Remote sensing image road track correctness checking method and device based on deep learning

• Main Authors: YUAN SHENGGU, DING YONGQIANG, LI CHENG, SHENG GUANGXIAO, QIN LUOYU, CAI HONGYUE
• Format: Patent
• Language: Chinese; English
• Published: 22.12.2020
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The embodiment of the invention provides a remote sensing image road track correctness checking method, device and equipment based on deep learning and a computer readable storage medium. The method comprises the following steps: acquiring road check reference remote sensing image data; inputting the reference image into a pre-trained road identification model to obtain a corresponding road identification result; and checking according to the to-be-checked road vector data and the road identification result. In this way, a wrong road track can be quickly positioned, and the method is particularly suitable for remote areas such as mountainous areas and environments with hard on-site checking operation conditions. 本公开的实施例提供了基于深度学习的遥感影像道路轨迹正确性核查方法、装置、设备和计算机可读存储介质。所述方法包括获取道路核查参考遥感影像数据；将所述参考影像输入预先训练得到的道路识别模型中，得到对应的道路识别结果；根据待核查道路矢量数据与所述道路识别结果进行核查。以此方式，可以快速定位到道路错误轨迹，特别适合于山区等偏远地区、现场核查作业条件艰苦的环境。