基于多源信息融合的土壤含水量估算

遥感信息在大面积土壤水分监测中具有不可替代的优势。通过对试验区域的气象数据、土壤类型数据、土壤和水体的光谱特征曲线、多时相遥感影像数据等进行预处理，提取图像信息和属性数据，并对土地利用类型和植被覆盖度进行划分。基于土壤的光谱响应机制建立像元反射光谱信息分解模型，以此计算出该区域土壤容积含水率。结果表明该方法对于低植被区的监测精度较高(理论精度89.78%)，可作为土壤水分监测预警的依据。

Soil Moisture Estimation Based on Multi-source Information Fusion

Remote Sensing (RS) based soil moisture monitoring and estimating has advantages than conventional methods in large scale land survey and environmental evaluations due to its convenience in data acquisition.The procedure of soil moisture estimation based on multi\|source information fusion is proposed.First,the multi\|source data,such as weather information,soil type,spectrum characteristic curves and RS images are preprocessed to get the attribute data.The land utilization classification and vegetation cover intensity are calculated based on the formerly obtained data and the real survey data.Then a spectral decomposition model is proposed for soil moisture estimation.A case study is provided and the computational results of soil moisture show that the approximate estimation precision is 89.78% in lands with lower vegetation cover intensity.Moreover,the computational results fit fairly well with the measured data and show that the proposed model is feasible to be used for drought monitoring and early warning.