Analysis Of Vegetation Classification Algorithms On Satellite Images And Medical Images

In recent days, the remote sensing algorithms are used in the medical field for improving the visualization of the medical images. Because, the medical images are generally in the gray scale image format for better visualization the colour Doppler or spectrograms are used but they are expensive. To overcome this drawback the remote sensing algorithm is applied to the medical images to group the pixels and visualize in different colours. The image processing techniques is used to classify the vegetation region into 16 samples. The image pre-processing is done by Wiener filter to remove the noise. Feature extraction is carried out by Grey Level Co-occurrence Matrix (GLCM) and the spectral bands are optimized by Particle Swarm Optimization (PSO) .The classification of vegetation region is classified by Extreme Learning Machine. In this, the comparisons of the remote sensing algorithms like IRVM-MFO, ELM-DF and ELM-PSO for the Indian pines and Salinas Dataset. Among these the ELM- Dragon Fly algorithm produced the best results for both the sets. Hence, this ELM-DF is applied to the Brain tissue region segmentation. In this paper the analysis is performed to find the efficient method for vegetation classification by comparing with other methods. Simulations are carried out on two datasets such as Indian Pine and Salinas scene. Performance metrics such as accuracy, specificity, and sensitivity have been evaluated that show the efficiency of the proposed classifier.