基于残差稠密神经网络的银河系去融合
Galaxy Deblending using Residual Dense Neural networks
摘要：
我们提出了一种新的神经网络方法去融合天文数据中的星系图像使用残差密集神经网络（RDN）架构。我们训练网络上的合成星系图像类似于场星系的典型安排与有限的点扩散函数（PSF）和现实的噪声水平。我们的方法的主要新颖之处在于使用了两个不同的神经网络：i）将单个星系邮票从复合物中分离出来的去混合网络，ii）计算剩余星系数量的分类器网络。去混合的过程是从合成图像中一次迭代地剥离一个星系，直到图像不包含由分类器或其他停止标准确定的其他物体。通过查看两个网络输出的一致性，我们可以评估去混合的质量。我们在不同的质量箱中表征通量和形状重建，并将我们的deblender与行业标准SExtractor进行比较。我们还讨论了可变的PSF和噪声水平的项目未来可能的扩展。
英文摘要：
We present a new neural network approach for deblending galaxy images in astronomical data using Residual Dense Neural network (RDN) architecture. We train the network on synthetic galaxy images similar to the typical arrangements of field galaxies with a finite point spread function (PSF) and realistic noise levels. The main novelty of our approach is the usage of two distinct neural networks: i) a deblending network which isolates a single galaxy postage stamp from the composite and, ii) a classifier network which counts the remaining number of galaxies. The deblending proceeds by iteratively peeling one galaxy at a time from the composite until the image contains no further objects as determined by the classifier, or by other stopping criteria. By looking at the consistency in the outputs of the two networks, we can assess the quality of the deblending. We characterize the flux and shape reconstructions in different quality bins and compare our deblender with the industry standard, SExtractor. We also discuss possible future extensions for the project with variable PSFs and noise levels.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
