Fast eye localization from thermal images using neural networks

Abstract

The paper presents a fast algorithm for eye localization from thermal images. Due to blurred edges, lower quality of thermal images compared with visible-light images, the process may be more complicated. On the other hand, it seems relatively easy to designate specific areas of the eyes because of their relatively highest brightness resulting from the highest temperature in these face areas. It turns out, however, that the highest pixel brightness does not always determine unequivocally the position of the eyes. Therefore, the algorithm described here proposes a specialized method for segmentation of images containing eye areas and a specialized classifier. The first part of the algorithm is a block for generating features describing eye areas. The second stage is a decision block built based on a neural network allowing for effective classification of pre-designated areas. A suitable combination of these blocks into a single system allowed for correct analysis of images with rich geometric diversity (scale, position, orientation) and brightness distribution, reaching more than 91 % of correct localizations and time of analysis of a single image of hundredths of a second. Algorithms of this type can be used as the initial stages of facial recognition, emotion recognition, quantitative analysis of the face or modern human-computer interfaces. The described algorithm may be used for both conventional thermal images and thermograms (after prior conversion of temperature values into brightness values), which affects its potential application in medicine. The benefits of this study are twofold: fast localization of characteristic points of the eyes and the possibility of processing despite the typical problems of thermovision and image transformations.

Using Facial Landmark Detection on Thermal Images as a Novel Prognostic Tool for Emergency Departments

“…This study has investigated how to do eye localization on thermal images. The study builds on the fact that the eyes have the highest brightness in the face area, but also points out that this does not necessarily mean that the single highest pixel value in the image is equal to the eye area (Marzec et al, 2016).…”

Infrared dataset generation for people detection through superimposition of different camera sensors

“…In [25] neural networks are used to classify objects in thermal images for search and rescue missions using UAVs. Deep neural networks are exploited yet again in [26] for fast eye tracking from thermal images. Finally, infrared images are also being used to enhance face recognition applications [27].…”

Where Do We Stand in the Domestic Dog (Canis familiaris) Positive-Emotion Assessment: A State-of-the-Art Review and Future Directions

“…Moreover, we would like to stress again the great potential and importance of affective computing in animal emotion research, especially in recognizing subtle differences among emotional responses or by detecting specific patterns in time series of single biosignals or by combining multimodal input data (Caridakis et al, 2007;Yin et al, 2017;Quesnel et al, 2018). For example, deep learning might be particularly helpful in developing a newer methodology for fast detection of regions of interest and accurate, reliable data extraction and analyses in infrared imaging (Marzec et al, 2016;Sonkusare et al, 2019).…”