Plant Leaf Sickness Recognition Using Image Processing

This study demonstrates the use of image processing techniques in detection of diseases in plants. The system, which we have demonstrated in this technique, is a software result for computerised detection and computing of the texture enumeration for the leaf diseases in plants. This processing system involves four major steps: (1) in this step, we take an RGB image as input and it’s colour transformation structure is created, (2) in the second step, we mask the green pixels and then we remove them using well defined threshold value, (3) the image subdivision and the extraction of the useful segments is done the third step, (4) then last step involves the calculation of the texture statics. This helps in the determination of the disease, if any.

Pair of Retina Recognition System Using Hopfield Neural Network Algorithm

image processingA process of pair of retina recognition system using feature fusion method has been proposed in this paper. Left and right retinal images of human have been used for the inputs of this retinal recognition system. Wavelet based retinal image pre-processing technique has been applied to process the retina images. After extracting the features from the left and right retinal images, features are concatenated using feature fusion technique. Principal Component Analysis based dimensional reduction technique has been applied to reduce the dimension of the combined feature vector. Finally, Hop field neural network algorithm has been used for classification. DRIVE retinal database has been used for the performance measurements of the proposed pair of retina recognition system.