摘要: |
海雾气象条件下船只高精度检测识别面临较大困难,传统的目标识别、定位方法效果差强人意。本文围绕海雾气象条件下不同类型船只的实时检测问题,提出一种基于YOLOv3深度学习的实时海上船只检测新思路。首先构建清晰图片和模糊图片(海雾、雨)的判别方法,实现图片清晰度分类处理;其次为提高海雾气象条件下海上船只的实时检测精度,消除海雾遮挡对目标识别的影响,运用暗通道先验去雾方法对含有海雾的图像实行去雾;最后基于YOLOv3深度学习算法对精细处理后的图像进行船只实时检测。实验结果表明该方法能够在海雾气象条件下高效、准确地检测到船只,对海上复杂环境条件下的船只实时检测研究具有一定的理论指导意义。 |
关键词: 船只检测 暗通道先验去雾 深度学习 YOLOv3 |
DOI: |
分类号: |
基金项目:国家自然科学基金项目(面上项目,重点项目,重大项目) |
|
Real-time detection of marine vessels under sea fog weather conditions based on YOLOv3 deep learning |
wangfei, liuxueqin, qinzhiliang, mabenjun, zhengyi
|
HARBIN ENGINEERING UNIVERSITY COLLGE OF UNDERWATER ACOUSTIC ENGNEERING
|
Abstract: |
High-precision detection and recognition of vessels under sea fog weather conditions face greater difficulties, and traditional target recognition methods are difficult to achieve satisfactory results. This paper focuses on the detection of different types of vessels under complex sea fog weather, and proposes a new idea of real-time marine vessel detection based on YOLOv3 deep learning. First, a discriminative method for clear pictures and fuzzy pictures (sea fog, rain) is constructed to realize the classification processing of picture sharpness. Then, in order to improve the detection precision of marine vessels under complex sea fog weather dark channel prior dehazing algorithm is used to suppress the impact of sea fog occlusion for target recognition. Finally, real-time detection of vessels is performed on the finely processed images based on the YOLOv3 deep learning algorithm. The experimental results show that the method can be used to detect vessels efficiently and accurately under sea fog weather conditions, which have certain theoretical guidance significance for the research of vessel target recognition under complex marine conditions. |
Key words: Vessel detection Dark channel prior dehazing algorithm Deep learning YOLOv3 |