Recognition of Plants by Leaves Digital Image and Neural Network

To identify different plants by leaves digital image is one key problem in precision farming. By the combination of image processing and neural network, Most of the image blocks of different plants could be correctly classified. Firstly, the image enhancement processing can make objects in the source image clear. Secondly, due to the different shapes and sizes of image blocks of leaves, they could be separated and extracted from sources. Then, by using image analysis tools from Matlab, these characters such as sizes, radius, perimeters, solidity, and eccentricity could be calculated. Then, using them as input data, create a radial basis function neural networks. Divide the input data into two parts. Select one part to train the network and the other to check the validity of the model. Finally, input data from other image frames under the same condition could be used to check the model. In this work, the total accuracy is about 80%. These methods was simple and highly effective, So they could easily be integrated into auto machines in the field, which can largely saving labor and enhance productivity.