Toward Ultralightweight Remote Sensing With Harmonic Lenses and Convolutional Neural Networks

Никоноров, Артем; Petrov, Maksim V.; Bibikov, S. A.; Yakimov, Pavel; Kutikova, Viktoriya V.; Yuzifovich, Yuriy V.; Морозов, А. А.; Скиданов, Р. В.; Kazanskiy, Nikolay L.

doi:10.1109/jstars.2018.2856538

Cited by 55 publications

(17 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As demonstrated in previous works [ 20 , 34 ], diffractive optics can cause two types of image degradation: local degradation, which is caused by chromatic aberration, and non-local, content-aware chromatic shift, which is caused by the redistribution of energy between the secondary diffractive orders of the lens. Since these degradations affect areas larger than 200 pixels in width in our setup [ 34 ], we use a CNN with a receptive field wider than 200 pixels, which is based on the modified U-Net [ 21 ] architecture.…”

Section: Deep Learning-based Image Reconstructionmentioning

confidence: 99%

“…In previous works [ 16 , 17 , 18 , 19 ], deep learning-based image reconstruction was successfully used to compensate for the chromatic distortions typical in an optical system with a harmonic diffractive lens. Despite good results of the deep learning-based correction as measured by the peak signal-to-noise ratio (PSNR) on a test set at about 27 dB [ 16 , 20 ], reconstructed real scene images showed visible reconstruction artifacts. These artifacts were caused by the following features specific to real scenes as opposed to the training set: high dynamic range (HDR), camera gain, and lossy video compression.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Hybrid Refractive-Diffractive Lens with Reduced Chromatic and Geometric Aberrations and Learned Image Reconstruction

Evdokimova

Подлипнов

Ivliev

et al. 2022

Sensors

Self Cite

View full text Add to dashboard Cite

In this paper, we present a hybrid refractive-diffractive lens that, when paired with a deep neural network-based image reconstruction, produces high-quality, real-world images with minimal artifacts, reaching a PSNR of 28 dB on the test set. Our diffractive element compensates for the off-axis aberrations of a single refractive element and has reduced chromatic aberrations across the visible light spectrum. We also describe our training set augmentation and novel quality criteria called “false edge level” (FEL), which validates that the neural network produces visually appealing images without artifacts under a wide range of ISO and exposure settings. Our quality criteria (FEL) enabled us to include real scene images without a corresponding ground truth in the training process.

show abstract

Section: Deep Learning-based Image Reconstructionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Hybrid Refractive-Diffractive Lens with Reduced Chromatic and Geometric Aberrations and Learned Image Reconstruction

Evdokimova

Подлипнов

Ivliev

et al. 2022

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…Among the works of Russian scientists, both fundamental research on computer optics methods [21][22][23] and physical processes of image registration [24,25] and applied research aimed at the development of integrated satellite monitoring data processing systems [26,27] can be highlighted. The results of these studies make it possible to improve both the hardware characteristics of remote sensing systems, for example, by significantly facilitating the registration system [28], and to improve the quality of the monitoring data obtained [29,30]. Much attention is being paid to the color properties of images.…”

Section: Related Workmentioning

confidence: 99%

Segmentation of Agricultural Parcels in Satellite Images Based on Historical Vegetation Index Data

Pavlova¹,

Timofeev²,

Bocharov³

et al. 2021

Preprint

View full text Add to dashboard Cite

This paper considered the issue of agricultural fields boundary recognition in satellite images. A novel algorithm based on the aggregated history of vegetation index data obtained via open satellite data, Sentinel-2, was proposed. The proposed algorithm included several basic steps, namely the detection of parcel regions on aggregated index data; the calculation of aggregated edge maps; the segmentation of parcel regions using the edges obtained; the computation of connected components and their contour extraction. In this paper, we showed that the use of aggregated vegetation index data and boundary maps allow for much more accurate agricultural field segmentation compared to the instant vegetation index approach. The quality of segmentation within regions of Russia and the Ukraine was estimated. The dataset that was used and Python implementation of the proposed algorithm were provided.

show abstract

“…There was an aberration of the coma, based on the form of the PSF. Even at such a PSF, the objective has a resolution of approximately 50 l/mm, which is acceptable for practical uses, that is, this objective is an imaging system and can be used in combination with modern photosensor arrays, whereas the use of post-capture image processing [ 27 , 28 , 29 , 30 ] enables the acquisition of an image comparable in quality with that of refractive objectives. Thus, the proposed imaging system could have practical uses in areas where compactness is of greater significance than the lens aperture.…”

Section: Numerical Ray Tracing For the Proposed Optical Systemmentioning

confidence: 99%

Compact Imaging Systems Based on Annular Harmonic Lenses

Скиданов

Strelkov

Volotovskiy

et al. 2020

Sensors

View full text Add to dashboard Cite

In this study, a configuration of a compact imaging objective based on a reflecting annular harmonic lens was proposed. Light propagation through the proposed optical system was comprehensively modeled using a dedicated special program and the ZEMAX software, with the latter used to derive the point spread function (PSF). Several relationships were used to describe the connection between key parameters of the objective, including its focal length, field of view, and thickness. We demonstrated that it was possible to design a compact imaging objective whose overall length could be one to two orders of magnitude smaller than its focal length. Using direct laser writing, a reflecting annular harmonic lens was fabricated and used in the proposed objective scheme. The performance of the objective was experimentally studied by imaging a light source and a test pattern. The performance of the compact imaging objective based on a reflecting annular harmonic lens was verified in principle. A PSF value of approximately 16 microns was experimentally obtained, for a lens with a diameter of 25 mm with a focal length of 100 mm.

show abstract

Toward Ultralightweight Remote Sensing With Harmonic Lenses and Convolutional Neural Networks

Cited by 55 publications

References 31 publications

Hybrid Refractive-Diffractive Lens with Reduced Chromatic and Geometric Aberrations and Learned Image Reconstruction

Hybrid Refractive-Diffractive Lens with Reduced Chromatic and Geometric Aberrations and Learned Image Reconstruction

Segmentation of Agricultural Parcels in Satellite Images Based on Historical Vegetation Index Data

Compact Imaging Systems Based on Annular Harmonic Lenses

Contact Info

Product

Resources

About