Present-day high-performance imagery equipment allows the capture of images with as high as 6Mhz frame rate that enables the capture and analysis of physical phenomena such as neural activity and shock waves[2] to deforestation and effects of global warming [3]. Still, the process requires a huge amount of human intervention to achieve high levels of accuracy in terms of analysis. Automation these fields is a widely researched area to provide accurate analysis[4] and therefore solutions.
Imperceptible variations in human body when magnified can give information about the state of a person like a heartbeat rate, which can be used for diagnosis. One change that happen in the human body due to blood circulation or rhythmic beat of the heart is the change in colour of the skin. This variation can not be perceived by human eyes. If the spatial flow patterns of the blood can be extracted from a sequence of images or video, the pulse rate can be estimated. The colour details are better extracted from the forehead and cheeks of a person. Apart from the skin colour/tone, variations in subtle motions of different body parts[5] like the blood vessels in the wrist can also be detected and the pulse rate can be extracted using temporal and spatial processing. This work aims at capturing the temporal variations and extracting temporal signals of interest to estimate the pulse rate from a video.
The work is presented as follows. Chapter 2 discusses the related work in this field. Chapter 3 elaborates on the proposed work and the application of Eulerian Video Magnificationalgorithm for enhancing the subtle motions. Chapter 4 presents the experimental results followed by a Conclusion in Chapter 5.
[2] K Goda, KK Tsia, and B Jalali, “Serial time-encoded amplified imaging for real-time observation of fast dynamic phenomena”,
Nature, vol. 458, no.7242, 2009, pp.1145–1149.
[3] Google Earth Engine. http://earthengine.google.org/.
[4] William T. Freeman and Wojciech Matusik, “Analyzing images through time.”, NSF Award 1111415, September 2011.
[5] Michael Rubinstein, Neal Wadhwa, Fredo Durand, and William T. Freeman, “Revealing invisible changes in the world”,
Science, vol. 339, no. 6119,February 2013, pp. 519.
