在夜间低照度环境下，轮胎式龙门起重机(RTG)常常需要执行自动驾驶导航任务，针对导航过程中地面引导线图像边缘难以检测的问题，提出一种彩色加权引导滤波-Retinex(CWGIF-Retinex)算法.首先，有针对性地改进加权引导滤波算法的权值估计方案，用矢量梯度方法提取经过Retinex算法处理的彩色图像边缘作为权值；进而，利用CWGIF算法估算引导线图像的照度分量，并通过单尺度Retinex算法计算得到反射分量；最后，通过算法计算得到色彩与轮廓边缘均增强的引导线图像.实验证明，CWGIF-Retinex算法相较于传统算法能够更好地增强弱光照区域的色彩，同时还能起到消除光晕伪影、颜色保真和增强引导线边缘对比度的效果.
Aiming at solving the difficulties in night-time edge detection of guideline during the autopilot navigation of rubber tyred gantry (RTG), a color weighted guided image filtering-Retinex (CWGIF-Retinex) algorithm is proposed. Firstly, the weight estimation of weighted guided image filtering is improved by using vector gradient method. And the edges of color images processed by Retinex are extracted as weights. Secondly, the illuminance components of guideline image are estimated by using CWGIF algorithm. And then, the reflection components are calculated by using the single-scale Retinex algorithm. Finally, an image with enhanced colors and edges is obtained. The results of experiments indicate that CWGIF-Retinex algorithm compared with traditional algorithms can enhance the colors of low illumination area while achieving the goals of removing halo effect, color fidelity, and enhancing the contrast of guideline edges.
［1］蒋阳, 曹杰, 豆泽阳, 等.基于截尾全变差Retinex算法的图像复原技术［J］.兵器装备工程学报, 2018, 39(5): 125-128.
JIANG Yang, CAO Jie, DOU Zeyang, et al. Image restoration based on truncated total variation for retinex algorithm［J］. Journal of Ordnance Equipment Engineering, 2018, 39(5), 125-128.
［2］张瑞. 基于Retinex理论的图像增强算法研究［D］.重庆: 重庆邮电大学, 2017.
ZHANG Rui. Research on image enhancement algorithm based on retinex theory［D］. Chongqing: Chongqing University of Posts and Telecommunications, 2017.
［3］肖创柏, 赵宏宇, 禹 晶.基于引导滤波的Retinex快速夜间彩色图像增强技术［J］.北京工业大学学报, 2013, 39(12): 1868-1873.
XIAO Chuangbai, ZHAO Hongyu, YU Jing. Rapid retinex algorithm for night color image enhancement based on guided filtering［J］. Journal of Beijing University of Technology, 2013, 39(12): 1868-1873.
［4］LI Z G, ZHENG J H, ZHU Z J, et al. Weighted guided image filtering［J］. IEEE Transactions on Image Processing, 2015, 24(1): 120-129.
［5］龙 鹏, 鲁华祥.LoG边缘算子改进的加权引导滤波算法［J］.计算机应用, 2015, 35(9): 2661-2665.
LONG Peng, LU Huaxiang. Weighted guided image filtering algorithm using Laplacian-of-Gaussian edge detector［J］. Journal of Computer Applications, 2015, 35(9): 2661-2665.
［6］HE K, SUN J, TANG X O. Guided image filtering［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(6): 1397-1409.
［7］DI ZENZO S.A note on the gradient of a multi-image［J］. Computer Vision, Graphics, and Image Processing, 1986, 33(1): 116-125.
［8］LAND E H. The retinex theory of color vision［J］. Scientific American, 1977, 237(6): 108-128.
［9］RAHMAN Z, JOBSON D J, WOODELL G A, et al. Image enhancement, image quality, and noise ［C］ //Photonic Devices and Algorithms for Computing VII. San Diego, United States: SPIE, 2005: 59070N.
［10］李钊洋.视频监控夜间图像增强技术的研究［D］.广州: 华南理工大学, 2015.
LI Zhaoyang. Research on nighttime image enhancement for video surveillance［D］. Guangzhou: South China University of Technology, 2015.
