Remote sensing identification of early planting information of rapeseed in mountainous areas

As a major oilseed crop with high ornamental value, Accurate and timely determination of their distribution area under cultivation is essential to ensure food security achieve sustainable development goals. The bright yellow flowers are unique feature rapeseed compared other crops. Therefore, the flower index was previously used detect on aerial images or medium-resolution satellites. However, impact its topographic terrain structure confuses planting structure, while signal weak in early stage growth. it remains challenging accurately identify southwest mountains. In this study, new mountain rape (MWRI) proposed based Sentinel-2 time series images. First, NDVI characteristics vegetation were filter out non-vegetated areas. Secondly, non-pure image elements enhanced by combining reflection variation curves red, green, NIR, SWIR bands. MRM method extract Chongqing, typical mountainous growing China. Three different indices: normalized difference yellowness (NDYI), (RYI) also calculated for comparison, validation using high-resolution interpretation samples Google earth showed that MWRI has higher recognition accuracy OA above 0.97, indices between 0.9 0.95. results indicate an effective distinguish from