Selected annotated instance segmentation sub-volumes from a large scale CT data-set of a historic aircraft.

The Me 163 was a Second World War fighter airplane and is currently displayed in the Deutsches Museum in Munich, Germany. A complete computed tomography (CT) scan was obtained using a large scale industrial CT scanner to gain insights into its history, design, and state of preservation. The CT data enables visual examination of the airplane's structural details across multiple scales, from the entire fuselage to individual sprockets and rivets. However, further processing requires instance segmentation of the CT data-set. Currently, there are no adequate computer-assisted tools for automated or semi-automated segmentation of such large scale CT airplane data. As a first step, an interactive data annotation process has been established. So far, seven 512 × 512 × 512 voxel sub-volumes of the Me 163 airplane have been annotated, which can potentially be used for various applications in digital heritage, non-destructive testing, or machine learning. This work describes the data acquisition process, outlines the interactive segmentation and post-processing, and discusses the challenges associated with interpreting and handling the annotated data.

3D segmentation of plant root systems using spatial pyramid pooling and locally adaptive field-of-view inference.

Background: The non-invasive 3D-imaging and successive 3D-segmentation of plant root systems has gained interest within fundamental plant research and selectively breeding resilient crops. Currently the state of the art consists of computed tomography (CT) scans and reconstruction followed by an adequate 3D-segmentation process. Challenge: Generating an exact 3D-segmentation of the roots becomes challenging due to inhomogeneous soil composition, as well as high scale variance in the root structures themselves. Approach: (1) We address the challenge by combining deep convolutional neural networks (DCNNs) with a weakly supervised learning paradigm. Furthermore, (2) we apply a spatial pyramid pooling (SPP) layer to cope with the scale variance of roots. (3) We generate a fine-tuned training data set with a specialized sub-labeling technique. (4) Finally, to yield fast and high-quality segmentations, we propose a specialized iterative inference algorithm, which locally adapts the field of view (FoV) for the network. Experiments: We compare our segmentation results against an analytical reference algorithm for root segmentation (RootForce) on a set of roots from Cassava plants and show qualitatively that an increased amount of root voxels and root branches can be segmented. Results: Our findings show that with the proposed DCNN approach combined with the dynamic inference, much more, and especially fine, root structures can be detected than with a classical analytical reference method. Conclusion: We show that the application of the proposed DCNN approach leads to better and more robust root segmentation, especially for very small and thin roots.