Road extraction method of high-resolution remote sensing image based on multi-branch cascaded cavity space pyramid

• Main Authors: LIU AILIN, LI XUNGEN, MA QI, LI ZIXUAN, PAN MIAN, ZHANG ZHAN, LYU SHUAISHUAI, MEN FEIFEI, ZHOU SHANGCHAO
• Format: Patent
• Language: Chinese; English
• Published: 02.10.2020
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention discloses a road extraction method for a high-resolution remote sensing image based on a multi-branch cascaded cavity space pyramid, and the method comprises the following steps: S1, carrying out the data preprocessing: carrying out the data amplification of data in a Massachusetts data set; S2, building a model, extracting a feature map of the remote sensing road image by using a convolutional neural network, extracting feature information on the road image in combination with a multi-branch cascaded cavity space pyramid, and performing parallel sampling by the multi-branch cascaded cavity space pyramid on given input by cascaded cavity convolution with different cavity ratios, which is equivalent to capturing context information of the image in multiple modes; S3, designinga loss function to optimize network parameters, and establishing an evaluation index F1 and an evaluation index MIOU; and S4, testing the test sample set by using the trained model, and obtaining a final segmented image thro