Parallel-beam X-ray CT datasets of apples with internal defects

本数据集由数学与计算机科学中心创建，包含72,192张二维CT图像，涉及94个内部有缺陷的苹果。数据集大小庞大，每张图像对应50个平行束投影，用于模拟各种稀疏或有限视角的断层扫描问题。创建过程中，使用了定制的高灵活性CT扫描仪，并进行了详细的图像重建和优化。该数据集主要应用于图像重建、分割、自动缺陷检测以及机器学习中标签偏差的测试和方法开发，旨在解决在有限数据条件下提高CT图像质量的问题。

This dataset was developed by the Center for Mathematics and Computer Science, and comprises 72,192 2D CT images derived from 94 apples with internal defects. Boasting a substantial scale, each image is paired with 50 parallel-beam projections, enabling the simulation of diverse sparse-view or limited-angle computed tomography problems. During the dataset creation process, a custom high-flexibility CT scanner was employed, alongside rigorous image reconstruction and optimization procedures. This dataset is primarily utilized for tasks including image reconstruction, segmentation, automated defect detection, as well as the testing and methodology development for label bias in machine learning. Its primary objective is to tackle the challenge of enhancing CT image quality under limited data acquisition conditions.