An RGB camera or webcam can be used to accurately estimate the physiological features of the biomedical symptomatic issue from the signals received. In this regard, our current work proposes a method based on the most well-known image and multi-dimensional signal processing algorithms to estimate the pulse and breathing rates from facial skin by photoplethysmography wave examination. The current study uses signal processing and image processing techniques to eliminate noise that can be caused by head movements, poor signal strength, changes in lighting, or acquisition devices. To collect the most PPG data feasible, we will base our image processing on the three distinct color spaces: RGB band, Luv band, and Lab band on the one hand, and skin detection on the other. Afterward, the signal processing part combines two well-known methods: fast-PCA, a distributed PCA algorithm, and serial-EMD, a method for multi-dimensional signal decomposition. The latter method can achieve higher accuracy and a decomposition time that is more reduced compared with the existing methods. Our results of the extracted PPG waves and the IMF modes signals are depicted further in the result section.