一种融合面向对象与深度学习的地表覆盖监测成果质检技术

A quality inspection technology for surface coverage monitoring results combining the object-oriented and deep learning

地表覆盖监测成果为自然资源调查监测和统计提供了持续、可靠的参考数据。本文结合地表覆盖监测成果应用和自然资源调查监测的实际需求,针对地表覆盖分类成果检查的关键质量要求,提出了一种以矢量数据为基础,融合面向对象和深度学习等遥感影像智能化信息提取技术辅助分类信息正确的质量检查技术方法,并结合工程实践开展了技术路线验证。结果表明,该方法不仅为地表覆盖监测成果质检提供了技术支撑,也提高了地表覆盖分类检查的效率与正确性,可为自然资源调查监测工程质量控制提供技术参考。

Land cover monitoring results provide continuous and reliable reference data for natural resources investigation,monitoring and statistics.This paper combines the application of land cover monitoring results and the actual demand for natural resources investigation and monitoring.Aiming at the key quality requirements of land cover classification results inspection,a quality inspection technology method based on vector data,fusion of the object-oriented and deep learning and other remote sensing image intelligent information extraction technologies to assist the accuracy of classification information is proposed.And combined with engineering practice the technical route verification is carried out.The results show that this technical method not only provides technical support for the quality inspection of the land cover monitoring results,but also improves the efficiency and accuracy of the inspection of the correctness of the land cover classification,and can provide technical references for the quality control of natural resources survey and monitoring projects.