Date Palm Crop Yield Estimation – A Framework

Abstract

Saudi Arabia is the home land of the date  palm tree and the dates are considered to be one of the  most important national products. As the dates are part  of their heritage, therefore, Saudi Arabia is the largest  consumer of dates. Saudi Arabia has more than fifty  date-processing facilities, which process large amounts of  these products. Right now, Saudi Arabia ranks second in  the production of dates. There are more than twenty-five  million date palm trees that cover more than 150,000  hectares of land in Saudi Arabia. Date production is  estimated to be more than 1.1 million tons each year  which accounts for around seventy two percent of the  total agricultural output of Saudi Arabia. It is very vital  to predict the yield so that stakeholders will be prepared  to market their product in a better way. Crop yield  estimation can be done either through conventional  method or through image processing methods. The  former are often costly, complex, time consuming methods, that cannot be applied on a large scale. So, it is  essential to employ those methods for crop yield  estimation that are time saving as well as cost effective  and image processing method fulfils these conditions.  Since image processing extracts different features from  an image that can be used not only in recognizing  different types of crops but also estimating their yield.  Recently, crop yield estimation have been developed  using Artificial Neural Networks (ANN) have exhibited  improved performance and self-adaptability as compared  to traditional statistical methods. Keeping in view the  importance of this topic, this paper presents a  framework for crop yield estimation through image  processing by using the ANN.