TY - JOUR
T1 - Road Structure Refined CNN for Road Extraction in Aerial Image
AU - Wei, Yanan
AU - Wang, Zulin
AU - Xu, Mai
N1 - Publisher Copyright:
© 2017 IEEE.
PY - 2017/5
Y1 - 2017/5
N2 - In this letter, we propose a road structure refined convolutional neural network (RSRCNN) approach for road extraction in aerial images. In order to obtain structured output of road extraction, both deconvolutional and fusion layers are designed in the architecture of RSRCNN. For training RSRCNN, a new loss function is proposed to incorporate the geometric information of road structure in cross-entropy loss, thus called road-structure-based loss function. Experimental results demonstrate that the trained RSRCNN model is able to advance the state-of-the-art road extraction for aerial images, in terms of precision, recall, F-score, and accuracy.
AB - In this letter, we propose a road structure refined convolutional neural network (RSRCNN) approach for road extraction in aerial images. In order to obtain structured output of road extraction, both deconvolutional and fusion layers are designed in the architecture of RSRCNN. For training RSRCNN, a new loss function is proposed to incorporate the geometric information of road structure in cross-entropy loss, thus called road-structure-based loss function. Experimental results demonstrate that the trained RSRCNN model is able to advance the state-of-the-art road extraction for aerial images, in terms of precision, recall, F-score, and accuracy.
KW - Convolutional neural network (CNN)
KW - machine learning
KW - road extraction
UR - https://www.scopus.com/pages/publications/85015620235
U2 - 10.1109/LGRS.2017.2672734
DO - 10.1109/LGRS.2017.2672734
M3 - 文章
AN - SCOPUS:85015620235
SN - 1545-598X
VL - 14
SP - 709
EP - 713
JO - IEEE Geoscience and Remote Sensing Letters
JF - IEEE Geoscience and Remote Sensing Letters
IS - 5
M1 - 7876793
ER -