A Framework Based on Nesting of Convolutional Neural Networks to Classify Secondary Roads in High Resolution Aerial Orthoimages

Abstract

Remote sensing imagery combined with deep learning strategies is often regarded as an ideal solution for interpreting scenes and monitoring infrastructures with remarkable performance levels. In addition, the road network plays an important part in transportation, and currently one of the main related challenges is detecting and monitoring the occurring changes in order to update the existent cartography. This task is challenging due to the nature of the object (continuous and often with no clearly defined borders) and the nature of remotely sensed images (noise, obstructions). In this paper, we propose a novel framework based on convolutional neural networks (CNNs) to classify secondary roads in high-resolution aerial orthoimages divided in tiles of 256 × 256 pixels. We will evaluate the framework’s performance on unseen test data and compare the results with those obtained by other popular CNNs trained from scratch.

Grammar Guided Genetic Programming for Network Architecture Search and Road Detection on Aerial Orthophotography

“…Regarding execution times, on average k-fold (k = 5) evaluation takes 52.8 h on IGN1 server (4× Tesla V100 GPUs). [43], the mean accuracies for the same dataset are reported, among others, for ResNet-50 (91.7%), VGG (93.9%) and for two different ensembles (94.6% and 95.6%, respectively). Here, ensemble refers to the combination of multiple networks into one meta-classifier, namely their results are combined to improve predictions.…”

“…It is frequent to rely on a deep ANN to perform the whole road detection/segmentation process by either developing a novel network or using a preexisting architecture. Some researchers use custom network architectures [35,36], others customize well-known models [30,[37][38][39][40], use those well-known models with transfer learning [41,42] or combine them into road-detection ensembles [43].…”

Generative Learning for Postprocessing Semantic Segmentation Predictions: A Lightweight Conditional Generative Adversarial Network Based on Pix2pix to Improve the Extraction of Road Surface Areas

“…Remotely sensed images have been used lately by researchers in machine vision applications such as object identification [1,2], detection [3], or extraction [4]. At the same time, deep learning algorithms proved to be useful for classification tasks and land use analysis [5] in satellite imagery data [6,7]-an important remote sensing application, where semantic segmentation techniques (based on supervised learning) are applied to assign a land cover class to every pixel of an image.…”

A Deep Learning-Based Solution for Large-Scale Extraction of the Secondary Road Network from High-Resolution Aerial Orthoimagery

“…Given that we are tackling a complex segmentation task with a limited dataset, we carried out the experiments on the data split describe above, allowing a higher ratio of tiles to be used for training [47][48][49].…”

“…The models take an RGB image of size 256 × 256 × 3 and output a segmentation map with the number of channels equal to the number of filters in the last transposed convolution layer. This third dimension is squeezed down using a 1-by-1 convolution layer to the number of classes and Given that we are tackling a complex segmentation task with a limited dataset, we carried out the experiments on the data split describe above, allowing a higher ratio of tiles to be used for training [47][48][49].…”