测绘通报 ›› 2025, Vol. 0 ›› Issue (3): 133-137,167.doi: 10.13474/j.cnki.11-2246.2025.0323

• 技术交流 • 上一篇    

改进的DeepLabV3+模型用于四川县域大小春作物识别

张璇1,2, 杨本勇1,2, 文武3, 邓维熙1,2, 周何帆1,2   

  1. 1. 自然资源部第三地理信息制图院, 四川 成都 610100;
    2. 自然资源部数字制图与国土信息应用 重点实验室, 四川 成都 610100;
    3. 四川省都江堰水利发展中心, 四川 成都 611830
  • 收稿日期:2024-10-31 发布日期:2025-04-03
  • 通讯作者: 杨本勇。E-mail:615727320@qq.com
  • 作者简介:张璇(1988—),女,硕士,研究方向为自然资源遥感监测。E-mail:kris325@qq.com
  • 基金资助:
    基于遥感的县域农业灌溉用水定量估算分析项目

Improved DeepLabV3+ model for spring crops identification in Sichuan counties

ZHANG Xuan1,2, YANG Benyong1,2, WEN Wu3, DENG Weixi1,2, ZHOU Hefan1,2   

  1. 1. Third Geoinformation Mapping Institute, Ministry of Natural Resources, Chengdu 610100, China;
    2. Ministry of Natural Resources Key Laboratory of Digital Cartography and Land Information Application, Chengdu 610100, China;
    3. Sichuan Dujiangyan Irrigation Project Water Conservancy Development Center, Chengdu 611830, China
  • Received:2024-10-31 Published:2025-04-03

摘要: 传统的农业灌溉水资源分配模式存在严重的分配不均及浪费现象,通过遥感技术准确获取作物空间分布数据能解决农业灌溉水资源分配中作物分布缺失的问题。本文以四川省成都市新津区为研究区,引入对比学习和特征增强机制改进DeepLabV3+模型,使用GF系列影像数据实现大小春作物的准确识别。结果表明,改进后的IM-DeepLabV3+模型在对新津地区油菜、小麦、水稻、玉米的识别精度上有所提升,分别达91.73%、89.93%、80.18%、72.08%,能够为农业灌溉水资源科学分配提供科学的作物分布数据支撑。

关键词: 农业遥感, 对比学习, 深度学习, 四川, 作物识别

Abstract: Traditional agricultural irrigation water resource allocation models suffer from significant inefficiencies and waste due to uneven distribution. Remote sensing technology can effectively address the issue of missing crop distribution data in irrigation planning by providing accurate spatial distribution information. This paper takes Xinjin district of Chengdu, Sichuan province as the research area. A mechanism which contains contrastive learning and feature enhancement is introduced to improve the DeepLabV3+ model, utilizing GF series satellite imagery to accurately identify both major and minor crops. The results indicate that the improved IM-DeepLabV3+ model can achieve recognition accuracies of 91.73%, 89.93%, 80.18%, and 72.08% for rapeseed, wheat, rice, and corn, respectively, which can provide scientific crop distribution data support for scientific allocation of agricultural irrigation water resources.

Key words: agricultural remote sensing, contrastive learning, deep learning, Sichuan, crop identification

中图分类号: