Paper Title
Assessing Diseased Lettuce Leaves Using Back Propagation With Feature Subset

Abstract
Abstract - There is a compelling effort made in the recent years to face several kinds of plant detection using sensors and state-of-art techniques. Automatic detection system is becoming vital for farming practices as it is rapidly sprouting field with the emergence of many applications in agriculture. Identification of faulty leaves yield in the prevention of volume losses that could incur in agricultural product. Moreover, it is a wearying process to differentiate and identify the healthy lettuce from its abnormal ones through human capabilities as it is time-consuming and also prone to errors in the identification process. Hence, there is a need to perform this task assisted by a machine capability which makes it faster with even greater accuracy. The objective of this research work is to design an active detection system using pattern recognition approach wherein different features of healthy and diseased leaves of the lettuce are studied. The technique employed in the approach is Back-Propagation Neural network and the results of this neural network are compared on different features set. Several process including image acquisition, image pre-processing, features extraction, subset creation and BPNN classification are incorporated to achieve the effectiveness of machine learning and thus achieving the maximum accuracy. Keywords - BPNN, Feature Extraction, Feature set, Image Processing, Lettuce Disease detection