基于HSV与形状特征融合的花椒图像识别

doi:10.13733/j.jcam.issn.2095-5553.2021.10.25

摘要/Abstract

摘要： 光照条件变化会对花椒目标识别率产生影响,关系到机器视觉技术能否有效应用于生产现场的花椒采摘。通过对HSV特性图像识别技术的分析,提出HSV和形状特征融合的花椒识别算法。该算法采用同态滤波方法对光照进行补偿,解决因为光照不均匀而导致的花椒识别率低的问题,最后利用花椒圆度特征,排除树枝及树叶等的干扰,实现花椒的准确识别。利用同态滤波方法对光照进行补偿,对于光照不强或者发生遮挡的花椒图像有较大改善,通过试验得出其平均识别率达到94.0%,比单独采用HSV特性识别时,在顺光,背光和遮阴条件下,识别率分别提高4%,13%和21%,此外在遮阴条件下运行时间缩短14.6%。为遮阴条件下提高花椒识别率提供一种方法。

关键词: 花椒, 同态滤波, 形状特征, 图像识别

Abstract: Changes in lighting conditions will have an impact on the target recognition rate of Zanthoxylum bungeanum maxim. It is related to whether machine vision technology can be effectively applied to Zanthoxylum bungeanum picking at the production site. Based on the analysis of HSV characteristic image recognition technology, a prickly ash recognition algorithm combining HSV and shape features is proposed. The calculation uses a homomorphic filtering method to compensate for the illumination, which solves the problem of low recognition rate of Zanthoxylum bungeanum due to uneven illumination. The roundness feature of Zanthoxylum bungeanum is used to eliminate the interference of branches and leaves and realize accurate recognition. The average recognition rate reaches 94.0%, which is higher than when the HSV characteristic recognition is used alone. Under backlight and shading conditions, the recognition rate is increased by 4%, 13%, and 21%, respectively, and the running time under shading conditions is shortened by 14.6%. This paper provides a method to improve the recognition rate of Zanthoxylum bungeanum under shade conditions.

Key words: pepper, homomorphic filtering, shape feature, image recognition

中图分类号:

汪杰, 陈曼龙, 李奎, 丁敏, 王琨, . 基于HSV与形状特征融合的花椒图像识别[J]. 中国农机化学报, 2021, 42(10): 180-185.

Wang Jie, Chen Manlong, Li Kui, Ding Min, Wang Kun.. Prickly ash image recognition based on HSV and shape feature fusion[J]. Journal of Chinese Agricultural Mechanization, 2021, 42(10): 180-185.

参考文献

［１］
万芳新, 白明昌, 贺志洋, 等. 自然场景下花椒果实的识别［J］. 中国农机化学报, 2016, 37(10): 115-119.
Wan Fangxin, Bai Mingchang, He Zhiyang, et al. Identification of Chinese prickly ash under the natural scenes ［J］. Journal of Chinese Agricultural Mechanization, 2016, 37(10): 115-119.

［2］
Zhang J. Target extraction of fruit picking robot vision system ［J］. Journal of Physics Conference Series, 2019, 1423: 012061.

［3］
Tang S, Zhao D, Jia W, et al. Feature extraction and recognition based on machine vision application in lotus picking robot ［J］. International Conference on Computer & Computing Technologies in Agriculture, 2015.

［4］
吕继东, 吕小俊, 王艺洁, 等. 近色背景果实识别的多源图像融合［J］. 中国农机化学报, 2020, 41(3): 141-146.

Lü Jidong, Lü Xiaojun, Wang Yijie, et al. Multisource image fusion for fruit recognitionin similarcolor background ［J］. Journal of Chinese Agricultural Mechanization, 2020, 41(3): 141-146.

［5］
赵利平, 吴德刚. 融合GA的三点定位夜间苹果目标的识别算法研究［J］. 中国农机化学报, 2020, 41(5): 134-138. 
Zhao Liping, Wu Degang. Research on threepoint location algorithm based on GA for night apple targets ［J］. Journal of Chinese Agricultural Mechanization, 2020, 41(5): 134-138.

［6］
熊俊涛, 林睿, 刘振, 等. 夜间自然环境下荔枝采摘机器人识别技术［J］. 农业机械学报, 2017, 48(11): 28-34.
Xiong Juntao, Lin Rui, Liu Zhen, et al. Visual technology of picking robot to detect litchi at nighttime ［J］. Transactions of the Chinese Society of Agricultural Machinery, 2017, 48(11): 28-34.

［7］
刘小刚, 范诚, 李加念, 等. 基于卷积神经网络的草莓识别方法［J］. 农业机械学报, 2020, 51(2): 237-244.
Liu Xiaogang, Fan Cheng, Li Jianian, et al. Strawberry recognition method based on convolutional neural network ［J］. Transactions of the Chinese Society of Agricultural Machinery, 2020, 51(2): 237-244.

［8］
〖JP3〗闫建伟, 赵源, 张乐伟, 等. 基于残差网络的自然环境中刺梨果实的识别［J］. 中国农机化学报, 2020, 41(10): 191-196. 
Yan Jianwei， Zhao Yuan， Zhang Lewei, et al. Recognition of Rosa roxbunghil fruit in natural environment based on residual network［J］.Journal of Chinese Agricultural Mechanization, 2020, 41(10): 191-196.

［9］
赵贤国. 基于几何特征的花粉图像识别算法研究［D］. 南京: 南京信息工程大学, 2018.
Zhao Xianguo. Research on pollen image recognition algorithm based on geometric features ［D］. Nanjing: Nanjing University of Information Technology, 2018.

［10］
〖JP3〗邱津怡, 罗俊, 李秀, 等. 基于卷积神经网络的多尺度葡萄图像识别方法［J］. 计算机应用, 2019, 39(10): 2930-2936. 
〖JP3〗Qiu Jinyi, Luo Jun, Li Xiu, et al. Multiscale grape image recognition method based on convolutional neural network ［J］. Computer Applications, 2019, 39(10): 2930-2936.

［11］
苗中华, 沈一筹, 王小华, 等. 自然环境下重叠果实图像识别算法与试验［J］. 农业机械学报, 2016, 47(6): 21-26. 
Miao Zhonghua, Shen Yichou, Wang Xiaohua, et al. Recognition algorithm and experiment of overlapping fruit images in natural environment ［J］. Transactions of the Chinese Society of Agricultural Machinery, 2016, 47(6): 21-26.

［12］
Wang B, Brown D, Gao Y, et al. MARCH: Multiscalearchheight description for mobile retrieval of leaf images ［J］. Informationences, 2015, 302: 132-148.

［13］
Zhuang J J, Luo S M, Hou C J, et al. Detection of orchard citrus fruits using a monocular machine visionbased method for automatic fruit picking applications ［J］. Computers and Electronics in Agriculture, 2018, 152: 64-73.

［14］
刘佳敏, 何宁. 基于改进同态滤波的低对比度图像增强［J］. 计算机应用与软件, 2020, 37(3): 220-224.
Liu Jiamin, He Ning. Lowcontrast image enhancement based on improved homomorphic filtering ［J］. Computer Applications and Software, 2020, 37(3): 220- 224.

［15］
张亚飞, 谢明鸿. 基于HSI和局部同态滤波的彩色图像增强算法［J］. 计算机应用与软件, 2013, 30(12): 303-307. 
〖JP3〗Zhang Yafei, Xie Minghong. Color image enhancement algorithm based on HSI and local homomorphic filtering ［J］. Computer Applications and Software, 2013, 30(12): 303-307.

［16］
焦竹青, 徐保国. 基于同态滤波的彩色图像光照补偿方法［J］. 光电子激光, 2010, 21(4): 602-605.
Jiao Zhuqing, Xu Baoguo. Color image illumination compensation based on homomorphic filtering ［J］. Journal of Optoelectronics•Laser, 2010, 21(4): 602-605.

［17］
齐锐丽, 陈曼龙, 杨宗浩, 等. 基于HSV模型与改进的OTSU算法花椒图像分割［J］. 中国农机化学报, 2019, 40(11): 155-160.
Qi Ruili, Chen Manlong, Yang Zonghao, et al. Image segmentation of Sichuan pepper based on HSV model and improved OTSU algorithm ［J］. Journal of Chinese Agricultural Mechanization, 2019, 40(11): 155-160.

［18］
宋彦, 谢汉垒, 宁井铭, 等. 基于机器视觉形状特征参数的祁门红茶等级识别［J］. 农业工程学报, 2018, 34(23): 279-286.
Song Yan, Xie Hanlei, Ning Jingming,et al. Grading Keemun black tea based on shape feature parameters of machine vision ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(23): 279-286.

[1]	于雪莹, 高继勇, 王首程, 李庆盛, 王志强. 基于生成对抗网络和混合注意力机制残差网络的苹果病害识别[J]. 中国农机化学报, 2022, 43(6): 166-174.
[2]	雷雪梅, 张光强, 姚旗, 刘伟渭, 邱帅. 基于卷积神经网络的农机图像自动识别研究[J]. 中国农机化学报, 2022, 43(5): 140-147.
[3]	陈旭, 关卓怀, 李海同, 沐森林, 张敏, 吴崇友. 油菜联合收获机含杂率在线检测系统设计与试验[J]. 中国农机化学报, 2022, 43(10): 127-134.
[4]	戴雨舒, 仲晓春, 孙成明, 杨俊, 刘涛, 刘升平, . 基于图像处理和Deeplabv3+模型的小麦赤霉病识别[J]. 中国农机化学报, 2021, 42(9): 209-215.
[5]	项小东, 翟蔚, 黄言态, 刘薇, . 基于Xception-CEMs神经网络的植物病害识别[J]. 中国农机化学报, 2021, 42(8): 177-186.
[6]	梁裕巧;李洪兵;罗洋;康云川;宋小令;向伟;. 基于NB-IoT的花椒生长环境因子监测系统[J]. 中国农机化学报, 2021, 42(6): 166-173.
[7]	闫建伟;张乐伟;赵源;张富贵;. 改进RetinaNet的刺梨果实图像识别[J]. 中国农机化学报, 2021, 42(3): 78-83.
[8]	张亚军;. 基于改进支持向量机算法的农业害虫图像识别研究[J]. 中国农机化学报, 2021, 42(2): 146-152.
[9]	曹跃腾, 朱学岩, 赵燕东, 陈锋军, . 基于改进ResNet的植物叶片病虫害识别[J]. 中国农机化学报, 2021, 42(12): 175-181.
[10]	王林柏, 张博, 姚竟发, 杨志辉, 张君, 范晓飞. 基于卷积神经网络马铃薯叶片病害识别和病斑检测^*[J]. 中国农机化学报, 2021, 42(11): 122-129.
[11]	周宏威, 沈恒宇, 袁新佩, 李晓冬. 基于迁移学习的苹果树叶片病虫害识别方法研究^*[J]. 中国农机化学报, 2021, 42(11): 151-158.
[12]	赵敬坤, 陈松柏, 李忠意, 王洋, 张涛, 王帅, . 模糊综合评价法判断重庆花椒种植区土壤肥力水平[J]. 中国农机化学报, 2021, 42(10): 206-212.
[13]	李浩;徐航煌;郑恒宇;陈学永;. 基于无人机遥感图像的松材线虫病监测技术研究[J]. 中国农机化学报, 2020, 41(9): 170-175.
[14]	李娟;朱学岩;葛凤丽;白皓然;. 基于计算机视觉的水下海参识别方法研究[J]. 中国农机化学报, 2020, 41(7): 171-177.
[15]	刘安成;尉倩;崔新爱;王庆;韩桂军;高森;. 花椒采收现状及研究进展[J]. 中国农机化学报, 2019, 40(3): 84-87.