Multi View Face Detection in Cattle Using Infrared Thermography

Publication Type:
Chapter
Citation:
2020, 1174 CCIS pp. 223 - 236
Issue Date:
2020-01-01
Filename Description Size
MultiViewFaceDetectioninCattleUsingInfraredThermography-revised_paper.pdfPublished version1.92 MB
Adobe PDF
Full metadata record
© 2020, Springer Nature Switzerland AG. Face detection in thermal imaging has been used widely in human for different purposes such as surveillance, obtaining physiological reading: respiratory and heart rate from face region via thermal imaging. Physiological reading via infrared thermal imaging used in emotion and stress detection as well as polygraph analysis. In animal as general and cattle in specific, face region localized manually in order to obtain temperature for eyes, nose and mouth, which used for stress, diseases and inflammation detection. In order to develop a future automated system for monitoring health conditions in cattle, it required to detect the face region automatically. Based on author knowledge, there is no research done regarding face detection in cattle using infrared thermal images. Unlike the human, cattle keep roaming, which lead to a change in the face and body orientation. The main objective of this paper is proposing a new method for Multi-view face detection in cattle with accuracy enhancement by using three classifiers and temperature thresholding. Classifiers are established by using Histogram Oriented Gradient (HOG) as features and Support vector machine (SVM) for classification. The results show that the proposed algorithm is performing well in term of Specificity, Recall and F-measure and detection rate compare to the currently used method in the literature.
Please use this identifier to cite or link to this item: