A tool for climate smart crop insurance: Combining farmers’ pictures with dynamic crop modelling for accurate yield estimation prior to harvest

• 查看报告

作者：

B.K.Singh; DulalChakraborty; NaveenKalra; JayaSingh

关键词：

ClimateChange;  Resilience;  RiskandInsurance; 

年份：

2018

总页数：

27

语种：

English

信息来源：

国际粮食政策研究所（IFPRI）

摘要：

The study found that dynamic crop models have the accuracy to predict normal to high yields, but there are limits to their ability to capture low yields. On the other hand, the machine learning (CNN) model has better ability to capture lower yields. It is worth noting that the crop model only took into consideration mainly the weather data to predict yields; it is handicapped by the paucity of detailed management information deployed by farmers. However, the pictures sent by farmers reflected more yield-determining characteristics that reflected crop health and yield and that were then captured by the CNN. Finally, among the picture characteristics parameters, if “GCC & H” correlations are high, this could be a good indicator of low yield.