Performance estimation of a real-time rosette imager

Stoltz, Gene; Stoltz, Marnus

doi:10.1117/12.2575060

Cited by 4 publications

(6 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…5 Using the Rosette Imager, it has been shown that non-grid based sampling can match the performance of a traditional staring array (grid-based sampling). 3 These results are not unexpected as compressed sensing require signals to be sparse in the sampling domain and natural scenes are not sparse in the spatial domain.…”

Section: Introductionmentioning

confidence: 94%

“…The regularisation algorithm is chosen to support real-time reconstruction from structured sampling and thus the Fourier Domain Regularisation Inversion (FDRI) algorithm is an adequate choice. 3 The FDRI algorithm only approximates an l1-optimisation algorithm but does match the performance of Nesterov's algorithm (NESTA) and Basis Pursuit Denoising (BPDN). 9 It is noted that in the the paper an output image size of 64x64 will be used with 19.1% sampling, equating to 784 samples.…”

Section: Compressed Sensingmentioning

confidence: 99%

See 1 more Smart Citation

Improving spatial domain based image formation through compressed sensing

Stoltz¹,

Nel²

2020

Preprint

Self Cite

View full text Add to dashboard Cite

In this paper, we improve image reconstruction in a single-pixel scanning system by selecting an detector optimal field of view. Image reconstruction is based on compressed sensing and image quality is compared to interpolated staring arrays. The image quality comparisons use a "dead leaves" data set, Bayesian estimation and the Peak-Signal-to-Noise Ratio (PSNR) measure.Compressed sensing is explored as an interpolation algorithm and shows with high probability an improved performance compared to Lanczos interpolation. Furthermore, multi-level sampling in a single-pixel scanning system is simulated by dynamically altering the detector field of view. It was shown that multi-level sampling improves the distribution of the Peak-Signal-to-Noise Ratio.We further explore the expected sampling level distributions and PSNR distributions for multi-level sampling. The PSNR distribution indicates that there is a small set of levels which will improve image quality over interpolated staring arrays. We further conclude that multi-level sampling will outperform single-level uniform random sampling on average.

show abstract

Section: Introductionmentioning

confidence: 94%

Section: Compressed Sensingmentioning

confidence: 99%

Improving spatial domain based image formation through compressed sensing

Stoltz¹,

Nel²

2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…A rosette-scan pattern has also been used to sparsely sample the scene and reconstruct images using CS. [7][8][9] The image quality (for a fixed image size) for CS-based systems is associated with the number of samples taken. Since the development of high-specification DMD devices, high framerates were achievable in real-world applications.…”

Section: Introductionmentioning

confidence: 99%

Simulation of a tomographic single-pixel imager using rosette scanning

Duvenage,

du Plessis

2023

Electro-Optical and Infrared Systems: Technology and Applications XX

View full text Add to dashboard Cite

Spin-scan and conical-scan tomographic scanning (TOSCA) imagers have produced good-quality and costeffective images and video in both the infrared (IR) and optical wavelengths. A novel rosette-scan implementation of TOSCA single-pixel imaging is presented below. Previous conical-scan TOSCA imagers implemented a reticle with a fixed number of thin slits. This resulted in a fixed angular resolution which implied a fixed image resolution. The feasibility of a rosette-scan implementation using similar processing techniques to conical-scan TOSCA imagers will be demonstrated. The rosette-scan implementation would only require a reticle with a single thin slit, instead of a reticle with a number of thin slits at fixed angles. The single thin-slit reticle can be rotated to be perpendicular to the line-scan angle of each rosette petal. The number of scan angles can be dynamically changed to achieve different trade-offs between resolution and frame rate by varying the rotational speeds of the prisms and the single thin slit reticle.

show abstract

“…Additionally, the use of rosette-scan patterns with CS reconstruction has been considered [12,13]. Rosette-scan systems move a narrow instantaneous field of view (IFOV) over the scene, as shown in Figure 2a, with the scan pattern (the number of petals) and the scan rate (the speed with which the IFOV moves) determining the clarity of the reconstructed image.…”

Section: Introductionmentioning

confidence: 99%

“…Rosette-scan systems move a narrow instantaneous field of view (IFOV) over the scene, as shown in Figure 2a, with the scan pattern (the number of petals) and the scan rate (the speed with which the IFOV moves) determining the clarity of the reconstructed image. Rosette-scan inherently emphasises the centre region of the scene, but there are applications where the majority of image information is assumed to be predominantly centred within the scene [13,14].…”

Section: Introductionmentioning

confidence: 99%

Rosette‐scan tomographic single‐pixel imager

Duvenage,

du Plessis

2023

Electronics Letters

View full text Add to dashboard Cite

Spin‐scan tomographic scanning imagers utilise a reticle with a fixed number of thin slits and nutating optics, and have produced images and video at both infrared and optical wavelengths. However, the fixed nature of these systems reduces their versatility by restricting changes that can be made after manufacture. An alternative rosette‐scan tomographic scanning imager based on similar concepts is proposed to overcome this limitation. A single thin‐slit reticle is rotated to be perpendicular to the line scan angle of each rosette petal, thereby allowing tomographic reconstruction. The parameters of the proposed system can be dynamically changed to achieve different trade‐offs between resolution and frame rate. Initial results obtained from a proof‐of‐concept prototype demonstrate the feasibility of this approach.

show abstract

Performance estimation of a real-time rosette imager

Cited by 4 publications

References 16 publications

Improving spatial domain based image formation through compressed sensing

Improving spatial domain based image formation through compressed sensing

Simulation of a tomographic single-pixel imager using rosette scanning

Rosette‐scan tomographic single‐pixel imager

Contact Info

Product

Resources

About