基于Unet模型的河湖“四乱”样本数据集

本数据集基于Unet模型的Google Earth遥感影像“四乱”目标识别方法，针对黄河流域“四乱”问题，利用Google Earth遥感影像构建样本库，并利用Unet模型进行“四乱”目标的自动检测与识别，能够大大提高“四乱”问题目标的检测精度与效率，从而为黄河流域的生态环境保护和高...

This dataset is established for the target recognition method of the "four chaos" (four types of illegal activities: illegal channel occupation, illegal sand excavation, illegal waste piling, and illegal construction) in Google Earth remote sensing images based on the Unet model. Aiming at the "four chaos" issues in the Yellow River Basin, a sample library is constructed using Google Earth remote sensing images, and the Unet model is utilized to automatically detect and recognize the targets of "four chaos". This approach can greatly improve the detection accuracy and efficiency of targets related to the "four chaos" issues, thereby providing support for the ecological environment protection and high-

搜集汇总

背景与挑战

背景概述

该数据集是针对黄河流域河湖“四乱”（乱占、乱建、乱堆、乱采）问题构建的遥感影像样本库，基于Google Earth影像采集了932张原始图片，并通过数据增强扩展至9320张，用于训练和验证Unet模型以实现自动检测。数据集采用Unet模型进行语义分割，在训练样本较少的情况下表现出高性能，模型验证精度达0.962，旨在提升“四乱”目标检测的精度和效率，为黄河流域生态保护提供支持。