摘要
针对水下介质分布不均匀引起的光吸收和散射导致图像颜色失真、对比度低、细节模糊,严重影响ORBSLAM2算法前端特征提取与匹配鲁棒性的问题,文中提出一种基于颜色校正与改进的自适应对比度增强的图像处理方法。首先,利用一种改进的颜色通道补偿和颜色通道拉伸方法去除色偏;其次,采用改进的自适应对比度增强方法提高图像的亮度与对比度;最后,将彩色校正图像与对比度增强图像在HSV空间中融合。此外,将提出的算法与一些其他算法进行主客观的评价,并将处理好的图片进行特征提取和匹配。结果表明,该算法处理的图片不仅在主客观评价上均优于对比算法,而且增加了特征点的匹配数量,为水下图像处理提供了借鉴。
The light absorption and scattering caused by the uneven distribution of underwater media lead to image color distortion,low contrast and blurred details,which seriously affects the robustness of front-end feature extraction and matching of ORB-SLAM2 algorithm,so an image processing method based on color correction and improved adaptive contrast enhancement is proposed.An improved color channel compensation and stretching method is utilized to remove color cast first,and then an improved adaptive contrast enhancement method is adopted to improve the brightness and contrast of the image.Finally,the color-corrected image and the contrast-enhanced image are fused in the HSV(hue,saturation,value)space.In addition,the proposed algorithm and other algorithms are evaluated objectively and subjectively,and the processed images are used for feature extraction and matching.The results show that the images processed by the proposed algorithm not only outperform the comparison algorithms in both objective and subjective evaluations,but also increase the number of matched feature points.To sum up,the proposed algorithm provides reference for underwater image processing.
作者
刘明
肖汉城
LIU Ming;XIAO Hancheng(School of Electrical and Information Engineering,Yunnan Minzu University,Kunming 650500,China)
出处
《现代电子技术》
北大核心
2025年第1期40-46,共7页
Modern Electronics Technique
基金
国家自然科学基金资助项目(52061042)。
关键词
水下视觉SLAM
水下图像增强
自适应对比度增强
颜色校正
颜色补偿
特征匹配
underwater visual SLAM
underwater image enhancement
adaptive contrast enhancement
color correction
color compensation
feature matching
