Abstract
This chapter describes efficient visual analysis methods towards the unobtrusive monitoring of elderly people using common low-cost hardware. In particular, two real-time approaches are demonstrated for (a) Pupil Size Estimation and (b) Pulse Rate Estimation. In both cases, a main concern in the development has been to keep low the overall complexity, so that the hosting hardware that executes the proposed methods is left with free resources, in order to be able to also execute other processes in the context of a health monitoring framework. In addition, all experiments have been carried out using typical, low-cost visual sensors, i.e. simple webcams. Finally, the respective benchmarks needed to evaluate the two methodologies are described in this chapter.
TopIntroduction
With the recent increase of the elderly population, especially in most of the developed countries, computerized health monitoring can provide useful information to clinicians and carers. More particularly, unobtrusive every day health monitoring can be an important component of every day home automation systems. In addition, to allow for unobtrusive monitoring to gain acceptance, one should seek efficient methods of monitoring using common low-cost hardware. This chapter focuses on presenting visual analysis techniques for two important physiological measurements namely: (a) pupil size estimation and (b) pulse rate estimation. Pulse rate is probably the most basic medical measurement, while pupil size can be a valuable source of information, since, apart from pathological cases, it can reveal emotional state, fatigue and the ageing as well.
In both cases, the presented algorithmic schemata, which are adopted in order to analyze visual information and extract the two physiological signs, need to satisfy the following conditions:
- 1.
Unobtrusiveness: This is achieved through the nature of the acquisition of the visual information. The user can be monitored by a camera while in front of a laptop or behind a mirror which certainly falls into an unobtrusive scenario.
- 2.
Low-Cost Infrastructure: The acquisition of the visual information in the presented methods is achieved through a typical web camera. In addition, the main processing procedures are executed in a low-cost computer unit.
- 3.
Low Computational Complexity: In the context of a general health monitoring framework, other procedures may have to be executed on the same computer. Therefore it is essential to implement fast procedures that leave free resources in the hosting hardware, so that other processes can also be executed in parallel.
The chapter is organized as follows:
- •
Section 2 provides a general background description for both tasks of pulse rate estimation and heart rate estimation.
- •
Section 3 describes the methodology followed in the heart rate estimation task along with the respective benchmark which has been devised to evaluate the method, as well as performance evaluation results.
- •
Section 4 describes the pupil size estimation method along with the resulting performance measures and results.
- •
The final section presents the future and ongoing work along with the concluding remarks.
Key Terms in this Chapter
IRIS: A circular area of the eye which is responsible for controlling the diameter and size of the pupil.
Viterbi Algorithm: A dynamic programming method that aims at finding the most likely sequence of hidden states that results in a sequence of observed events. This optimal sequence is usually referred to as the Viterbi path.
(Color) Saturation: The degree of difference between the color and gray in respect to the color's own brightness. Strongly related to colorfulness and chroma. In simple words, as saturation increases, colors appear more pure.
Color Constancy: A set of methodologies and techniques applied on an image to ensure that the perceived color of a scene remains constant under varying illumination conditions.
Sclera: The white area of the eye, a protective outer layer
Pupil: A hole located in the center of the iris of the eye. It is responsible for allowing light to enter the retina.
Pulse Oximeter: A non-invasive method for monitoring a patient's oxygen saturation based on the photoplethysmography principle.
Photoplethysmography: A technique to detect blood volume changes in the microvascular bed of tissue using visual analysis methods.