标题： 基于深度学习的天文学调查检测、实例分割和分类（DeepDISC）：Detectron2实现与Hyper Suprime-Cam数据演示 显示英文标题

标题： Detection, Instance Segmentation, and Classification for Astronomical Surveys with Deep Learning (DeepDISC): Detectron2 Implementation and Demonstration with Hyper Suprime-Cam Data

摘要： 下一代广域深空天文调查将在2020年代及以后提供前所未有的图像数量。 随着观测的灵敏度和深度增加，将检测到更多的混合源。 这种现实可能导致测量偏差，污染关键的天文学推论。 我们通过Facebook AI Research的Detectron2仓库中可用的新深度学习模型，对来自Hyper Suprime-Cam (HSC)的大规模多波段共加图像执行对象识别、去混叠和分类的同步任务。 我们使用现有的检测/去混叠代码和分类方法来训练一系列深度神经网络，包括最先进的变换器。 一旦训练完成，我们发现变换器优于传统的卷积神经网络，并且对不同的对比度缩放更稳健。 变换器能够检测和去混叠与真实情况非常接近的对象，实现了中位数边界框交并比为0.99。 使用来自哈勃空间望远镜的高质量类别标签，我们发现表现最好的网络可以在整个测试样本中以接近100%的完整性和纯度对星系进行分类，并在HSC i波段亮度达到25星等时，对恒星的分类完整性和纯度分别超过60%和80%。 该框架可以扩展到其他即将开展的深空调查，如空间和时间遗产调查以及罗马空间望远镜的调查，以实现快速源检测和测量。 我们的代码DeepDISC可在https://github.com/grantmerz/deepdisc公开获取。 显示英文摘要

摘要： The next generation of wide-field deep astronomical surveys will deliver unprecedented amounts of images through the 2020s and beyond. As both the sensitivity and depth of observations increase, more blended sources will be detected. This reality can lead to measurement biases that contaminate key astronomical inferences. We implement new deep learning models available through Facebook AI Research's Detectron2 repository to perform the simultaneous tasks of object identification, deblending, and classification on large multi-band coadds from the Hyper Suprime-Cam (HSC). We use existing detection/deblending codes and classification methods to train a suite of deep neural networks, including state-of-the-art transformers. Once trained, we find that transformers outperform traditional convolutional neural networks and are more robust to different contrast scalings. Transformers are able to detect and deblend objects closely matching the ground truth, achieving a median bounding box Intersection over Union of 0.99. Using high quality class labels from the Hubble Space Telescope, we find that the best-performing networks can classify galaxies with near 100% completeness and purity across the whole test sample and classify stars above 60% completeness and 80% purity out to HSC i-band magnitudes of 25 mag. This framework can be extended to other upcoming deep surveys such as the Legacy Survey of Space and Time and those with the Roman Space Telescope to enable fast source detection and measurement. Our code, DeepDISC is publicly available at https://github.com/grantmerz/deepdisc.