Automatic Recognition of Rice Fields from Multitemporal Satellite Images

A technology for automatic recognition of rice fields by using multitemporal satellite images is proposed. The principle of this technology is applying a region-based classification by means of integrating geographical data and domain knowledge with multitemporal Images. Based on the principle, three methods of investigating the temporal NDVI profile to detect rice fields were implemented. They are Profile Matching (PM), Peak Detection (PD), and Difference Classification (DC). All of the methods were tested on a set of multitemporal SPOT XS images (12 epochs) collected during the second rice season of 1993. Comparing to the traditional supervised classification using a single image epoch, all the methods can easily improve the accuracy about 20%. The PM and PD methods can also determine the planting time of a rice field, but they do not provide a better result than the DC method. When the number of image epochs is small the PM and PD methods may not work, but the DC method works well even if there are only 2 or 3 epochs. All of the methods do not require any training data for classification. We expect that this approach will dramatically reduce the needs of human work and increase the efficiency of the rice inventory work.