Building facade images for classifying building stories and identifying building typologies

The study, titled 'Effectiveness of Image Augmentation Techniques on Detection of Building Characteristics from Street View Images Using Deep Learning,' This paper investigates the impact of eight distinct image augmentation techniques—Brightness, Contrast, Perspective, Rotate, Scale, Shear, Translate, and a combined method termed 'A Sum of Techniques'—on two specific tasks: classifying building stories and identifying building typologies. The primary aim of this research is to enhance accuracy in these tasks using the aforementioned augmentation techniques. Additionally, this study emphasizes reproducibility of the proposed methods. All images used in the research were annotated specifically for the classification task.

本研究标题为《基于深度学习的街景图像建筑特征检测中的图像增强技术有效性》。本研究探究了8种不同的图像增强技术——亮度（Brightness）增强、对比度（Contrast）增强、透视（Perspective）变换、旋转（Rotate）、缩放（Scale）、斜切（Shear）变换、平移（Translate），以及名为“综合复合增强法（A Sum of Techniques）”的组合方法——对两项特定任务的影响：建筑层数分类与建筑类型识别。本研究的核心目标为通过上述图像增强技术提升这两项任务的分类准确率。此外，本研究着重强调所提出方法的可复现性。本研究中使用的所有图像均针对该分类任务进行了专属标注。