Fast Compressive 3D Single-pixel Imaging
3
Citation
5
Reference
10
Related Paper
Citation Trend
Abstract:
In this work, we demonstrate a modified photometric stereo system with perfect pixel registration, capable of reconstructing continuous real-time 3D video at ~8 Hz for 64 × 64 image resolution by employing evolutionary compressed sensing.Keywords:
Stereo imaging
Hyperspectral image super resolution (SR) reconstruction has been studied widely and many algorithms have been proposed. In this paper, a novel super resolution reconstruction method was designed by employing a joint spectral-spatial sub-pixel mapping model which aims to obtain the probabilities of sub-pixels to belong to different land cover classes by dividing mixed pixels into several sub-pixels. Given these sub-pixel probabilities, the resolution enhanced image can be further generated. The proposed approach has been evaluated using both synthetic and real hyperspectral images and compared with other well-known methods. The visual and quantitative comparisons confirm the effectiveness of the proposed method.
Sub-pixel resolution
Full spectral imaging
Land Cover
Cite
Citations (11)
The consequences of changes in spatial resolution for application of thermal imagery in plant phenotyping in the field are discussed. Where image pixels are significantly smaller than the objects of interest (e.g., leaves), accurate estimates of leaf temperature are possible, but when pixels reach the same scale or larger than the objects of interest, the observed temperatures become significantly biased by the background temperature as a result of the presence of mixed pixels. Approaches to the estimation of the true leaf temperature that apply both at the whole-pixel level and at the sub-pixel level are reviewed and discussed.
Sub-pixel resolution
Cite
Citations (84)
Compressed Sensing is a new sampling theorem,it points out that if a signal can be compressed under some conditions,that a very accurate reconstruction can be obtained from a relatively small number of non-traditional samples.On the basis of compressed sensing,the paper presents multiscale compressed sensing.The numerical experiments demonstrate that multiscale compressed sensing can give better quality reconstruction than a literal deployment of the compressed sensing methodology.
Signal reconstruction
SIGNAL (programming language)
Cite
Citations (0)
This paper presents a novel approach, compressive mobile sensing, to use mobile sensors to sample and reconstruct sensing fields based on compressive sensing. Compressive sensing is an emerging research field based on the fact that a small number of linear measurements can recover a sparse signal without losing any useful information. Using compressive sensing, the signal can be recovered by a sampling rate that is much lower than the requirements from the well-known Shannon sampling theory. The proposed compressive mobile sensing approach has not only the merits of compressive sensing, but also the flexibility of different sampling densities for areas of different interests. A special measurement process makes it different from normal compressive sensing. Adopting importance sampling, compressive mobile sensing enables mobile sensors to move adaptively and acquire more samples from more important areas. A motion planning algorithm is designed based on the result of sparsity analysis to locate areas of more interests. At last, experimental results of 2-D mapping are presented as an implementation compressive mobile sensing.
SIGNAL (programming language)
Cite
Citations (5)
The compressive sensing theory has been successfully applied to image compression in the past few years. Recently, deep network-based compressive sensing image reconstruction algorithms have been proposed, which reduce the computational complexity compared with traditional iterative reconstruction algorithms. But most of those are patch-based reconstruction methods, which leads to blocky artifacts for the full image assembled by patch reconstruction. In this paper, we propose a novel image reconstruction network (CSReNet) from patch compressive sensing measurements. Different from other deep network-based algorithms, our network can not only recovery image from patch compressive sensing measurements, also remove the blocky artifacts. There are two modules, reconstruction module and removal module in our network. Experimental results on test data show that our proposed network outperforms several compressive sensing reconstruction algorithms with patch-based CS measurements.
Reconstruction algorithm
Cite
Citations (7)
Compressed sensing theory provides a new approach to acquire data as a sampling technique and makes sure that an original sparse signal can be reconstructed from few measurements. The construction of compressed sensing matrices is a central problem in compressed sensing theory. In this paper, the deterministic compressed sensing matrices with characters of finite fields are constructed and the coherence of the matrices are computed. Furthermore, the maximum sparsity of recovering the original sparse signals by using our compressed sensing matrices is obtained. Meanwhile, a comparison is made with the compressed sensing matrices constructed by DeVore based on polynomials over finite fields. In the numerical simulations, our compressed sensing matrix outperforms DeVore’s matrix in the process of recovering original sparse signals.
Mutual coherence
Matrix (chemical analysis)
SIGNAL (programming language)
Restricted isometry property
Cite
Citations (0)
Limited by the spatial resolution of hyperspectral satellites, mixed pixels are widely existed in remote sensing data. It is a hot spot in the field of remote sensing on using the proportions of different land covers to improve the spatial resolution of hyperspectral images. Sub-pixel mapping (SPM) is an effective means to further explore the spatial distribution of different land covers in mixed pixels. The sub-pixel mapping method based on BP Neural Network is one of the effective methods. It used proportion data and classification of different sub-pixels in geometrical shapes as the training data to train the neural network. The trained model can be used to optimize the spatial resolution of real land image. However, the BP Neural Network model does not take spatial correlation into account. This paper proposed a sub-pixel mapping method based on BPNN and improved sub-pixel swapping model (BPNN_IPSM). The artificial image and real land image taken by Landsat8 were used to be tested. Experiments and comparisons showed that the BPNN_IPSM presented in this paper is an efficient approach in sub-pixel mapping.
Spatial correlation
Cite
Citations (0)
We present a quantitative evaluation of a Maximum a Posteriori image reconstruction (USC-MAP) implemented on the Concorde microPET-R4 camera. This iterative reconstruction algorithm uses a Bayesian reconstruction technique for the reconstruction of PET images which includes an accurate modeling of the camera response, the Poisson distribution of coincidence data, and the physics of positron decay. The spatial resolution measurements were made by using a small cylindrical chamber containing a thin plastic tube filled with concentrated /sup 18/F-FDG, scanned at different radial positions in the field of view. A specially designed miniature rat heart phantom and a cylindrical chamber containing four different size spheres were used to measure recovery coefficients mimicking cardiac and small lesion quantitative applications. The images were reconstructed with FORE+2D-FBP, 2D-OSEM, 3DRP, and USC-MAP. The experiments show that substantial gain in spatial resolution is achieved using this novel image reconstruction algorithm. In particular, we showed that the spatial resolution degradation at large radial offsets due to the depth of interaction is in large part corrected with USC-MAP, a consequence of the accurate system detection modeling. The gain in resolution results in recovery of activity concentration in hot spheres is achieved with USC-MAP as compared to FORE-2D-FBP. The recovery coefficient increases by 9% in spheres with diameter of 1.24 cm and 62% in spheres with diameter of 0.39 cm in the experiment with /sup 18/F-FDG. More significant improvement is observed in the case of long positron range isotope such as /sup 76/Br.
Reconstruction algorithm
Field of view
Cite
Citations (5)
The color X-ray camera SLcam(R) is a full-field, single photon detector providing scanning free, energy and spatially resolved X-ray imaging. Spatial resolution is achieved with the use of polycapillary optics guiding X-ray photons from small regions on a sample to distinct energy dispersive pixels on a CCD. Applying sub-pixel resolution, signals from individual capillary channels can be distinguished. Accordingly the SLcam(R) spatial resolution can be released from pixel size being confined rather to a diameter of individual polycapillary channels. In this work a new approach to sub-pixel resolution algorithm comprising photon events also from the pixel centers is proposed. The details of the employed numerical method and several sub-pixel resolution examples are presented and discussed.
Sub-pixel resolution
Sample (material)
Cite
Citations (17)
Super-resolution mapping (SRM) is a technique for exploring spatial distribution information of the land-cover classes at finer spatial resolution. The soft-then-hard super-resolution mapping (STHSRM) algorithm is a type of SRM algorithm that first estimates the soft class values for sub-pixels at the target fine spatial resolution and then predicts the hard class labels for sub-pixels. The sub-pixel shifted images from the same area can be incorporated to improve the accuracy of STHSRM algorithm. In this article, multiscale sub-pixel shifted images (MSSI) based on the fine-scale model and the coarse-scale model are utilized to increase the accuracy of STHSRM. First, class fraction images are derived from multiple sub-pixel shifted coarse spatial resolution images by soft classification. Then using the sub-pixel/sub-pixel spatial attraction model as fine-scale and the sub-pixel/pixel spatial attraction model as coarse scale, all MSSI can be derived from fraction images. The MSSI for each class are then integrated to obtain the desired fine spatial resolution images. Finally, the integrated fine spatial resolution images are used to allocate classes for sub-pixel. Experiments on two synthetic remote sensing images and a real hyperspectral remote sensing imagery show that the proposed method produces higher mapping accuracy result.
Sub-pixel resolution
Cite
Citations (21)