Extraction and Monitoring of Cotton Area and Growth Information Using Remote Sensing at Small Scale: A Case Study in Dingzhuang Town of Guangrao County, China

Cotton area and growth information are important basis for cotton production and management. This article took Dingzhuang Town in Guangrao County of Shandong Province as the study area and chose CBERS01 and HJ1B satellite images as the information source. Selecting similar phases with obvious cotton information, the area of cotton was acquired by decision tree classification model according to spectrum characteristics of typical objects after pre-processing. Regular changes of vegetation index and cotton growth condition in spatial and time were analyzed according to four different time remote sensing images of cotton growing season in 2009. The results showed that the extraction accuracy of cotton area was over 90%. In the past 10 years, cotton planting area increased 7529.4 hm². With the growing of cotton in every period, the growth information of cotton showed different spatial and time distribution regularities. Monitoring results were consistent with surveyed cotton yield. This study manifested that the method can timely achieve and dynamically monitor the cotton area and growth information at small-scale. It can also provide basis for early prediction of cotton production. This research has positive significance to improve the levels of cotton cultural production and management.