摘要
基于试验设计方法实现了番茄采摘机器人夜间照明系统设计。提出了一种基于前景与背景类间方差和类内方差比值F的图像可分割性评价指标作为确定最佳试验方案即最佳照明系统的评价指标;考虑了光源种类、光源布局及图像采集距离3个试验因子;采用正交试验表L_（18）（6×3^6）安排试验。试验结果表明：光源种类和光源布局是影响番茄采摘机器人夜间照明系统的显著因子,图像采集距离为不显著因子;光源种类因子各水平中,荧光灯照射时图像的F最大,为2.159;光源布局因子各水平中,对角布局时图像的F最大,为2.234。因此,所设计的番茄采摘机器人夜间照明系统的最佳组合为：荧光灯、对角布局。将试验结果与基于归一化R-G色差的OTSU自动阈值图像分割算法的分割效果进行了对比,对比结果验证了基于该图像可分割性评价指标F的夜间照明系统设计方法的有效性。
A design method for night lighting system of tomato picking robots based on experiments design was presented. In the design of experiments,an image segmentation evaluation index,which could determine the optimal night lighting system,was proposed. It was calculated as the ratio of F between intra-class and inter-class variances of foreground and background of tomato images captured at night.Three factors,including light source type,light source layout and image acquisition distance,were considered. Moreover,orthogonal experiment table L_（18）（ 6 × 3~6） was used to arrange experiments. Testing results showed that,the light source type and light source layout were two significant factors for night lighting system,while the image acquisition distance was not. So,it is focused on the influence of light source type and light source layout in the night lighting system. Meanwhile,the F value of fluorescent lamp was 2. 159 which was the highest among those values of six types of light source type,and that of the diagonal layout was 2. 234 which was the highest among the values of three types of light source layout. Therefore,in the current trial,the optimal light was fluorescent lamp,and the best light source layout was diagonal layout for the night lighting system of tomato picking robot. F values of tomato images captured at night were compared with image segmentation results using the OTSU algorithm based on normalized R-G color difference. Comparative results showed that the image segmentation results were better for images with higher F. The image segmentation evaluation index based on F and the design method for night lighting system are effective.
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2016年第7期8-14,共7页
Transactions of the Chinese Society for Agricultural Machinery
基金
国家自然科学基金项目(61105024)
关键词
番茄
采摘机器人
夜间照明系统
试验设计
图像分割
tomato
picking robot
night lighting system
experiment design
image segmentation
