An Enhanced Approach of Underwater Image Processing for Species Classification

Due to the various environmental conditions that surround seas, underwater object detection is a challenging undertaking. Diverse strategies had been employed to boost the characteristics of pictures taken under the water. Pre-processing of images and extraction of features methods have been proposed in this study. The picture's spatial properties have been enhanced with the use of pre-processing. The enhancement of image is a major concern before extracting the features of an image. A robust algorithm has been implemented which converts the raw image into enhanced version. To improve the intensity, contrast, and sharpness, an enhancement algorithm has been devised. Furthermore, a comparison analysis that extracts the feature utilizing feature extraction approaches has been presented. The main goal of pre-processing the image was by enhancing the picture by applying histogram equalization methods. The features extracted from the enhanced version of the image using the feature extraction technique i.e., SURF (Speeded up robust features). The error rate and time consumption has been taken to compare the results. The outcomes shows that rate of error and time consumption by the proposed approach proves to be better than SIFT and PCA techniques.