Road Extraction from Remote Sensing Images Using Parallel Softplus Networks

Road extraction from remote sensing images plays an important role in traffic management, urban planning, automatic vehicle navigation and emergency management. It is a hot issue that how to extract effectively road information from remote sensing images. Here, a new model, namely parallel softplus network (PSNet), has been proposed, which uses parallel network structure and softplus activation function. Specially, the model uses a new weight initialization for extraction effectiveness. Moreover, compared with the popular models, it extracts more complete and continuous road information on the same road remote sensing images. Meanwhile, it outperforms other extraction models, with a high F1-score . Experimental results indicate that it is a promising model, which effectively extracts road information from remote sensing images with a little noise.

[1]  Chang Luo,et al.  A novel softplus linear unit for deep convolutional neural networks , 2017, Applied Intelligence.

[2]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[3]  H. Kerem Cigizoglu,et al.  Integration of Information for Environmental Security , 2008 .

[4]  Geoffrey E. Hinton,et al.  Machine Learning for Aerial Image Labeling , 2013 .

[5]  Min Wang,et al.  Very high resolution remote sensing image classification with SEEDS-CNN and scale effect analysis for superpixel CNN classification , 2018, International Journal of Remote Sensing.

[6]  S. C. Kremer,et al.  Gradient Flow in Recurrent Nets: the Difficulty of Learning Long-Term Dependencies , 2001 .

[7]  George Papandreou,et al.  Rethinking Atrous Convolution for Semantic Image Segmentation , 2017, ArXiv.

[8]  Corso Elvezia.  Bridging Long Time Lags by Weight Guessing and \long Short Term Memory" , 1996 .

[9]  Yuantao Jiang.  Research on road extraction of remote sensing image based on convolutional neural network , 2019, EURASIP J. Image Video Process..

[10]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[11]  The Role Of Remote Sensing And Gis For Security , 2008 .

[12]  Ami Wiesel,et al.  Learning to Detect , 2018, IEEE Transactions on Signal Processing.

[13]  Christian Heipke,et al.  EVALUATION OF AUTOMATIC ROAD EXTRACTION , 2007 .

[14]  Seyed Majid Azimi,et al.  Road Segmentation in SAR Satellite Images With Deep Fully Convolutional Neural Networks , 2018, IEEE Geoscience and Remote Sensing Letters.

[15]  Ying Wang,et al.  Accurate urban road centerline extraction from VHR imagery via multiscale segmentation and tensor voting , 2015, Neurocomputing.

[16]  Jitendra Malik,et al.  Learning to detect natural image boundaries using local brightness, color, and texture cues , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[17]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[18]  Sepp Hochreiter,et al.  Untersuchungen zu dynamischen neuronalen Netzen , 1991 .

[19]  Yoshua Bengio,et al.  Gradient Flow in Recurrent Nets: the Difficulty of Learning Long-Term Dependencies , 2001 .

[20]  Dar A. Roberts,et al.  Using paired thermal and hyperspectral aerial imagery to quantify land surface temperature variability and assess crop stress within California orchards , 2019, Remote Sensing of Environment.

[21]  Lennart Svensson,et al.  LIDAR-Camera Fusion for Road Detection Using Fully Convolutional Neural Networks , 2018, Robotics Auton. Syst..