Single Image Super-Resolution Method Using CNN-Based Lightweight Neural Networks

There are many studies that seek to enhance a low resolution image to a high resolution image in the area of super-resolution. As deep learning technologies have recently shown impressive results on the image interpolation and restoration field, recent studies are focusing on convolutional neural network (CNN)-based super-resolution schemes to surpass the conventional pixel-wise interpolation methods. In this paper, we propose two lightweight neural networks with a hybrid residual and dense connection structure to improve the super-resolution performance. In order to design the proposed networks, we extracted training images from the DIVerse 2K (DIV2K) image dataset and investigated the trade-off between the quality enhancement performance and network complexity under the proposed methods. The experimental results show that the proposed methods can significantly reduce both the inference speed and the memory required to store parameters and intermediate feature maps, while maintaining similar image quality compared to the previous methods.

A practical approach on non-regular sampling and universal demosaicing of raw image sensor data

Non-regular sampling is a well-known method to avoid aliasing in digital images. However, the vast majority of single sensor cameras use regular organized color filter arrays (CFAs), that require an optical-lowpass filter (OLPF) and sophisticated demosaicing algorithms to suppress sampling errors. In this paper a variety of non-regular sampling patterns are evaluated, and a new universal demosaicing algorithm based on the frequency selective reconstruction is presented. By simulating such sensors it is shown that images acquired with non-regular CFAs and no OLPF can lead to a similar image quality compared to their filtered and regular sampled counterparts. The MATLAB source code and results are available at: http://github. com/PhilippBackes/dFSR

Near, far, wherever you are: simulations on the dose efficiency of holographic and ptychographic coherent imaging

Different studies in X-ray microscopy have arrived at conflicting conclusions about the dose efficiency of imaging modes involving the recording of intensity distributions in the near (Fresnel regime) or far (Fraunhofer regime) field downstream of a specimen. A numerical study is presented on the dose efficiency of near-field holography, near-field ptychography and far-field ptychography, where ptychography involves multiple overlapping finite-sized illumination positions. Unlike what has been reported for coherent diffraction imaging, which involves recording a single far-field diffraction pattern, it is found that all three methods offer similar image quality when using the same fluence on the specimen, with far-field ptychography offering slightly better spatial resolution and a lower mean error. These results support the concept that (if the experiment and image reconstruction are done properly) the sample can be near or far; wherever you are, photon fluence on the specimen sets one limit to spatial resolution.

Patient-Assisted Compression in Screening Mammography: Patient Experience and Image Quality

Objective Screening mammography is critical to reduce breast cancer mortality, yet many women cite pain from compression as a reason they avoid this test. We evaluated patient experience and image quality in screening patients opting for a handheld patient-assisted compression (PAC) device. Methods After institutional review board approval, women screened between February and July 2018 with a synthetic 2D/tomosynthesis mammography unit were offered use of a handheld PAC device. Patient experience through survey, image quality, compression thickness, compression force, and average glandular dose were evaluated and compared between women opting for PAC and women opting for technologist-controlled compression (TC). Multivariable ordinal logistic and linear regression models were estimated to control for age and breast density. In addition, for women opting for PAC, image quality obtained with their current PAC mammogram was compared with that obtained with their prior TC mammogram, by using Wilcoxon/Pearson tests. Results Seventy-three percent of women preferred their mammogram experience with PAC compared with their prior mammogram without PAC. Women using PAC reported decreased anxiety compared with those using TC, after controlling for age and breast density (adjusted odds ratio [aOR] 0.22 [95% confidence interval (CI): 0.09–0.49]). There were no significant differences in image quality, compression thickness, or average glandular dose in exams for women using PAC compared with exams for women using TC. Women using PAC had significantly more compression force than women using TC had (P = 0.012). Conclusions Mammography with PAC improves patient experience and results in similar image quality compared with mammography with TC.

Overview of the Near-IR S0 Galaxy Survey (NIRS0S)

An overview of the results of the near-IR S0 galaxy survey (NIRS0S) is presented. NIRS0S is a magnitude- (mB⩽12.5 mag) and inclination- (<65∘) limited sample of ̃200 nearby galaxies, mainly S0s, but includes also Sa and E galaxies. It uses deepKs-band images, typically reaching a surface brightness of 23.5 mag arcsec−2. Detailed visual and photometric classifications were made, for the first time coding also the lenses in a systematic manner. As a comparison sample, a similar sized spiral galaxy sample with similar image quality was used. Based on our analysis, the Hubble sequence was revisited: following the early idea by van den Bergh we suggested that the S0s are spread throughout the Hubble sequence in parallel tuning forks as spirals (S0a, S0b, S0c, etc.). This is evidenced by our improved bulge-to-total (B/T) flux ratios in the S0s, reaching small values typical to late-type spirals. The properties of bulges and disks in S0s were found to be similar to those in spirals and, also, the masses and scale parameters of the bulges and disks to be coupled. It was estimated that the spiral bulges brighter than −20 mag inK-band are massive enough to be converted into the bulges of S0s merely by star formation. Bars were found to be fairly robust both in S0s and spirals, but, inspite of that, bars might evolve significantly within the Hubble sequence.