岳学军,王林惠,兰玉彬,刘永鑫,凌康杰,甘海明.基于DCP和OCE的无人机航拍图像混合去雾算法[J].农业机械学报,2016,47(s1):419-425.
Yue Xuejun,Wang Linhui,Lan Yubin,Liu Yongxin,Ling Kangjie,Gan Haiming.Algorithm of Defogging UAV’s Aerial Images Based on DCP and OCE[J].Transactions of the Chinese Society for Agricultural Machinery,2016,47(s1):419-425.
摘要点击次数: 718
全文下载次数: 320
基于DCP和OCE的无人机航拍图像混合去雾算法   [下载全文]
Algorithm of Defogging UAV’s Aerial Images Based on DCP and OCE   [Download Pdf][in English]
投稿时间:2016-07-20  
DOI:10.6041/j.issn.1000-1298.2016.S0.064
中文关键词:  图像去雾  无人机  暗通道先验  优化对比度增强  图像分割
基金项目:广东省科技计划项目(2015A020224036、2014A020208109)、国家重点研发计划项目(2016YFD0200700)和广东省水利科技创新项目(2016-18)
作者单位
岳学军 华南农业大学
国际农业航空施药技术联合实验室 
王林惠 华南农业大学
国际农业航空施药技术联合实验室 
兰玉彬 国际农业航空施药技术联合实验室
华南农业大学 
刘永鑫 华南农业大学
国际农业航空施药技术联合实验室 
凌康杰 华南农业大学
国际农业航空施药技术联合实验室 
甘海明 华南农业大学
国际农业航空施药技术联合实验室 
中文摘要:针对无人机在雨雾天气下的农田航拍图像退化问题,考虑无人机自身特性,提出了一种基于DCP和OCE的无人机航拍图像混合去雾算法。首先判断原始图像天空区域的存在,利用Canny边缘检测算法对带天空区域的原始图像进行分割并做高斯羽化处理,再采用膨胀和腐蚀等形态学操作进行二值化区域填充,去除非天空区域中对应亮度低的区域,提取天空区域和非天空区域。非天空区域图像采用基于导向滤波的暗通道先验算法去雾。天空区域图像采用基于代价函数的优化对比度算法去雾。本试验分别从主观视觉性和无参考量化评测两方面对100幅航拍图像去雾结果作出评价,试验结果表明,所提算法在对带雾图像去雾后,出现Halo现象的概率相较于DCP算法降低了95%,其综合评测均值指数提高了26%,去雾效果明显,色彩还原度高,没有明显的过渡区域和偏色现象,处理速度可达33帧/s,平均速度相较于DCP算法提高了32%,能满足实时性要求。
Yue Xuejun  Wang Linhui  Lan Yubin  Liu Yongxin  Ling Kangjie  Gan Haiming
South China Agricultural University;International Joint Laboratory of Agricultural Aviation Application Techniques,South China Agricultural University;International Joint Laboratory of Agricultural Aviation Application Techniques,International Joint Laboratory of Agricultural Aviation Application Techniques;South China Agricultural University,South China Agricultural University;International Joint Laboratory of Agricultural Aviation Application Techniques,South China Agricultural University;International Joint Laboratory of Agricultural Aviation Application Techniques and South China Agricultural University;International Joint Laboratory of Agricultural Aviation Application Techniques
Key Words:image defogging  unmanned aerial vehicle  dark channel prior  optimized contrast enhancement  image segmentation
Abstract:Aiming at degeneration of UAV’s farmland aerial images in rainy or foggy weather, this article proposed an integrated method based on dark channel prior(DCP) and optimized contrast enhancement(OCE). Firstly, this method identified the existence of sky from the original image, and then segmented and feathered the original images in Gauss method with Canny edge detection algorithm. Secondly, the non sky regions, where luminance value was low, were filled with binarized dilation and morphological erosion removed. Finally, both non sky and sky regions within a frame were processed by dark channel and optimized contrast enhancement algorithm respectively. In this research, the effect of the proposed algorithm was evaluated by no reference quantitative assessment using 100 aerial photographing images. The experimental results showed that the possibility of Halo effect, compared to DCP, was reduced by 95% while its comprehensive evaluation average index was raised by 26%. Consequently, without obvious transition region and color cast, the proposed method filtered the fog of the aerial images well and the color reproduction was high. Meanwhile, the average processing efficiency in the proposed algorithm was up to 32% (33 frame per second) higher than classical DCP. Hence, the method could met the real time requirements.

Transactions of the Chinese Society for Agriculture Machinery (CSAM), in charged of China Association for Science and Technology (CAST), sponsored by CSAM and Chinese Academy of Agricultural Mechanization Science(CAAMS), started publication in 1957. It is the earliest interdisciplinary journal in Chinese which combines agricultural and engineering. It always closely grasps the development direction of agriculture engineering disciplines and the published papers represent the highest academic level of agriculture engineering in China. Currently, nearly 8,000 papers have been already published. There are around 3,000 papers contributed to the journal each year, but only around 600 of them will be accepted. Transactions of CSAM focuses on a wide range of agricultural machinery, irrigation, electronics, robotics, agro-products engineering, biological energy, agricultural structures and environment and more. Subjects in Transactions of the CSAM have been embodied by many internationally well-known index systems, such as: EI Compendex, CA, CSA, etc.

   下载PDF阅读器