基于根茬检测的秋后残膜回收导航路径提取方法

doi:10.13229/j.cnki.jdxbgxb20200500

摘要/Abstract

摘要：

针对秋后残膜回收作业视觉导航中的路径规划问题，提出一种作业路径提取方法。提取根茬、残膜和行间茎秆碎叶三类图片的颜色空间特征和纹理特征，按8∶2构成训练集和测试集；以检测根茬为目标，搭建随机森林模型对样本图片进行分类，通过特征重要性和相关性对特征进行降维，使用网格搜索确定随机森林模型的最优模型参数；基于根茬主干保持的直立特点，把检测到的每一个根茬上下顶点作为特征点，通过最小二乘法对所有特征点进行拟合即为作业导航线。试验结果表明，通过特征选择将30维特征降到16维，单幅图像处理时间从0.24 s降到0.16 s。在模型最优参数下，测试集准确率为92.5%。选取450幅10×20像素图片进行分类测试，准确率为91.8%，根茬检测准确率（敏感性）为90.7%。选择200幅ROI区域大小为100×200像素的田间作业图片进行导航线拟合试验，其包括不同天气、不同俯角以及非正常驾驶图片，成功拟合出171幅图像。该根茬检测方法具有较高的稳定性和准确性，可为不同作物根茬检测以及路径提取提供参考。

关键词: 机器视觉, 残膜回收, 导航路径, 特征选择, 随机森林, 最小二乘法

Abstract:

In view of the path planning problem in the visual navigation of the residual film recovery operation after the autumn, a method of path extraction for residual film recovery operation is proposed. The color space and texture features of the three types of images of stubble, residual film and inter-row noises were extracted to constitute a training set and test set with ratio of 8:2. With the goal of detected root, a Random Forest (RF) model is built to classify the sample images. The feature dimensions are reduced by the feature importance and correlation, the optimal model parameters of the RF model are determined by grid search. Based on the upright characteristics of the stubble, the upper and lower vertices of each stubble are detected as the feature points. All the feature points are fitted by the least square method as the navigation line. The experimental results show that the 30 feature dimensions are reduced to 16 feature dimensions by feature selection, and the processing time of a single image is reduced from 0.24 seconds to 0.16 seconds. Under the optimal parameters of the model, the accuracy of the test set is 92.5%. The classification test was selected for 450 images of 10× 20 pixels, the accuracy rate is 91.8%, the accuracy rate (sensitivity rate) of stubble detection is 90.7%. 200 field operation images with ROI area size of 100×200 pixels were selected for the navigation line fitting experiment, which included different weather, different depression angles and abnormal driving images, 171 images were successfully fitted. The stubble detection method has high stability and accuracy, and can provide reference for root crop detection and path extraction of different crops.

Key words: machine vision, residual film recycling, navigation path, feature selection, random forest, least square method

中图分类号:

李景彬,杨禹锟,温宝琴,坎杂,孙雯,杨朔. 基于根茬检测的秋后残膜回收导航路径提取方法[J]. 吉林大学学报(工学版), 2021, 51(4): 1528-1539.

Jing-bin LI,Yu-kun YANG,Bao-qin WEN,Za KAN,Wen SUN,Shuo YANG. Method of extraction of navigation path of post-autumn residual film recovery based on stubble detection[J]. Journal of Jilin University(Engineering and Technology Edition), 2021, 51(4): 1528-1539.

图/表 20

图1

图2

图3

图4

表1

颜色特征计算公式"

特征	计算公式
R分量平均值	$R m = ∑ j = 1 20 ∑ i = 1 10 R (i, j) 10 × 20$
G分量平均值	$G m = ∑ j = 1 20 ∑ i = 1 10 G (i, j) 10 × 20$
B分量平均值	$B m = ∑ j = 1 20 ∑ i = 1 10 B (i, j) 10 × 20$
H分量平均值	$H m = ∑ j = 1 20 ∑ i = 1 10 H (i, j) 10 × 20$
S分量平均值	$S m = ∑ j = 1 20 ∑ i = 1 10 S (i, j) 10 × 20$
V分量平均值	$V m = ∑ j = 1 20 ∑ i = 1 10 V (i, j) 10 × 20$

表1

图5

表2

灰度共生矩阵特征计算公式"

特征	计算公式
二阶矩（asm）	$a s m = ∑ i = 1 k ∑ j = 1 k (G (i, j)) 2$
熵（ent）	$e n t = - ∑ i = 1 k ∑ j = 1 k G (i, j) l o g G (i, j)$
相关性（cor）	$c o r = ∑ i = 1 k ∑ j = 1 k (i j) G (i, j) - u i u j s i s j$
差异性（dis）	$d i s = ∑ i = 1 k ∑ j = 1 k G (i, j) \| i - j \|$
对比度（con）	$c o n = ∑ n = 0 k - 1 n 2 ∑ i = 1 k ∑ j = 1 k G (i, j), \| i - j \| = n$
逆差矩（hom）	$h o m = ∑ i = 1 k ∑ j = 1 k G (i, j) 1 + (i - j) 2$

表2

图6

图7

表3

图8

图9

表4

表5

图10

表6

分类效果评价"

指标	公式	数值
准确率/%	$A c = T P + T N T P + T N + F P + F N$	91.8
敏感性/%	$S e = T P T P + F N$	90.7
特异性/%	$S p = T N T N + F P$	92.3

表6

表7

图11

图12

图13

参考文献 29

1	汤一平,王伟羊,朱威,等. 基于机器视觉的茶陇识别与采茶机导航方法[J]. 农业机械学报, 2016, 47(1): 45-50.
	Tang Yi-ping, Wang Wei-yang, Zhu Wei, et al. Tea ridge identification and navigation method for tea-plucking machine based on machine vision[J]. Transactions of the Chinese Society for Agricultural Machinery, 2016, 47(1): 45-50.
2	关卓怀,陈科尹,丁幼春,等.水稻收获作业视觉导航路径提取方法[J].农业机械学报,2020, 51(1):19-28.
	Guan Zhuo-huai, Chen Ke-yin, Ding You-chun,et al. Visual navigation path extraction method in rice harvesting[J]. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(1):19-28.
3	李云伍,徐俊杰,王铭枫,等. 丘陵山区田间道路自主行驶转运车及其视觉导航系统研制[J]. 农业工程学报, 2019, 35(1): 52-61.
	Li Yun-wu, Xu Jun-jie,Wang Ming-feng,et al. Development of autonomous driving transfer trolley on field roads and its visual navigation system for hilly areas[J]. Transactions of the Chinese Society of Agricultural Engineering,2019, 35(1): 52-61.
4	廖娟,汪鹞,尹俊楠,等. 基于分区域特征点聚类的秧苗行中心线提取[J]. 农业机械学报, 2019, 50(11): 34-41.
	Liao Juan, Wang Yao, Yin Jun-nan, et al. Detection of seedling row centerlines based on sub-regional feature points clustering[J]. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(11): 34-41.
5	赵岩,陈学庚,温浩军,等. 农田残膜污染治理技术研究现状与展望[J]. 农业机械学报, 2017, 48(6):1-14.
	Zhao Yan,Chen Xue-geng,Wen Hao-jun,et al. Research status and prospect of control technology for residual plastic film pollution in farmland[J]. Transactions of the Chinese society for Agricultural Machinery, 2017, 48(6):1-14.
6	姬长英,周俊. 农业机械导航技术发展分析[J]. 农业机械学报, 2014, 45(9): 44-54.
	Ji Chang-ying, Zhou Jun. Current situation of navigation technologies for agricultural machinery[J]. Transactions of the Chinese Society for Agricultural Machinery, 2014, 45(9):44-54.
7	胡静涛,高雷,白晓平,等. 农业机械自动导航技术研究进展[J]. 农业工程学报, 2015, 31(10): 1-10.
	Hu Jing-tao, Gao Lei, Bai Xiao-ping, et al. Review of research on automatic guidance of agricultural vehicles[J]. Transactions of the Chinese Society of Agricultural Engineering, 2015, 31(10): 1-10.
8	李景彬,陈兵旗,刘阳. 棉花铺膜播种机导航路线图像检测方法[J]. 农业机械学报, 2014, 45(1): 40-45.
	Li Jing-bin, Chen Bing-qi, Liu Yang. Image detection method of navigation route of cotton plastic film mulch planter[J]. Transactions of the Chinese Society for Agricultural Machinery, 2014, 45(1): 40-45.
9	彭顺正,坎杂,李景彬. 矮化密植枣园收获作业视觉导航路径提取[J]. 农业工程学报, 2017, 33(9): 45-52.
	Peng Shun-zheng, Kan Za, Li Jing-bin. Extraction of visual navigation directrix for harvesting operation in short-stalked and close-planting jujube orchard[J]. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33(9):45-52.
10	刁智华,赵明珍,宋寅卯,等.基于机器视觉的玉米精准施药系统作物行识别算法及系统实现[J]. 农业工程学报,2015,31(7):47-52.
	Diao Zhi-hua, Zhao Ming-zhen, Song Yin-mao, et al. Crop line recognition algorithm and realization in precision pesticide system based on machine vision[J]. Transactions of the Chinese Society of Agricultural Engineering, 2015, 31(7):47-52.
11	宋宇,刘永博,刘路,等. 基于机器视觉的玉米根茎导航基准线提取方法[J]. 农业机械学报, 2017, 48(2): 38-44.
	Song Yu, Liu Yong-bo, Liu Lu, et al. Extraction method of navigation baseline of corn roots based on machine vision[J]. Transactions of the Chinese Society for Agricultural Machinery, 2017, 48(2):38-44.
12	孟庆宽,张漫,仇瑞承,等. 基于改进遗传算法的农机具视觉导航线检测[J]. 农业机械学报, 2014, 45(10): 39-46.
	Meng Qing-kuan, Zhang Man, Qiu Rui-cheng, et al. Navigation line detection for farm machinery based on improved genetic algorithm[J]. Transactions of the Chinese Society for Agricultural Machinery, 2014, 45(10): 39-46.
13	王春雷,卢彩云,陈婉芝,等. 基于遗传算法和阈值滤噪的玉米根茬行图像分割[J]. 农业工程学报, 2019, 35(16): 198-205.
	Wang Chun-lei, Lu Cai-yun, Chen Wan-zhi, et al.Image segmentation of maize stubble row based on genetic algorithm and threshold filtering noise[J]. Transactions of the Chinese Society of Agricultural Engineering,2019, 35(16): 198-205.
14	杨洋,张亚兰,苗伟,等. 基于卷积神经网络的玉米根茎精确识别与定位研究[J]. 农业机械学报, 2018, 49(10): 46-53.
	Yang Yang, Zhang Ya-lan, Miao Wei, et al. Accurate identification and location of corn rhizome based on faster R-CNN[J]. Transactions of the Chinese Society for Agricultural Machinery, 2018, 49(10): 46-53.
15	Fu F, Duan J, Zou X, et al. Banana detection based on color and texture features in the natural environment[J]. Computers and Electronics in Agriculture, 2019,167:105057.
16	Abdalla A, Cen H, Wan L,et al. Fine-tuning convolutional neural network with transfer learning for semantic segmentation of ground-level oilseed rape images in a field with high weed pressure[J]. Computers and Electronics in Agriculture,2019,167:105091.
17	Bakhshipour A, Jafari A. Evaluation of support vector machine and artificial neural networks in weed detection using shape features[J]. Computers and Electronics in Agriculture, 2018,145:153-160.
18	蔡道清,李彦明,覃程锦,等. 水田田埂边界支持向量机检测方法[J]. 农业机械学报, 2019, 50(6): 22-27, 109.
	Cai Dao-qing, Li Yan-ming, Qin Cheng-jin, et al. Detection method of boundary of paddy fields using support vector machine[J]. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(6): 22-27, 109.
19	黄小玉,李光林,马驰,等. 基于改进判别区域特征融合算法的近色背景绿色桃子识别[J].农业工程学报,2018,34(23):142-148.
	Huang Xiao-yu, Li Guang-lin, Ma Chi, et al. Green peach recognition based on improved discriminative regional feature integration algorithm in similar background[J]. Transactions of the Chinese Society of Agricultural Engineering, 2018,34(23):142-148.
20	王煜,高向东,陈子琴,等. 激光-MAG复合焊接过程金属蒸气和背部熔池图像分析[J]. 机械工程学报,2019,55(19):167-173.
	Wang Yu, Gao Xiang-dong, Chen Zi-qin, et al. Analysis of the images of metal vapor and bottom-molten pool in laser-MAG hybrid welding process[J]. Journal of Mechanical Engineering, 2019, 55(19):167-173.
21	Haralick R M. Statistical and structural approaches to texture[J]. Proceedings of the IEEE, 1979,67(5): 786-804.
22	Breiman L, Cutler R A. Random forests machine learning[J]. Journal of Clinical Microbiology, 2001, 2:199-228.
23	Criminisi A, Shotton J, Robertson D, et al. Regression forests for efficient anatomy detection and localization in CT studies[J]. Journal of Bacteriology, 1957,74(5):559-564.
24	Stumpf André, Kerle Norman. Object-oriented mapping of landslides using random forests[J]. Remote Sensing of Environment, 2011, 115(10):2564-2577.
25	石礼娟,卢军.基于随机森林的玉米发育程度自动测量方法[J].农业机械学报,2017,48(1):169-174.
	Shi Li-juan, Lu Jun. Automatic measurement method for maize ear development degree based on random forest[J]. Transactions of the Chinese Society for Agricultural Machinery, 2017,48(1):169-174.
26	Verikas A, Gelzinis A, Bacauskiene M. Mining data with random forests: a survey and results of new tests[J]. Pattern Recognition, 2011, 44(2):330-349.
27	Raschka S. Python Machine Learning[M]. Birmingham: Packt Publishing, 2014.
28	Lin P,Li X,Li D,et al. Rapidly and exactly determining postharvest dry soybean seed quality based on machine vision technology[J]. Scientific Reports,2019,9(1):17143.
29	刘一鸣,侯智超,李晓琴,等.基于卷积神经网络的肺结节检测方法[J].生物医学工程学杂志,2019,36(6):969-977, 985.
	Liu Yi-ming, Hou Zhi-chao, Li Xiao-qin, et al. Pulmonary nodule detection method based on convolutional neural network[J]. Journal of Biomedical Engineering,2019, 36(6):969-977, 985.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

分类特征	准确性/%	增长幅度/%
GLCM（0）	84.0
GLCM（0）+ GLCM（90）	90.1	6.1
GLCM（0）+ GLCM（90）+ GLCM（45）	91.5	1.4
GLCM（0）+ GLCM（90）+ GLCM（45）+GLCM（135）	92.5	1.0
GLCM（0）+ GLCM（90）+ GLCM（45）+GLCM（135）+HSV	93.1	0.6
GLCM（0）+ GLCM（90）+ GLCM（45）+GLCM（135）+HSV+RGB	93.3	0.2

随机森林参数	数值	随机森林参数	数值
n_estimators	50	min_samples_split	2
max_depth	11	max_features	4
min_samples_leaf	1

[1]	顾天奇,胡晨捷,涂毅,林述温. 基于移动最小二乘法的稳健重构方法[J]. 吉林大学学报(工学版), 2021, 51(2): 685-691.
[2]	高万夫,张平,胡亮. 基于已选特征动态变化的非线性特征选择方法[J]. 吉林大学学报(工学版), 2019, 49(4): 1293-1300.
[3]	刘恩泽,吴文福. 基于机器视觉的农作物表面多特征决策融合病变判断算法[J]. 吉林大学学报(工学版), 2018, 48(6): 1873-1878.
[4]	林金花, 王延杰, 王璐, 姚禹. 全局相机姿态优化下的快速表面重建[J]. 吉林大学学报(工学版), 2018, 48(3): 909-918.
[5]	田彦涛, 张宇, 王晓玉, 陈华. 基于平方根无迹卡尔曼滤波算法的电动汽车质心侧偏角估计[J]. 吉林大学学报(工学版), 2018, 48(3): 845-852.
[6]	刘杰, 张平, 高万夫. 基于条件相关的特征选择方法[J]. 吉林大学学报(工学版), 2018, 48(3): 874-881.
[7]	袁哲明, 张弘杨, 陈渊. 基于特征选择和支持向量机的HIV-1型蛋白酶剪切位点预测[J]. 吉林大学学报(工学版), 2017, 47(2): 639-646.
[8]	张飞, 单忠德, 任永新, 聂军刚, 刘丰. 缸盖缺陷检测系统线阵相机现场标定方法[J]. 吉林大学学报(工学版), 2017, 47(2): 540-545.
[9]	周晓东, 张雅超, 谭庆昌, 张维君. 基于结构光视觉技术的圆柱度测量新方法[J]. 吉林大学学报(工学版), 2017, 47(2): 524-529.
[10]	赵东, 臧雪柏, 赵宏伟. 基于果蝇优化的随机森林预测方法[J]. 吉林大学学报(工学版), 2017, 47(2): 609-614.
[11]	祁若龙, 张伟, 王铁军, 李正. 仿人头颈部机器人跟踪运动控制[J]. 吉林大学学报(工学版), 2016, 46(5): 1595-1601.
[12]	秦余钢, 马勇, 张亮, 李腾飞. 基于改进最小二乘算法的船舶操纵性参数辨识[J]. 吉林大学学报(工学版), 2016, 46(3): 897-903.
[13]	姚向明, 赵鹏, 禹丹丹. 基于平均策略的城市轨道交通动态O-D矩阵估计[J]. 吉林大学学报(工学版), 2016, 46(1): 92-99.
[14]	张波,王文军,魏民国,成波. 基于机器视觉的驾驶人使用手持电话行为检测[J]. 吉林大学学报(工学版), 2015, 45(5): 1688-1695.
[15]	赵健, 张进, 朱冰. 基于路面识别的汽车驱动力模糊滑模控制[J]. 吉林大学学报(工学版), 2015, 45(4): 1036-1042.