Automatic method for detection of characteristic areas in thermal face images

Abstract

The use of thermal images of a selected area of the head in screening systems, which perform fast and accurate analysis of the temperature distribution of individual areas, requires the use of profiled image analysis methods. There exist methods for automated face analysis which are used at airports or train stations and are designed to detect people with fever. However, they do not enable automatic separation of specific areas of the face. This paper presents an algorithm for image analysis which enables localization of characteristic areas of the face in thermograms. The algorithm is resistant to subjects' variability and also to changes in the position and orientation of the head. In addition, an attempt was made to eliminate the impact of background and interference caused by hair and hairline. The algorithm automatically adjusts its operation parameters to suit the prevailing room conditions. Compared to previous studies (Marzec et al., J Med Inform Tech 16:151-159, 2010), the set of thermal images was expanded by 34 images. As a result, the research material was a total of 125 patients' thermograms performed in the Department of Pediatrics and Child and Adolescent Neurology in Katowice, Poland. The images were taken interchangeably with several thermal cameras: AGEMA 590 PAL (sensitivity of 0.1°C), ThermaCam S65 (sensitivity of 0.08°C), A310 (sensitivity of 0.05°C), T335 (sensitivity of 0.05°C) with a 320×240 pixel optical resolution of detectors, maintaining the principles related to taking thermal images for medical thermography. In comparison to (Marzec et al., J Med Inform Tech 16:151-159, 2010), the approach presented there has been extended and modified. Based on the comparison with other methods presented in the literature, it was demonstrated that this method is more complex as it enables to determine the approximate areas of selected parts of Multimed Tools Appl (2015)

Human Psychophysiological State Recognition Based on Analysis of Thermograms of Face and Neck Regions

“…Шея. На основании известной информации о размере головы, антропометрическом соотношении областей лица (так называемого «золотого сечения» лица) [19,20] и результатах анализа примеров термограмм в работе [21] был предложен детектор, который в сочетании с модификациями, описанными в работе [22] позволяет выде-лить такие участки лица, как лоб, брови, глаза и нос. Выделение областей рта и шеи базируется на геометрии лица: область рта ограничивалась областью носа (вверху) и последующим сужением в области шеи (внизу).…”

Fast eye localization from thermal images using neural networks

“…In order to carry out the tests, cropped images sized 30 × 30 pixels located near the centre of the head were selected - Fig.1. Regions of the centre of the head were determined automatically based on the previously developed solutions [9,10].…”

“…The silhouette area was selected and the approximate centre of the head was determined using the methods described in [9,10]. Determination of the centre of the head and a cropped image covering the centre sized 30 × 30 pixels provided the test images.…”

“…The exact localization of certain elements of the face opens up possibilities for further analysis of this type of images. Examples of using such methods in the case of thermal imaging can be found in biometric systems [1,2], as an initial block before the proper recognition process, or in medicine [9,10], where the process of locating facial areas allows for further automated analysis of thermograms. They can be also used in the creation of human-computer interfaces whose task is to facilitate cooperation with the computer [3] or increase safety [6].…”

“…A particularly important element of the face are the eyes because they carry a large amount of information [7,19]. Therefore, many described solutions are directly related to eye localization [18,8,13,20,21] or eye localization is part of a larger system of analysis of the entire face [1,10,14,19]. The algorithm for eye localization from thermal images presented in this paper enables to automate this process.…”