly, the convolutional neural network (CNN) can reconstruct target images from 2D measurements by building end-to-end mapping functions to learn data features. For example, U-Net and related improvements have achieved good results in GI image reconstruction tasks [7,8]. In addition, CNN-based image reconstruction algorithms have advantages in extracting the local spatial features of images.…”

Section: Introductionmentioning

confidence: 99%

Improved U-Net3+ With Spatial–Spectral Transformer for Multispectral Image Reconstruction

Chen

Li³

et al. 2023

IEEE Photonics J.

View full text Add to dashboard Cite

Multispectral image reconstruction, which aims to recover a three-dimensional (3D) spatial-spectral signal from a two-dimensional measurement in a spectral camera based on ghost imaging via sparsity constraint (GISC), has been attracting much attention recently. However, faced with abundant 3D spectral data, the reconstruction quality cannot meet the visual requirements. Based on the robust data processing capability of deep learning, a novel network called SSTU-Net3+ is constructed by improving U-Net3+ with a spatial-spectral transformer (SST). To enhance the feature representation of images during reconstruction, mixed pooling modules and new convolution processes are proposed to improve the performance of the encoder and decoder, with U-Net3+ as the backbone. To boost the quality of reconstructed images, with split and concatenate (Concat) operations, we construct SST modules by exploiting both spatial and spectral correlations of multispectral images to refine the spatial and spectral features. Furthermore, we employ the SST in the decoder to reconstruct the desired 3D cube. Given similar network parameters, experiments on GISC spectral imaging data show that, compared to convolutional neural network-based methods, the average peak signal-to-noise ratio of images reconstructed using SSTU-Net3+ is improved by 3%, the structural similarity is enhanced by 3%, and the spectral angle mapping is cut by 12%. Particularly, compared to differential ghost imaging and compressed sensing, the reconstruction quality of SSTU-Net3+ has been significantly improved. SSTU-Net3+ can process a large amount of 3D multispectral image data more efficiently and construct the target image more accurately than the abovementioned methods. 

show abstract

“…To overcome this difficulty, several recent works (e.g. [21][22][23][24][25][26][27][28]) have addressed improving the speed and quality of GI by using deep learning networks. In general, introducing deep learning methods into GI can significantly improve the quality of the reconstructed image.…”

Section: Introductionmentioning

confidence: 99%

“…Besides, in [23,26], the depth neural network was considered to improve the imaging quality and imaging speed, and the sampling ratio was further reduced. In addition, in [28], only one imaging was needed to obtain a clearer reconstructed image, which greatly reduced Single-arm GI setup, where the natural light is adopted as the light source; the aperture is to control the luminous flux, lens-1 and lens-2 are the extended lens and collection lens, respectively, and the bucket detector is to detect and record the total light intensity.…”

Section: Introductionmentioning

confidence: 99%

Single-arm ghost imaging via conditional generative adversarial network

Zhao

Zhang

2021

Laser Phys. Lett.

View full text Add to dashboard Cite

In this study, we develop a single-arm ghost imaging (GI) framework based on a conditional generative adversarial network (cGAN) to improve the image quality and extend the application scenarios of GI. A set of one-dimensional (1D) bucket signals generated by a single-arm GI system and their corresponding ground-truth counterparts is employed to train the cGAN. This allows us to reconstruct a low-noise image from a new 1D bucket signal, while the sequence of random patterns is unnecessary. The results show that the proposed method significantly improves the image quality relative to the conventional GI methods.

show abstract

Ghost imaging based on Y-net: a dynamic coding and decoding approach

Cited by 28 publications

References 45 publications

SP-ILC: Concurrent Single-Pixel Imaging, Object Location, and Classification by Deep Learning

SP-ILC: Concurrent Single-Pixel Imaging, Object Location, and Classification by Deep Learning

Improved U-Net3+ With Spatial–Spectral Transformer for Multispectral Image Reconstruction

Single-arm ghost imaging via conditional generative adversarial network

Contact Info

Product

Resources

About