Division of Translational Surgical Oncology (TSO), National Center for Tumor Diseases (NCT/UCC) Dresden, Fetcherstr. 74, 01039, Dresden, Germany.
Department of Mathematics and Computer Science, University of Marburg, Hans-Meerwein-Str. 6, 35032, Marburg, Germany.
BMC Med Imaging. 2021 Aug 5;21(1):119. doi: 10.1186/s12880-021-00650-z.
Object detection and image segmentation of regions of interest provide the foundation for numerous pipelines across disciplines. Robust and accurate computer vision methods are needed to properly solve image-based tasks. Multiple algorithms have been developed to solely detect edges in images. Constrained to the problem of creating a thin, one-pixel wide, edge from a predicted object boundary, we require an algorithm that removes pixels while preserving the topology. Thanks to skeletonize algorithms, an object boundary is transformed into an edge; contrasting uncertainty with exact positions.
To extract edges from boundaries generated from different algorithms, we present a computational pipeline that relies on: a novel skeletonize algorithm, a non-exhaustive discrete parameter search to find the optimal parameter combination of a specific post-processing pipeline, and an extensive evaluation using three data sets from the medical and natural image domains (kidney boundaries, NYU-Depth V2, BSDS 500). While the skeletonize algorithm was compared to classical topological skeletons, the validity of our post-processing algorithm was evaluated by integrating the original post-processing methods from six different works.
Using the state of the art metrics, precision and recall based Signed Distance Error (SDE) and the Intersection over Union bounding box (IOU-box), our results indicate that the SDE metric for these edges is improved up to 2.3 times.
Our work provides guidance for parameter tuning and algorithm selection in the post-processing of predicted object boundaries.
目标检测和感兴趣区域的图像分割为跨学科的众多流水线提供了基础。需要强大而准确的计算机视觉方法才能正确解决基于图像的任务。已经开发了多种算法来仅检测图像中的边缘。受限于从预测对象边界创建一条细的、一像素宽的边缘的问题，我们需要一种在保留拓扑结构的同时去除像素的算法。多亏了骨架化算法，对象边界被转换为边缘；用确切位置来对比不确定性。
为了从不同算法生成的边界中提取边缘，我们提出了一个计算管道，该管道依赖于：一种新颖的骨架化算法、一种非详尽的离散参数搜索，以找到特定后处理管道的最佳参数组合，以及使用来自医学和自然图像领域的三个数据集（肾脏边界、NYU-Depth V2、BSDS 500）进行广泛评估。虽然骨架化算法与经典拓扑骨架进行了比较，但通过集成来自六个不同工作的原始后处理方法，评估了我们的后处理算法的有效性。
使用最先进的指标，基于有符号距离误差 (SDE) 的精度和召回率以及边界框的交并比 (IOU-box)，我们的结果表明，这些边缘的 SDE 指标提高了 2.3 倍。
我们的工作为预测对象边界的后处理中的参数调整和算法选择提供了指导。
