Search

Fast and accurate image retrieval is an important and challenging task in massive image data scenarios. As the core technology of image retrieval tasks, deep metric learning aims at learning effective embedding representations that possess two properties among data points: positive concentrated and negative separated. In this work, we propose a multilevel similarity-aware method based on deep local descriptors for deep metric learning. We take the rich interclass similarity relationship based on the deep local invariant descriptors from the data into account to optimize sampling strategies for mining informative samples. The method dynamically adjusts the margin between data points to better match the true similarity relationship between classes. Specifically, for images in a batch, we first obtain deep local descriptors and calculate the similarity matrix of the channel, pixel, and spatial levels. Then, depending on the calculated comprehensive similarity matrix, we propose a multilevel similarity-aware loss function through the deviation between pairwise distance and violate margin to make full use of informative samples. The experimental results demonstrate that our proposed method outperforms other state-of-the-art methods in terms of fine-grained image retrieval and clustering tasks.

The X-ray spectrum emitted from laser-produced plasma contains plentiful information. X-ray spectrometer is a powerful tool for plasma diagnosis and studying the information and evolution of the plasma. X-ray concave (elliptical) curved crystals analyzer was designed and manufactured to investigate the properties of laser-produced plasma. The experiment was carried out on Mianyang Xingguang-II Facility and aimed at investigating the characteristics of a high density iron plasma. Experimental results using KAP, LIF, PET, and MICA curved crystal analyzers are described, and the spectra of Au, Ti laser-produced plasma are shown. The focusing crystal analyzer clearly gave an increase in sensitivity over a flat crystal.

In order to measure spatially and temporarily resolved laser-produced plasma X-ray spectra in 0.2-2 nm region, a novel two-channel elliptically-bent crystal spectrograph has been developed. Dispersive elements are LiF, PET, Mica, and KAP crystals, which cover Bragg angles in the range of 30-67.5 degrees. Eccentricity and focal distance of twin ellipses are 0.9586 and 1350 mm, respectively. Spatially resolved spectrum is photographically recorded with an X-ray film or X-CCD camera in one channel, and temporarily resolved one is photographically recorded with an X-ray streak camera in another channel, thus spatially and temporarily resolved spectra can be simultaneously obtained. Spectral images were acquired with X-CCD and PET in SHENGUANG-II laser facility, and experimental results show that the spectral resolution of the spectrograph is about 0.002 nm.