The invention relates to a
coastal zone regional function long
time sequence identification method based on multi-source
big data, and the method comprises the following steps: firstly, constructing
coastal zone regional function land classification based on human production, life and
ecology, and carrying out the initial classification of a long
time sequence image through employing a
random forest algorithm based on a multi-source
big data fusion platform; secondly, spatial separation of an
impervious surface and a
water body is achieved through a scanning line seed filling
algorithm and geometric feature analysis, and functional types related to
cultivated land transformation, offshore cultivation land, a
salt pan and reclamation are corrected based on a temporal and
spatial change logic rule; and finally, according to a
classification result, change ranges and time stages of grain production, offshore culture and reclamation are extracted. According to the invention, the long-time-sequence identification precision of the regional functions of the
coastal zone is improved, and the method is particularly suitable for regional function land transformation and
change detection of reclamation construction caused by long-time-sequence, large-range and high-density offshore human activities, and can be directly applied to auxiliary
decision making of
spatial planning and regional policies of the coastal zone.