分类： 天文学 >> 天文学 提交时间： 2024-03-29 合作期刊: 《Research in Astronomy and Astrophysics》
摘要： Galaxy morphology classifications based on machine learning are a typical technique to handle enormous amounts of astronomical observation data, but the key challenge is how to provide enough training data for the machine learning models. Therefore this article proposes an image data augmentation method that combines few-shot learning and generative adversarial networks. The Galaxy10 DECaLs data set is selected for the experiments with consistency, variance, and augmentation effects being evaluated. Three popular networks, including AlexNet, VGG, and ResNet, are used as examples to study the effectiveness of different augmentation methods on galaxy morphology classifications. Experiment results show that the proposed method can generate galaxy images and can be used for expanding the classification model's training set. According to comparative studies, the best enhancement effect on model performance is obtained by generating a data set that is 0.5–1 time larger than the original data set. Meanwhile, different augmentation strategies have considerably varied effects on different types of galaxies. FSL-GAN achieved the best classification performance on the ResNet network for In-between Round Smooth Galaxies and Unbarred Loose Spiral Galaxies, with F1 Scores of 89.54% and 63.18%, respectively. Experimental comparison reveals that various data augmentation techniques have varied effects on different categories of galaxy morphology and machine learning models. Finally, the best augmentation strategies for each galaxy category are suggested.
分类： 天文学 >> 天文学 提交时间： 2023-02-19
摘要： With the development of a series of Galaxy sky surveys in recent years, the observations increased rapidly, which makes the research of machine learning methods for galaxy image recognition a hot topic. Available automatic galaxy image recognition researches are plagued by the large differences in similarity between categories, the imbalance of data between different classes, and the discrepancy between the discrete representation of Galaxy classes and the essentially gradual changes from one morphological class to the adjacent class (DDRGC). These limitations have motivated several astronomers and machine learning experts to design projects with improved galaxy image recognition capabilities. Therefore, this paper proposes a novel learning method, ``Hierarchical Imbalanced data learning with Weighted sampling and Label smoothing" (HIWL). The HIWL consists of three key techniques respectively dealing with the above-mentioned three problems: (1) Designed a hierarchical galaxy classification model based on an efficient backbone network; (2) Utilized a weighted sampling scheme to deal with the imbalance problem; (3) Adopted a label smoothing technique to alleviate the DDRGC problem. We applied this method to galaxy photometric images from the Galaxy Zoo-The Galaxy Challenge, exploring the recognition of completely round smooth, in between smooth, cigar-shaped, edge-on and spiral. The overall classification accuracy is 96.32\%, and some superiorities of the HIWL are shown based on recall, precision, and F1-Score in comparing with some related works. In addition, we also explored the visualization of the galaxy image features and model attention to understand the foundations of the proposed scheme.
分类： 天文学 >> 天文学 提交时间： 2023-02-19
摘要： With the development of a series of Galaxy sky surveys in recent years, the observations increased rapidly, which makes the research of machine learning methods for galaxy image recognition a hot topic. Available automatic galaxy image recognition researches are plagued by the large differences in similarity between categories, the imbalance of data between different classes, and the discrepancy between the discrete representation of Galaxy classes and the essentially gradual changes from one morphological class to the adjacent class (DDRGC). These limitations have motivated several astronomers and machine learning experts to design projects with improved galaxy image recognition capabilities. Therefore, this paper proposes a novel learning method, ``Hierarchical Imbalanced data learning with Weighted sampling and Label smoothing" (HIWL). The HIWL consists of three key techniques respectively dealing with the above-mentioned three problems: (1) Designed a hierarchical galaxy classification model based on an efficient backbone network; (2) Utilized a weighted sampling scheme to deal with the imbalance problem; (3) Adopted a label smoothing technique to alleviate the DDRGC problem. We applied this method to galaxy photometric images from the Galaxy Zoo-The Galaxy Challenge, exploring the recognition of completely round smooth, in between smooth, cigar-shaped, edge-on and spiral. The overall classification accuracy is 96.32\%, and some superiorities of the HIWL are shown based on recall, precision, and F1-Score in comparing with some related works. In addition, we also explored the visualization of the galaxy image features and model attention to understand the foundations of the proposed scheme.
分类： 物理学 >> 地球物理学、天文学和天体物理学 提交时间： 2023-12-15 合作期刊: 《Research in Astronomy and Astrophysics》
摘要： Machine learning has become a crucial technique for classifying the morphology of galaxies as a result of the meteoric development of galactic data. Unfortunately, traditional supervised learning has significant learning costs since it needs a lot of labeled data to be effective. FixMatch, a semi-supervised learning algorithm that serves as a good method, is now a key tool for using large amounts of unlabeled data. Nevertheless, the performance degrades significantly when dealing with large, imbalanced data sets since FixMatch relies on a fixed threshold to filter pseudo-labels. Therefore, this study proposes a dynamic threshold alignment algorithm based on the FixMatch model. First, the class with the highest amount has its reliable pseudo-label ratio determined, and the remaining classes' reliable pseudo-label ratios are approximated in accordance. Second, based on the predicted reliable pseudo-label ratio for each category, it dynamically calculates the threshold for choosing pseudo-labels. By employing this dynamic threshold, the accuracy bias of each category is decreased and the learning of classes with less samples is improved. Experimental results show that in galaxy morphology classification tasks, compared with supervised learning, the proposed algorithm significantly improves performance. When the amount of labeled data is 100, the accuracy and F1-score are improved by 12.8% and 12.6%, respectively. Compared with popular semi-supervised algorithms such as FixMatch and MixMatch, the proposed algorithm has better classification performance, greatly reducing the accuracy bias of each category. When the amount of labeled data is 1000, the accuracy of cigar-shaped smooth galaxies with the smallest sample is improved by 37.94% compared to FixMatch.
点击量
通过
点击量
下载量
评论
待评议
