English

应用科学学报 ›› 2025, Vol. 43 ›› Issue (6): 990-1002.doi: 10.3969/j.issn.0255-8297.2025.06.008

• 信号与信息处理 • 上一篇    

基于零参考网络的直升机桨叶低光图像增强

郭彦纯1, 熊邦书1, 黎文超2, 温书远2   

  1. 1. 南昌航空大学 信息工程学院, 江西 南昌 330069;
    2. 北京航空航天大学 仪器科学与光电工程学院, 北京 100191
  • 收稿日期:2025-01-25 发布日期:2025-12-19
  • 通信作者: 熊邦书,教授,研究方向为图像处理、计算机视觉和直升机故障诊断。E-mail:xiongbs@126.com E-mail:xiongbs@126.com
  • 基金资助:
    国家自然科学基金（No. 62473187, No. 62365014, No. 62401244）

Low-Light Image Enhancement of Helicopter Blades Based on Zero-Shot Network

GUO Yanchun1, XIONG Bangshu1, LI Wenchao2, WEN Shuyuan2   

  1. 1. School of Information Engineering, Nanchang Hangkong University, Nanchang 330069, Jiangxi, China;
    2. School of Instrumentation and Optoelectronic Engineering, Beihang University, Beijing 100191, China
  • Received:2025-01-25 Published:2025-12-19

PDF

27

摘要： 直升机桨叶低光图像增强是直升机桨叶运动参数测量不可或缺的预处理环节。针对图像对比度低、噪声干扰、成对低光/正常光训练数据获取困难等问题，提出了一种基于零参考网络的直升机桨叶低光图像增强方法。首先，构建了基于曝光矫正S曲线的光照矫正模块，利用网络估计的最佳S曲线参数直接增强图像的光照分量；其次，设计了自适应卷积网络提取图像噪声，抑制图像噪声干扰；最后，设计了基于零参考的损失函数，用于整体网络自适应训练，避免对成对训练集的依赖。实验结果表明，本文方法在多个数据集上的客观评价指标和视觉质量均优于当前先进算法，网络计算量仅为1.24 G，满足机载条件下直升机桨叶低光图像实时增强的需求。

关键词: 低光图像增强, 零参考网络, Retinex, 桨叶图像

Abstract: Helicopter blade low-light image enhancement is an indispensable preprocessing step for helicopter blade motion parameter measurement. Aiming at the problems of low image contrast, noise interference, and the scarcity of paired low-light/normal-light training data, this paper proposes a zero-shot network-based enhancement method. Firstly, an illumination correction module based on an exposure correction S-curve is constructed, where the illumination component of the image is directly enhanced using the optimal Scurve parameters estimated by the network. Secondly, an adaptive convolutional network is designed to extract and suppress image noise. Finally, a zero-shot-based loss function is designed to enable the adaptive training of the entire network without reliance on paired training sets. Experimental results show that the objective evaluation indicators and visual quality of the proposed method on multiple datasets are superior to the state-of-the-art algorithms. Moreover, the network computing capacity required is only 1.24 G, which meets the needs of real-time enhancement of helicopter blade low-light images under airborne conditions.

Key words: low-light image enhancement, zero-shot network, Retinex, paddle image

中图分类号: