# TIN Simplify

The algorithm divides the points into square cells so that each point lies in an individual cell. In the first step, the first point is taken from the first cell, according to the order of the points in the input point file or point cloud.

Around the first point, the fulfillment of condition ∆h (**Elevation difference**) is checked for all adjacent points. All points within a given radius, including those in adjacent cells, are considered as adjacent points (see figure below).

When all points with a higher value of ∆h than the user-defined one are excluded, the described procedure continues for the next point in the first cell. The process then continues for all points in the adjacent cell in the x-coordinate direction and so on to the last cell.

- Radius for simplification step
- Radius for simplification step

The algorithm automatically proposes the value of the Radius for simplification step parameter. It does this by calculating the **Bounding Box** based on the given input points. It then takes the larger of the dimensions ∆x and ∆y (length and width). 0.5% of the obtained value is calculated, which is rounded off and displayed in the command line as the proposed value for the user.

Larger cell size (i.e. the Radius for simplification step parameter) brings lower accuracy of the final TIN surface. For greater accuracy of the simplified TIN network, it makes more sense to enter a smaller cell size.