Mzh's Studio.

Skeletion paper notes

字数统计: 977阅读时长: 4 min
2020/11/12 Share

Skeletonization of Plant Point Cloud Data Using Stochastic Optimization Framework(有待进一步研究)

问题:

Problems with the existing approaches on skeletonizing plant point cloud data. Original point cloud is shown as black point cloud and the skeleton points are shown as red points. (Left) The problem of zigzag structure, where the skeleton does not follow the centerline of the stem and tends to deviate toward the branching point (Xu et al., 2007) (only the main stem skeleton is shown in the figure). (Middle) The problem of biologically irrelevant skeleton points which falls beyond the boundary of the input data (shown at the top part), and inability to capture the geometric details for some branches (Delagrange et al., 2014; Ziamtsov and Navlakha, 2019). (Right) Overlooking tiny geometrical structures (Xu et al., 2007).

现有的关于植物点云数据框架化的方法存在问题: 原始点云显示为黑点云,骨架点显示为红点。 (左)锯齿形结构问题,即骨架不跟随茎的中心线,而倾向于向分支点偏移(Xu等,2007)(图中仅显示了主茎骨架)。 (中)生物学上不相关的骨架点超出输入数据的边界(显示在顶部)的问题,以及无法捕获某些分支的几何细节的问题(Delagrange等人,2014年; Ziamtsov和Navlakha,2019年) )。 (右)俯瞰微小的几何结构(Xu等,2007)

img

抛出三个问题:

①Z字型结构,需要解决的关键问题

②无效不不准确的骨架点计算:骨架点在数据外部;数量不足区域无法表示分支的实际情况,完全分支无法用骨架点表示,解决方法是用插值或点近似策略生成更多骨架点,

③细小分支被视为噪声

初始化骨架,然后参数化表示骨架,用图表示,由于节点突变,所以考虑节点曲率求导为0?随机建模方法有点难理解,需要时间

An Accurate Skeleton Extraction Approach From 3D Point Clouds of Maize Plants

Skeleton extraction from 3D plant point cloud data is an essential prior for myriads of phenotyping studies. Although skeleton extraction from 3D shapes have been studied extensively in the computer vision and graphics literature, handling the case of plants is still an open problem.

Drawbacks of the existing approaches include the zigzag structure of the skeleton, nonuniform density of skeleton points, lack of points in the areas having complex geometry structure, and most importantly the lack of biological relevance.

自适应采样:目的:确定关键点,保持分支的几何特性,在枝条的交点采用不同的球形半径采样技术

确定是否是交叉点,引入点的方向堵来描述线性趋势,就是协方差确定方向度量衡,跟L1这块做的思路一致

方法效果一般可借鉴思路:骨架点连接部分,如何剔除非同一骨架点连接的情况。以及修正骨架。

①该文章思路是认为局部骨架点能够成近似平面,根据点距离拟合平面的距离(并非简单的欧氏距离,是带有区里权重的一个参数)来判断点是否是在某个骨架分支上。(平面是一个借鉴思路,实际上我也这样做,但是问题是,很多时候并不是平面,因此得看具体数据做出判断)

②在骨架修正方面,该文章思路找垂直于某个骨架点切方向的平面然后,找到该切平面的聚类点集进行重新拟合骨架点。

Knowledge and Heuristic-Based Modelingof Laser-Scanned Trees

需要先验经验(手动选择模型根节点,或者模型严格满足平行xyz坐标系,选择最低点)

TopoAngler: Interactive Topology-based Extraction of Fishes

对CT鱼群的扫描进行自动化的分割识别

instead of forcing the user to detect fishes in the dataset and manually construct bounding boxes, our system makes use of a hierarchical segmentation to find features of interest in the dataset

通过分级分割获取数据特征

Towards “smart canopy” sorghum: discovery of the genetic control of leaf angle across layers

顶部直立,底部平铺,最大程度提高光的拦截和转换效率,增产

角度分布室内室外与基因型相反?(室内外不一致?)