We outline the design of a fire spread smoldering apparatus and methodology for experimental micro-fires that allow for the exploration of the effects of environmental covariates on fire spread rate and fire spotting rate distributions. The fire smoldering experiments are measured using a digital camera at a birds-eye view above the experiment. The movies are segmented and processed using anisotropic diffusion and data-sharpening to capture the ignition and extinction times of individual pixels. We demonstrate how the processed data can be used to estimate fire rate of spreads for fire spread models using nonparametric kernel regression.