Authors: Randale Vaishnavi, Dr.Swaroopa Shastri
Abstract: Sub-sea image recovery is challenging because water has special optical properties, including absorption and scattering, that compromise the quality of the image. The captured images and videos frequently suffer from two displeasing problems: First, color distortion; and second, poor visibility. This is mainly because that the light is exponentially attenuated while penetrating through water and the strength of attenuation is color dependent. This study introduces a hybrid unsupervised method for underwater image restoration, integrating Support Vector Machines (SVM) with traditional Decision Tree algorithm. The suggested approach utilizes the capabilities of SVM in classification to improve the performance of Decision Tree methods in the restoration process. SVM is used in this approach to classify different underwater environments and conditions to facilitate more accurate utilization of restoration methods appropriate for each class. Decision Tree algorithm, in contrast, adjusts restoration parameters dynamically using the classifications given by the SVM. This hybrid model aims to improve color correction, contrast enhancement, and visibility restoration in underwater imagesResults show that the hybrid SVM and Decision Tree algorithm method surpasses classic Decision Tree algorithms based on visual quality and quantitative measures e.g., Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM). The comparison of a SVM And Decision Tree Algorithm, The SVM accuracy is 92% and The Decision Tree Algorithm 98%,The highest accuracy is Decision Tree algorithm.
DOI: http://doi.org/10.5281/zenodo.16742906.
