Identifikasi Jenis Penyakit Daun Tanaman Jagung Menggunakan Jaringan Saraf Tiruan Berbasis Backpropagation

Riyan Putra Ramadhan, Noveri Lysbetti Marpaung

Abstract


The identification of corn plant leaves can be done manually using human eye vision because based on the physical characteristics of the corn leaves that are affected by the disease will undergo changes in the shape and color of the leaves. However, it has a weakness when the corn leaves will be identified in many quantities and each age has a different judgment of the colors seen. In general, for the identification of diseases that attack the leaves of corn plants is processed by utilizing Digital Image Processing consisting of four main parts, namely Image acquisition, Preprocessing, Extraction of color traits, and Classification. The picture of corn leaves is taken using the camera, with a total of 200 images of corn leaves that have been infected by the disease, training and testing data on the system. The feature extraction method used is a Color Moment as the extraction of color features to get the value Mean, Standard Deviation, and Skewness as the input data for neural network processes. The classification method on this system uses a Backpropagation-based Neural Network with Matlab R2018a. The result of identification of disease type leaves of corn plants are four: Bercak Leaves, Bulai Leaves, Hawar Leaves, and Karat Leaves. The system is able to detect the disease of corn plant leaves with an accuracy rate of 90% and error 10%.
Keyword : Image Processing, Preprocessing, Color Moment, Backpropagation, Matlab R2018a

Full Text:

PDF

Refbacks

  • There are currently no refbacks.