Improving Road Safety during Nocturnal Hours by Characterizing Animal Poses Utilizing CNN-Based Analysis of Thermal Images

Animal–vehicle collision is a common danger on highways, especially during nighttime driving. Its likelihood affected not only by the low visibility hours, but also unpredictability of animals’ actions when vehicle nearby. Extensive research has shown that lack hours can be addressed using thermal imaging. However, to our knowledge, little been undertaken predicting animal action through an animal’s specific poses while moving. This paper proposes new system couples use two-dimensional convolutional neural network (2D-CNN) and image input, determine risk imposed in pose passing automobile hours. The proposed was tested set images presenting real-life scenarios animals roadside found classify accurately 82% cases. Overall, it provides valuable basis for implementing automotive tool minimize animal–vehicle collisions