智能除草装备苗草模式识别方法研究
|
摘要
精准苗草识别是靶向施药除草装备作业基础。为提高识别算法精度及效率,解决光照变化对识别图像分割精度影响,文章优化研究分割算法,引入加权系数,提高算法光照适应性;根据作物线性分布生长特点,采用烟花智能群体算法,对垄间杂草与作物识别与定位;田间图像采集与试验结果表明,加权分割方法可有效解决光照变化对分割效果影响,实际作物与垄间杂草识别率为98.7%和89.5%,满足苗草识别与导航要求,对导航技术与智能除草装备发展具有重要意义。
Accurate and efficient identification of seedling is the basis for application of target weeding equipment. In order to improve the accuracy and efficiency of the recognition algorithm and solve the influence of illumination variation on the recognition accuracy of image segmentation, firstly, the segmentation algorithm was optimized, and the weighting coefficient was introduced to improve the illumination adaptability of the algorithm. Secondly, the fireworks intelligent group algorithm was used to identify and locate the inter-row weeds and crops between the ridges according to the growth characteristics of crop linear distribution. Finally, the field images were collected and tested. The results showed that the weighted segmentation method could effectively solve the influence of illumination changes on the segmentation effect, the grass recognition rate was 98.7% and 89.5%, which met the requirements of seedling identification and navigation, and has important reference significance for the development of navigation technology and intelligent weeding.
Accurate and efficient identification of seedling is the basis for application of target weeding equipment. In order to improve the accuracy and efficiency of the recognition algorithm and solve the influence of illumination variation on the recognition accuracy of image segmentation, firstly, the segmentation algorithm was optimized, and the weighting coefficient was introduced to improve the illumination adaptability of the algorithm. Secondly, the fireworks intelligent group algorithm was used to identify and locate the inter-row weeds and crops between the ridges according to the growth characteristics of crop linear distribution. Finally, the field images were collected and tested. The results showed that the weighted segmentation method could effectively solve the influence of illumination changes on the segmentation effect, the grass recognition rate was 98.7% and 89.5%, which met the requirements of seedling identification and navigation, and has important reference significance for the development of navigation technology and intelligent weeding.
引文
[1]王金武,多天宇,唐汉,等.水田株间立式除草装置除草机理与试验研究[J].东北农业大学学报, 2016, 47(4):86-94.
[2] Meyer G E, Neto J C. Verification of color vegetation indices for automated crop imaging applications[J]. Computers&Electronics in Agriculture, 2008, 63(2):282-293.
[3] Guyer D E, Miles G E, Gaultney L D, et al. Application of machine vision to shape analysis in leaf and plant identification[J].Transactions of the Asae, 1993, 36(1):163-171.
[4] Meyer G E. Design of an optical weed sensor using plant spectral characteristics[J]. Transactions of the Asae, 2000, 44(2):409-419.
[5]边肇祺,张学工.模式识别[M]. 2版.北京:清华大学出版社,2000.
[6]王璨,李志伟.利用融合高度与单目图像特征的支持向量机模型识别杂草[J].农业工程学报, 2016, 32(15):165-174.
[7]陈亚军,赵博,李树君,等.基于多特征的杂草逆向定位方法与试验[J].农业机械学报, 2015, 46(6):257-262.
[8]岑益郎.计算机视觉技术在杂草识别中的应用[D].杭州:浙江大学, 2006.
[9]张红旗,刘宇,王春光.改进田间杂草识别图像预处理方法的研究[J].农机化研究, 2015(6):70-73.
[10]马锃宏.田间作物对靶喷施机器人关键技术研究[D].北京:中国农业大学, 2015.
[11]高国琴,李明.基于K-means算法的温室移动机器人导航路径识别[J].农业工程学报, 2014, 30(7):25-33.
[12]张志斌,罗锡文,臧英,等.基于颜色特征的绿色作物图像分割算法[J].农业工程学报, 2011, 27(7):183-189.
[2] Meyer G E, Neto J C. Verification of color vegetation indices for automated crop imaging applications[J]. Computers&Electronics in Agriculture, 2008, 63(2):282-293.
[3] Guyer D E, Miles G E, Gaultney L D, et al. Application of machine vision to shape analysis in leaf and plant identification[J].Transactions of the Asae, 1993, 36(1):163-171.
[4] Meyer G E. Design of an optical weed sensor using plant spectral characteristics[J]. Transactions of the Asae, 2000, 44(2):409-419.
[5]边肇祺,张学工.模式识别[M]. 2版.北京:清华大学出版社,2000.
[6]王璨,李志伟.利用融合高度与单目图像特征的支持向量机模型识别杂草[J].农业工程学报, 2016, 32(15):165-174.
[7]陈亚军,赵博,李树君,等.基于多特征的杂草逆向定位方法与试验[J].农业机械学报, 2015, 46(6):257-262.
[8]岑益郎.计算机视觉技术在杂草识别中的应用[D].杭州:浙江大学, 2006.
[9]张红旗,刘宇,王春光.改进田间杂草识别图像预处理方法的研究[J].农机化研究, 2015(6):70-73.
[10]马锃宏.田间作物对靶喷施机器人关键技术研究[D].北京:中国农业大学, 2015.
[11]高国琴,李明.基于K-means算法的温室移动机器人导航路径识别[J].农业工程学报, 2014, 30(7):25-33.
[12]张志斌,罗锡文,臧英,等.基于颜色特征的绿色作物图像分割算法[J].农业工程学报, 2011, 27(7):183-189.