摘要
为实现苹果采摘机器人的全天候作业,对采集到的夜间图像进行了相关研究.在分析夜视图像噪声的基础上,运用小波阈值方法进行图像的降噪处理,针对阈值算法的潜在缺点,通过构造模糊阈值函数,提出小波模糊阈值的夜视图像降噪算法.实验结果表明:从视觉上看小波模糊阈值降噪方法得到的低噪图像噪点明显减少;从客观数据比较,其相对峰值信噪比有较大幅度提高.新方法显示出在夜视图像降噪方面有着独特优势,为实现苹果采摘机器人的全天候作业打下基础.
In order to achieve the round-the-clock operation of apple harvesting robot,the relevant studies on the night vision image were studied.On the basis of the analysis of noise,the wavelet theo-ry was introduced into image processing system,using the wavelet threshold method for image noise reduction processing.Aiming at the potential disadvantages of threshold algorithm,through construc-ting fuzzy threshold functions,the wavelet fuzzy threshold de-noising algorithm was proposed.The experimental results show that,from the vision,the low noise images got by wavelet fuzzy threshold de-noising method are reduced sharply,and from the relative peak signal-to-noise ratio,the advantage of new de-noising method is obvious.The new method shows the unique advantages for the night vi-sion image noise reduction,and lays a solid foundation for the round-the-clock operation of apple har-vesting robot.
出处
《华中科技大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2015年第S1期509-512 520,520,共5页
Journal of Huazhong University of Science and Technology(Natural Science Edition)
基金
国家自然科学基金资助项目(61379101)
江苏省高校优势学科建设项目
高等学校博士学科点专项科研基金资助项目(20133227110024)
江苏省普通高校研究生科研创新计划资助项目(KYLX14-1062)
关键词
采摘机器人
夜视图像
小波降噪
模糊阈值
人工光源
峰值信噪比
harvesting robot
night vision image
wavelet de-noising
fuzzy threshold
artificial light
peak signal-to-noise ratio
