4 September 2015 Using artificial neural network and satellite data to predict rice yield in Bangladesh
Author Affiliations +
Abstract
Rice production in Bangladesh is a crucial part of the national economy and providing about 70 percent of an average citizen’s total calorie intake. The demand for rice is constantly rising as the new populations are added in every year in Bangladesh. Due to the increase in population, the cultivation land decreases. In addition, Bangladesh is faced with production constraints such as drought, flooding, salinity, lack of irrigation facilities and lack of modern technology. To maintain self sufficiency in rice, Bangladesh will have to continue to expand rice production by increasing yield at a rate that is at least equal to the population growth until the demand of rice has stabilized. Accurate rice yield prediction is one of the most important challenges in managing supply and demand of rice as well as decision making processes. Artificial Neural Network (ANN) is used to construct a model to predict Aus rice yield in Bangladesh. Advanced Very High Resolution Radiometer (AVHRR)-based remote sensing satellite data vegetation health (VH) indices (Vegetation Condition Index (VCI) and Temperature Condition Index (TCI) are used as input variables and official statistics of Aus rice yield is used as target variable for ANN prediction model. The result obtained with ANN method is encouraging and the error of prediction is less than 10%. Therefore, prediction can play an important role in planning and storing of sufficient rice to face in any future uncertainty.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Kawsar Akhand, Kawsar Akhand, Mohammad Nizamuddin, Mohammad Nizamuddin, Leonid Roytman, Leonid Roytman, Felix Kogan, Felix Kogan, Mitch Goldberg, Mitch Goldberg, } "Using artificial neural network and satellite data to predict rice yield in Bangladesh", Proc. SPIE 9610, Remote Sensing and Modeling of Ecosystems for Sustainability XII, 96100E (4 September 2015); doi: 10.1117/12.2186261; https://doi.org/10.1117/12.2186261
PROCEEDINGS
14 PAGES


SHARE
Back to Top