Identification of Mango Leaves by Using Artificial Intelligence
2019
MAQBOOL, Imran | QADRI, Salman | KHAN, Dost Muhammad | FAHAD, Muhammad
This research presents a classification based novel artificial intelligence approach of Mango Leaves recognition. The design andimplementation of an artificial neural network system that extracts specific shape and morphological features from mango plant leaves ofthree kinds is presented in this research. Modules of significant mango leaf image features are identified using a novel feature selectiontechnique. This technique reduces the dimensionality of the feature space leading to a simplified classification and identification schemeappropriate for real time classification systems for better results. In making the system complete, a full account is given of the necessaryimage processing methods that are applied to the binary images of mango plant leaves to ensure identification. These methods include theextraction of shapes from binary images.The proposed method inherits size and orientation invariance with respect to the image datasets and it can operate successfully even with leaves samples that are deformed due to dropout or due a number of holes drilled in them. A considerably very high classification ratioof 96% to 98% was achieved, even for the identification of deformed leaves.
显示更多 [+] 显示较少 [-]