A deep learning approach for detecting candidates of supernova remnants

Detecting supernova remnant(SNR) candidates in the interstellar medium is a challenging task because SNRs have weak radio signals and irregular shapes. The use of a convolutional neural network is a deep learning method that can help us extract various features from images. To extract SNRs from astronomical images and estimate the positions of SNR candidates, we design the SNR-Net model composed of a training component and a detection component. In addition, transfer learning is used to initialize the network parameters, which improves the speed and accuracy of network training. We apply a T-T plot(of the different brightness temperatures of map pixels at two different frequencies) to calculate the spectral index of SNR candidates. To accelerate the scientific computing process, we take advantage of innovative hardware architecture, such as deep learning optimized graphics processing units, which increases the speed of computation by a factor of 5. A case study suggests that SNR-Net may be applicable to detecting extended sources in the images automatically.

The distance measurements of supernova remnants in the fourth Galactic quadrant

We take advantage of red clump stars to build the relation of the optical extinction(AV) and distance in each direction of supernova remnants(SNRs) with known extinction in the fourth Galactic quadrant.The distances of nine SNRs are determined well by this method.Their uncertainties range from 10% to30%,which are significantly improved for eight SNRs,G279.0+1.1,G284.3–1.8,G296.1–0.5,G299.2–2.9,G308.4–1.4,G309.2–0.6,G309.8–2.6 and G332.4–0.4.In addition,SNR G284.3–1.8 with the new distance of 5.5 kpc is not likely associated with the PSR J1016–5857 at 3 kpc.

Radio Spectra of Three Supernova Remnants: G114.3+0.3, G116.5+1.1 and G116.9+0.2

Presented are new images of supernova remnants G114.3±0.3, G116.5＋1.1 and G116.9＋0.2 （CTB 1） at 408 MHz from the Canadian Galactic Plane Survey （CGPS）. We also use the 1420 MHz images from the CGPS in a study of their 408-1420 MHz spectral indices. The flux densities at 408 MHz and 1420 MHz, corrected for flux densities from compact sources within the SNRs, are 12±6 Jy and 9.8±0.8 Jy for G114.3＋0.3, 15.0±1.5 Jy and 10.6±0.6 Jy for G116.5＋1.1, 15.0＋1.5 Jy and 8.1±0.4 Jy for Gl16.9＋0.2. The integrated flux density-based spectral indices （Sv∝v^-α） are α=0.16±0.41, 0.28±0.09 and 0.49±0.09 for G114.3＋0.3, G116.5＋1.1 and G116.9＋0.2, respectively. Their T-T plot-based spectral indices are 0.68±0.48, 0.28±0.15, and 0.48±0.04, in agreement with the integrated flux density-based spectral indices. New flux densities are derived at 2695 MHz which are significantly larger than previous values. The new 408, 1420 and 2695 MHz flux densities and published values at other frequencies, where images are not available, are fitted after correcting for contributions from compact sources, to derive their multi-frequency spectral indices.