Enhancing Depth Image Estimation for Underwater Robots by Combining Image Processing and Machine Learning

TL;DR

This paper proposes a method to improve underwater depth estimation by enhancing image quality through processing techniques before applying deep learning models, addressing challenges posed by underwater conditions.

Contribution

It introduces an integrated approach combining image enhancement and deep learning for more accurate underwater depth estimation.

Findings

Enhanced image processing improves depth estimation accuracy

The approach performs well in low-light underwater environments

Results show significant improvement over baseline methods

Abstract

Depth information plays a crucial role in autonomous systems for environmental perception and robot state estimation. With the rapid development of deep neural network technology, depth estimation has been extensively studied and shown potential for practical applications. However, in particularly challenging environments such as low-light and noisy underwater conditions, direct application of machine learning models may not yield the desired results. Therefore, in this paper, we present an approach to enhance underwater image quality to improve depth estimation effectiveness. First, underwater images are processed through methods such as color compensation, brightness equalization, and enhancement of contrast and sharpness of objects in the image. Next, we perform depth estimation using the Udepth model on the enhanced images. Finally, the results are evaluated and presented to verify the effectiveness and accuracy of the enhanced depth image quality approach for underwater robots.