Skin perfusion pressure is a key parameter for diagnosing peripheral artery disease, measuring the point at which blood flow returns to the skin after sufficient external occlusion pressure is applied. This study explores skin perfusion pressure estimation using multi-spectral imaging and multi-wavelength phtoplethysmography data from the dominant foot of ten individuals. DC maps were generated for each participant using multi-linear regression model based on multi-wavelength photoplethysmography data. We evaluated the performance of similarity metrics, such as Bhattacharyya Distance, Mutual Information, and Jensen-Shannon Divergence in estimating skin perfusion pressure by analyzing changes in the DC maps during an occlusion protocol. Results indicate that Jensen-Shannon Divergence exhibited lower percentage errors, around 10%. Meanwhile, estimates using red wavelength photoplethysmography data as the dependent variable in the regression models showed better approximations, with percentage errors ranging from 9.05% to 10.64%. In contrast, estimates using Mutual Information demonstrated higher errors, ranging from 13.20% to 22.91%, likely due to foot movement. Future studies should consider image registration to mitigate the impact of foot movement on Mutual Information estimates and explore implementation in populations with peripheral artery disease.