We report an automated morphological classification of galaxies into S-wise spirals, Z-wise spirals, and non-spirals using big image data taken from Subaru/Hyper Suprime-Cam (HSC) Survey and a convolutional neural network(CNN)-based deep learning technique. The HSC i-band images are about 25 times deeper than those from the Sloan Digital Sky Survey (SDSS) and have a two times higher spatial resolution, allowing us to identify substructures such as spiral arms and bars in galaxies at z>0.1. We train CNN classifiers by using HSC images of 1447 S-spirals, 1382 Z-spirals, and 51,650 non-spirals. As the number of images in each class is unbalanced, we augment the data of spiral galaxies by horizontal flipping, rotation, and rescaling of images to make the numbers of three classes similar. The trained CNN models correctly classify 97.5% of the validation data, which is not used for training. We apply the CNNs to HSC images of a half million galaxies with an i-band magnitude of i 0.2, where we are hardly able to identify spiral arms in the SDSS images. Our attempt demonstrates that a combination of the HSC big data and CNNs has a large potential to classify various types of morphology such as bars, mergers and strongly-lensed objects.

中文翻译：

---

螺旋星系的自旋平价 II：使用来自斯巴鲁超级超级相机调查和深度学习的大数据的 8 万螺旋星系目录

我们使用取自 Subaru/Hyper Suprime-Cam (HSC) Survey 的大图像数据和基于卷积神经网络 (CNN) 的深层图像数据，将星系自动形态分类为 S 向螺旋、Z 向螺旋和非螺旋。学习技术。HSC i 波段图像比斯隆数字巡天 (SDSS) 图像深约 25 倍，空间分辨率高两倍，使我们能够识别 z>0.1 星系中的螺旋臂和棒等子结构。我们通过使用 1447 个 S 螺旋、1382 个 Z 螺旋和 51,650 个非螺旋的 HSC 图像训练 CNN 分类器。由于每类图像的数量不平衡，我们通过水平翻转、旋转和重新缩放图像来增加螺旋星系的数据，使三类的数量相似。经过训练的 CNN 模型正确分类了 97。5% 的验证数据，不用于训练。我们将 CNN 应用于 i-band 震级为 i 0.2 的五十万个星系的 HSC 图像，其中我们几乎无法识别 SDSS 图像中的旋臂。我们的尝试表明，HSC 大数据和 CNN 的结合具有对各种类型的形态进行分类的巨大潜力，例如条形、合并和强透镜物体。