2016年辣椒图像数据集

辣椒数据集是使用合成方法创建的农业计算机视觉任务数据集。该数据集不通过人工或机械拍摄采集图像，二是通过图像合成的方法，基于真实植物的经验测量对植物几何参数、颜色和纹理特征进行建模，然后进行计算绘制形成图像。该数据集由10,500张png 格式的合成彩色图像组成，对8类植物的组成部分进行像素级分割，包括茎、节、侧枝、叶、花序梗、果实和花。合成数据集为基准语义分割任务提供了良好的基础，但对于现实世界广泛应用需求，还需要实验图像或真实图像对象检测和分割算法进行微调。https://data.4tu.nl/articles/dataset/Synthetic_and_Empirical_Capsicum_Annuum_Image_Dataset/12706703

The Chili Pepper Dataset is an agricultural computer vision task dataset created using synthetic methods. Instead of collecting images via manual or mechanical shooting, this dataset models the geometric parameters, color and texture features of plants based on empirical measurements of real plants through image synthesis, then generates images via computational rendering. This dataset consists of 10,500 synthetic color images in PNG format, with pixel-level segmentation annotations for 8 types of plant components, including stems, nodes, lateral branches, leaves, pedicels, fruits and flowers. The synthetic dataset provides a solid foundation for benchmark semantic segmentation tasks, but for the requirements of real-world wide-ranging applications, it still needs to be fine-tuned with experimental images or real-image object detection and segmentation algorithms. https://data.4tu.nl/articles/dataset/Synthetic_and_Empirical_Capsicum_Annuum_Image_Dataset/12706703