This is due to the way the mesh doubles from its initial condition every iteration. We requested a 513-by-513 mesh for the terrain, and the generate_brownian_mesh produces a mesh of size by, where is the first argument. Note also that care was taken to ensure the population map size was the same size as the terrain map. Notice that for terrain, we use generate_terrain, but for the population data, we use generate_brownian_mesh, which is faster, but slightly less terrain-like. Title(colorbar(), 'Population Density (inhabitants/km^2)') Surf(xm, ym, max(hm, 0), 1e3*pm) % Make figure (and flatten ocean). % We can't have negative population density or people living underwater. Pm = generate_brownian_mesh(9, zeros(2, 2)) Consider the following, in which the terrain is automatically generated and a population density map is similarly generated. Other examples might include generation of the corrosion of a metallic surface or of 2D probability density distributions. This population map could be used to generate random emergencies (such that high population areas have more emergencies), stressing the simulated capabilities of the emergency response teams. These tools could be used to generate a population density map for the city, such that what might be a "mountain" when viewed as terrain is instead a high population area.
For instance, suppose one were simulating a city's ability to respond to an emergency. In addition to generating terrain, these methods can be used to generate other useful topologies. Surf(xm, ym, max(hm, 0), cm) % Make figure (and flatten ocean).Īxis equal vis3d off % Set aspect ratio and turn off axis. Note that this involves creating multiple meshes for grass, for trees, for rock, for dirt, and for water, so this function consists of numerous calls to generation methods and therefore takes substantially longer than to generate the terrain itself. Everything in between is a gradient of rockiness. Areas with 0 or lower definitely will not. That is, areas with a 1 or higher will definitely be colored as rock. E.g., steep areas are likely to be colored as rock, but the variation in the degree to which they take on rock colors and the variation in the rock color itself across the map are both meshes generated by a similar means. Note that this process uses the same type of underlying algorithm to generate the variation in different colors. We can add some natural-looking colors to this.
Note that this is zoomed in on the median off all of the generated points in order to capture the detailed middle instead of the relatively uneventful edges. Shading interp % Interpolate color across faces.Ĭamlight left % Add a light over to the left somewhere. Set(gca, 'Position', ) % Fill the figure window. We'll add lighting too to make the detail easier to see. Plot the triangles represented by all the points. Finally, we'll wrap up with a few options for controlling the result. We'll then proceed to a discussion of approximately how they work. First, we'll show how a couple of these methods might be used.
0 kommentar(er)
