An Approach For Road Network Detection From Satellite Images Using Neural Networks
Humans can identifycityobjectseasilyin remotesensing images, but this has turned out to be a problem forcomputers.In this paper, we propose an approach for road detection as objectdetection model based on neural networks from satellite images. The model is based on Multilayer Perceptron (MLP) which is one of the most preferred artificial neural network architecture in classification and prediction problems. According the neural network, the RGB values and measures of central tendency are used for deciding the pixel belongs to road or not. The output image generated with results.
Index Terms—Object Detection, Road Extraction, Neural Networks