ABSTRACT:2D image matching problem is often stated as an image-to-shape or shape-to-shape matching problem. Such shape-based matching techniques should provide the matching of scene image fragments registered in various lighting, weather and season conditions or in different spectral bands. Most popular shape-to-shape matching technique is based on mutual information approach. Another wellknown approach is a morphological image-to-shape matching proposed by Pytiev. In this paper we propose the new image-to-shape matching technique based on heat kernels and diffusion maps. The corresponding Diffusion Morphology is proposed as a new generalization of Pytiev morphological scheme. The fast implementation of morphological diffusion filtering is described. Experimental comparison of new and aforementioned shape-based matching techniques is reported applying to the TV and IR image matching problem.
Commission II, WG II/5KEY WORDS: infrared images, augmented reality, object recognition, deep convolutional neural networks
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.
Shape-based matching techniques should provide the matching of scene image fragments registered in various lighting, weather and season conditions or in different spectral bands. The most popular shape-to-shape matching technique is based on a mutual information approach. Another well-known approach is a morphological image-to-shape matching proposed by Pytiev. In this paper we propose a new image-to-shape matching technique based on heat kernels and diffusion maps. The corresponding Diffusion Morphology is proposed as a new generalization of Pytiev morphological scheme. The fast implementation of morphological diffusion filtering is described. An experimental comparison of the newly proposed and aforementioned image-to-shape and shape-to-shape matching techniques as applied to the TV and IR image matching problem is made.
An original method for object detection based on morphlet trees is proposed in the paper. It allows the robust detection of heterogeneous objects in images to be done without pre-training. Besides, the detection process simultaneously includes a preliminary segmentation, which can be later used for recognition. Also, there is another important characteristic: the proposed approach does not require the use of sliding windows and feature pyramids to detect different-scale objects.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.