Object recognition from underwater Images using deep cnn approach

Abstract

newline exploration, and underwater autonomous operation is becoming more and more essential to stay safe in the risky high-pressure deep-sea environment, because it is challenging to restore underwater images, the researcher is interested in light scattering and absorption phenomena to generate distortion in the images. newlineHowever, because of color distortion, noise, and dispersion, these methods have some difficulty recognizing underwater objects. Therefore, improving the underwater imagination is crucial for correctly identifying underwater things. The SSAG optimization-based deep CNN is developed in this research for image enhancement models to address this problem, the grouping behavior and adaptability of the guards and flocks, which serve as the search agents in the proposed optimization, are integrated to create the suggested social situation-aware guard optimization.