Abstract
Abstract. Deep convolutional neural networks have dramatically changed the landscape of the modern computer vision. Nowadays methods based on deep neural networks show the best performance among image recognition and object detection algorithms. While polishing of network architectures received a lot of scholar attention, from the practical point of view the preparation of a large image dataset for a successful training of a neural network became one of major challenges. This challenge is particularly profound for image recognition in wavelengths lying outside the visible spectrum. For example no infrared or radar image datasets large enough for successful training of a deep neural network are available to date in public domain. Recent advances of deep neural networks prove that they are also capable to do arbitrary image transformations such as super-resolution image generation, grayscale image colorisation and imitation of style of a given artist. Thus a natural question arise: how could be deep neural networks used for augmentation of existing large image datasets? This paper is focused on the development of the Thermalnet deep convolutional neural network for augmentation of existing large visible image datasets with synthetic thermal images. The Thermalnet network architecture is inspired by colorisation deep neural networks.
Highlights
Thermal cameras provide a robust solution for object detection and scene understanding
Nowadays methods based on deep neural networks show the best performance among image recognition and object detection algorithms
An intensive scholar attention to the field of convolutional neural networks (CNN) stimulated the development of extremely large image datasets with ground truth labelling of tens million of images (Deng et al, 2009, Lin et al, 2014, Everingham et al, 2015)
Summary
Thermal cameras provide a robust solution for object detection and scene understanding. Thermal imaging is widely exploited in the field of autonomous driving, where it helps to improve object detection rates significantly. Nowadays methods based on deep neural networks show the best performance among image recognition and object detection algorithms. CNN provide flexible solution for object detection in multispectral images. For a successful learning of a CNN a large training dataset with thousands of images is required. Most of datasets that are available online include only images captured in visible spectrum and could not be used for training of multispectral CNN. The key step in the development of the new generation of multispectral object detection algorithms using CNN is the generation of large multispectral datasets. The CNN was trained using NVIDIA DIGITS. (NVIDIA, 2016)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
