Analysis and Optimization of Aperture Design in Computational Imaging

Yedidia, Adam; Thrampoulidis, Christos; Wornell, Gregory W.

doi:10.1109/icassp.2018.8462521

Cited by 19 publications

(29 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Among other things, the analysis showed that when shot noise dominates thermal noise, randomly generated masks with lower transmissivity than 1/2 offered greater performance compared to spectrally flat patterns of transmissivity 1/2. This paper extends the work of [14] in multiple respects that may be broadly grouped into the following three main contributions.…”

Section: Introductionmentioning

confidence: 81%

“…Another issue is obtaining a good understanding of mask/lens combinations. This will require not only updates to the simple propagation model studied here and in [14], but also a refined understanding of the cost tradeoffs between lenses and apertures.…”

Section: Discussion and Future Workmentioning

confidence: 99%

“…In light of the increased importance of coded apertures, prior work [14] described a model under which they can be analyzed. This model uses far-field geometric optics to model light propagation and a sensor model that includes thermal and shot noise components.…”

Section: Introductionmentioning

confidence: 99%

“…This model uses far-field geometric optics to model light propagation and a sensor model that includes thermal and shot noise components. Together with mutual information (MI) as a performance metric, [14] compared the classical random on-off apertures [2,3] of varying intensity to the "spectrally flat" patterns with transmissivity 1/2 (same as the URA of [4]). Among other things, the analysis showed that when shot noise dominates thermal noise, randomly generated masks with lower transmissivity than 1/2 offered greater performance compared to spectrally flat patterns of transmissivity 1/2.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Near-optimal Coded Apertures for Imaging via Nazarov’s Theorem

Ajjanagadde

Thrampoulidis

Yedidia

et al. 2019

ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

We characterize the fundamental limits of coded aperture imaging systems up to universal constants by drawing upon a theorem of Nazarov regarding Fourier transforms. Our work is performed under a simple propagation and sensor model that accounts for thermal and shot noise, scene correlation, and exposure time. Focusing on mean square error as a measure of linear reconstruction quality, we show that appropriate application of a theorem of Nazarov leads to essentially optimal coded apertures, up to a constant multiplicative factor in exposure time. Additionally, we develop a heuristically efficient algorithm to generate such patterns that explicitly takes into account scene correlations. This algorithm finds apertures that correspond to local optima of a certain potential on the hypercube, yet are guaranteed to be tight. Finally, for i.i.d. scenes, we show improvements upon prior work by using spectrally flat sequences with bias. The development focuses on 1D apertures for conceptual clarity; the natural generalizations to 2D are also discussed.Index Termscoded aperture cameras, computational photography, optical signal processing, Fourier analysis

show abstract

Section: Introductionmentioning

confidence: 81%

Section: Discussion and Future Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Near-optimal Coded Apertures for Imaging via Nazarov’s Theorem

Ajjanagadde

Thrampoulidis

Yedidia

et al. 2019

ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

show abstract

“…The computation in (14) takes 2M 2 + 2M multiplications. We then feed our gradients to minfunc solver [33] with L-BFGS algorithm [34] to solve the non-linear optimization problem in (7).…”

Section: Algorithm Analysismentioning

confidence: 99%

Image and Depth Estimation with Mask-Based Lensless Cameras

Zheng

Asif

2019

2019 IEEE 8th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP)

View full text Add to dashboard Cite

Mask-based lensless cameras replace the lens of a conventional camera with a customized mask. These cameras can potentially be very thin and even flexible. Recently, it has been demonstrated that such mask-based cameras can recover light intensity and depth information of a scene. Existing depth recovery algorithms either assume that the scene consists of a small number of depth planes or solve a sparse recovery problem over a large 3D volume. Both these approaches fail to recover scene with large depth variations. In this paper, we propose a new approach for depth estimation based on alternating gradient descent algorithm that jointly estimates a continuous depth map and light distribution of the unknown scene from its lensless measurements. The computational complexity of the algorithm scales linearly with the spatial dimension of the imaging system. We present simulation results on image and depth reconstruction for a variety of 3D test scenes. A comparison between the proposed algorithm and other method shows that our algorithm is faster and more robust for natural scenes with a large range of depths.

show abstract