No abstract

The paper is devoted to the development of an effective semantic segmentation algorithm for automation of airport infrastructure labelling in RGB satellite images. This task is addressed using algorithms based on deep convolutional artificial neural networks, as they have proven themselves in a wide range of tasks, including the terrestrial imagery segmentation, where they show consistently high results. A new dataset was labelled for this particular task and a comparative analysis of different architectures and backbones was carried out. A conditional random field model (CRF) was used for postprocessing and accounting of contextual information and neighborhood of objects of different classes in order to eliminate outliers. Features of the solutions applied at all preparatory stages of the algorithm were described, including data preparation, neural network training and post-processing of the training results.

No abstract

Semantic segmentation is one of the important ways of extracting information about objects in images. State of the art neural network algorithms allow to perform highly accurate semantic segmentation of images, including aerial photos. However, in most of the works authors use high-quality low-noise images. In this work, we study the ability of neural networks to correctly segment images with intensive uncorrelated Gaussian noise. The study brings us three main conclusions. Firstly, it demonstrates that neural network algorithms are capable of working with extreme image distortions without using additional filtration or image recovery techniques. Secondly, the experiments quantitatively show that distortion intensity can be negated with increased training set size. Such process is similar to model’s quality improvement and generalization due to training dataset enlargement. Finally, we quantitatively demonstrate how image aggregation techniques affect training with noised data.