无人机遥感在仁沐新高速公路项目水土保持监测中的应用与评价
Application and Evaluation of UAV Remote Sensing in Soil and Water Conservation Monitoring of Renmuxin Expressway Project

无人机遥感技术在水土保持监测工作中的应用尚未形成统一成熟的方案，本文利用无人机遥感技术对仁沐新高速公路项目进行水土保持监测，旨在为该技术应用于水土保持监督管理、施工监理和水土保持验收等方面作出适用性评价并提供经验支撑和研究建议。结果表明，利用无人机遥感技术，能够准确获取扰动土地范围、水土流失面积、土地利用类型和水土保持措施等监测信息，并及时掌握监测区内的水土流失隐患和危害，计算结果的精度基本满足《生产建设项目水土保持监测规程(试行)》的要求。提升了工作效率和监测结果的精准性。但无人机遥感在水土保持监测中仍有两方面需要进一步研究：(1) 利用计算机的基准面测量法对弃渣场体积进行计算会导致结果存在偏差，软件测算体积和方量的方法仍需进一步研究；(2) 非监督分类影像处理结果精度较低，图像处理软件的自动识别仍需进一步细化。
The application of UAV remote sensing technology in soil and water conservation monitoring has not yet formed a unified and mature scheme. In this paper, UAV remote sensing technology is used to monitor soil and water conservation of Renmuxin expressway project. The purpose of this paper is to evaluate the applicability and provide empirical support and research suggestions for the application of this technology in soil and water conservation supervision and management, construction supervision and soil and water conservation acceptance. The results show that the monitoring information such as disturbed land area, soil erosion area, land use type and soil and water conservation measures can be accurately obtained by using UAV remote sensing technology, and the hidden dangers and hazards of soil and water loss in the monitoring area can be grasped in time. The accuracy of the calculation results basically meets the requirements of the soil and Water Conservation Monitoring regulations for production and Construction projects (trial). The work efficiency and the accuracy of monitoring results are improved. However, there are still two aspects that need to be further studied in UAV remote sensing in soil and water conservation monitoring: (1) The calculation of the volume of abandoned slag field by using the datum measurement method of computer will lead to the deviation of the results, and the method of calculating volume and square quantity by software still needs to be further studied; (2) The accuracy of unsupervised classification image processing results is low, and the automatic recognition of image processing software still needs to be further refined.