深度学习遥感图像标注系统研究

随着深度学习的发展,遥感影像处理技术也从传统机器学习算法向深度学习转变,然而,用于遥感图像的训练数据集却十分稀少,且数据标注困难.本文将GIS技术与图像标注技术相结合,基于Flask Web框架设计一个可用于海量遥感数据的标注系统.该系统可用于海量遥感数据的数据框标注、数据类别标注,以及目标关键点标注,同时能将标注数据...

Research on Deep Learning Remote Sensing Image Annotation System

With the development of deep learning,remote sensing image processing technology has changed from traditional machine learning algorithm to deep learning.However,the training data set for remote sensing image is very rare,and the data annotation is difficult.In this paper,GIS and image annotation technology are combined to development a remote sensing annotation system.Based on the flask web framework,a annotation system for massive remote sensing data is designed from the bottom.The system can be used for data frame annotation,data category annotation and target key point annotation of massive remote sensing data.At the same time,the labeling data can be exported to the most commonly used data set format such as COCO style and VOC2007 style for deep learning training.