基于GF-1影像的水稻苗情长势监测研究

Study on Rice Growth Monitoring Based on GF-1 Images

为了研究GF-1影像在水稻长势监测中的可行性,利用GF-1影像对信阳市部分县区进行水稻苗情的长势监测。该过程通过样点实地采集LAI与对应的GF-1影像提取EVI建立统计关系模型,选取拟合效果最佳的指数型函数进行区域LAI的反演,并依据相应生育期LAI分类标准划分苗情长势。结果表明,研究区2013年水稻抽穗期一类苗占68.56%,二类苗占24.92%,三类苗占6.52%,2014年拔节期一类苗占69.75%,二类苗占25.95%,三类苗占4.30%,苗情分类图能够为田间管理决策提供重要参考信息。表明利用GF-1影像进行地理条件复杂的水稻分布区的长势监测能够取得良好的应用效果,但仍需要大量地面调查数据不断修正改进模型,从而提高监测准确性。

In order to test the adaptability of GF-1 images in monitoring rice growth,rice growth was moni-tored using GF-1 images in parts of counties of Xinyang. It was achieved by establishing corresponding statistical relationship model between field collection LAI and GF-1 EVI of sample points,then selecting the best fitted exponential model to invert LAI and divide rice growth according to the classification crite-ria of LAI. In the year of 2013,the first,second and third class of rice seedlings in the study area accoun-ted for 68. 56%,24. 92% and 6. 52%,respectively,and in 2014,the first,second and third class of rice seedlings in the study area accounted for 69. 75%,25. 95% and 4. 30%,respectively. The classification diagram of rice seedlings growth provided an important reference of field management and decision-mak-ing information. GF-1 images had been made good application in rice growth monitoring in terrain complex region,but there still needed a lot of ground investigation data to improve model and monitoring accuracy.