This invention discloses a method and
system for monitoring rice planting area based on multi-source SAR data. The method includes
data acquisition and preprocessing, preliminary extraction of rice planting area, dynamic inversion of physiological parameters, phenological and
ripening system identification, and result output. On the one hand, this invention integrates multi-
source data to overcome the bottlenecks of cloud and
fog conditions and lighting conditions in optical
remote sensing. Simultaneously, it uses a
strongly coupled design and
algorithm based on segmentation and classification iterations, along with a linkage optimization mechanism, to improve adaptability to fragmented plots and complex
terrain, thus solving the problem of low planting area accuracy in complex scenarios. On the other hand, based on the linkage
adaptation and cross-validation of area extraction optimization results with physiological parameter inversion and phenological
ripening system identification, a two-way feedback
closed loop is used to complete secondary optimization. This avoids interference from non-rice plots, improving the accuracy of parameter inversion and phenological identification, while also verifying the rationality of area extraction through the latter, achieving a synergistic improvement in overall monitoring accuracy, exceeding 93%.