Camellia japonica CT slice images

We aim to obtain a three-dimensional (3D) structure of Camellia japonica flowers and propose a petal segmentation method using computer vision techniques. Petal segmentation on the slice images fails by simply applying the segmentation methods because the shape of the petals in slice CT images differs from that of the objects targeted by the latest instance segmentation methods. Moreover, because the number of images used to train the model is limited, it is difficult to apply segmentation methods that require a large amount of training data. To overcome these challenges, we crop long rectangles from the slice images and apply the segmentation method to segment the petals. Thanks to cropping, it is possible to segment the shape of the petals in the cropped images, and to increase training data. The experimental results show that the proposed method can segment petals on slice images with higher accuracy than the method without cropping. The 3D segmentation results were also obtained and visualized successfully.The authors would like to thank Suzuki and Ohtake Laboratory, Graduate School of Engineering, University of Tokyo, for CT data acquisition for this study.