Linear object resampling refers to the process of simplifying geometries (lines or region boundaries) by removing redundant nodes while preserving their original shapes according to specific rules.

Currently supported resampling methods for linear objects in vector datasets include Blend and Douglas-Peucker.

1. Resampling Using Blend Method

1. Take the polyline with 6 nodes shown below as an example to illustrate the calculation process:

   

2. As shown in the left figure below, construct perpendicular lines to segment 12 at node 2. Select points A and B on these perpendicular lines at "resampling distance m" from node 2. Connect node 1 to points A/B respectively and extend to form resampling areas (pink regions). Determine whether node 3 falls within these areas. If yes, delete node 2; otherwise retain it. Since node 3 is within the area, node 2 is removed, resulting in the simplified shape shown in the right figure.

   

3. Next, evaluate node 3 using the same method. As shown below, node 4 lies outside the resampling area (green region), thus node 3 is retained.

   

4. Continue evaluating remaining nodes to obtain the final resampling result as shown:

   

2. Resampling Using Douglas-Peucker Method

Using the same 6-node polyline example: Connect the first and last nodes to form a baseline. Find the node with maximum distance (d) from this baseline. If d > m (resampling tolerance), retain this node and split the line at this point for recursive processing. Repeat until no more simplification is possible.

1. In the figure below, d > m, so node 2 is retained. The line splits into two segments: nodes 1-2 and 2-3-4-5-6.

   

2. As shown in the left figure, the segment 1-2 cannot be simplified further. For segment 2-3-4-5-6, since d < m, intermediate nodes are removed leaving only nodes 2 and 6, producing the final result shown on the right.