ArcMap provides us a Kernel Density Estimation tool for both point data and line data. The principle of kernel density estimation for point features is easy to understand, it is the summation of densities of all points within a bandwidth around the focus point (Silverman (p.76)):
But, how to understand the principle of Kernel Density Estimation for line data? The official website of ArcMap says that "the use of kernel function for lines is adapted from kernel function for point densities" (How Kernel Density works—Help | ArcGIS Desktop ), but there is no information about how this adaption is conducted. Could somebody help me understand the principle of kernel density estimation for line data? Thank you!