杜鹃花各生长期识别与监测研究

doi:10.16180/j.cnki.issn1007-7820.2021.01.004

电子科技 ›› 2021, Vol. 34 ›› Issue (1): 17-22.doi: 10.16180/j.cnki.issn1007-7820.2021.01.004

杜鹃花各生长期识别与监测研究

裴晓芳^1,²,胡敏³

1.南京信息工程大学滨江学院,江苏无锡214105
2.南京信息工程大学江苏省大气环境与装备技术协同创新中心,江苏南京 210044
3.南京信息工程大学电子与信息工程学院,江苏南京 210044)

收稿日期:2019-10-23 出版日期:2021-01-15 发布日期:2021-01-22
作者简介:裴晓芳(1978-),女,副教授。研究方向:信号处理与应用。胡敏(1991-),女,硕士研究生。研究方向:基于机器学习的模式识别。
基金资助:
国家自然科学基金(61601229);南京信息工程大学滨江学院课题(2019BJYNG006)

Study on the Identification of Various Growth and Monitoring of Pest and Disease of Rhododendron

PEI Xiaofang^1,²,HU Min³

1. Binjiang College,Nanjing University of Information Science and Technology,Wuxi 214105,China
2. Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology,Nanjing University of Information Science and Technology,Nanjing 210044,China
3. School of Electronic and Information Engineering,Nanjing University of Information Science and Technology,Nanjing 210044,China

Received:2019-10-23 Online:2021-01-15 Published:2021-01-22
Supported by:
National Natural Science Foundation of China(61601229);Binjiang College,Nanjing University of Information Science & Technology Research Team(2019BJYNG006)

摘要/Abstract

摘要：

针对传统BoF算法缺乏空间信息的问题,文中提出一种改进式BoF算法,并将其应用于杜鹃花各生长期识别与病虫害监测问题。该算法在基于LAB的颜色特征中融入有序的空间信息,形成了新的空间颜色聚合特征来代替传统颜色直方图,有效解决了颜色特征变化尺度小的问题。该算法提取SURF特征代替原有的SIFT特征,通过一种多类特征学习算法融合颜色特征和SURF特征实现图像分类,并通过进一步分析叶片特征来快速识别杜鹃花植株的生长期与病害。经过仿真得知,基于LAB的颜色聚合向量的改进式BoF模型识别率达到了90.6%,较传统颜色直方图的图像分类方法图像检索速度增加3倍,更容易实现特征融合。

关键词: 改进式BoF算法, 空间颜色聚合特征, SURF, LAB, 多类特征学习, 叶片特征, 特征融合

Abstract:

In view of the problem that the traditional BoF algorithm lacks spatial information, an improved BoF algorithm is proposed and applied to the identification of various growth stages and the monitoring of pest and diseases of Rhododendron in this paper. The ordered spatial information is integrated into the LAB-based color features by this algorithm to form a new spatial color aggregation feature to replace the traditional color histogram, which effectively solved the problem of the small scale of color feature change. Besides, SURF features are extracted to replace the original SIFT features. The image classification is realized by a multi-class features, and the leaf features are further analyzed to quickly identify the growth period and disease of Rhododendron plants. Simulation results show that the recognition rate of the improved BoF model of the LAB-based color aggregation vector is 90.6%. Compared with the image classification method of the traditional histogram, the image retrieval speed is increased by 3 times, making it easier to implement the feature fusion.

Key words: improved BoF algorithm, spatial color aggregation feature, SURF, LAB, multi-class feature learning, leaf features, feature fusion

中图分类号:

TP391.4

裴晓芳, 胡敏. 杜鹃花各生长期识别与监测研究[J]. 电子科技, 2021, 34(1): 17-22.

PEI Xiaofang, HU Min. Study on the Identification of Various Growth and Monitoring of Pest and Disease of Rhododendron[J]. Electronic Science and Technology, 2021, 34(1): 17-22.

图/表 9

图1

图2

图3

图4

图5

图6

图7

图8

表1

参考文献 18

[1]	李康顺, 王福滨, 张丽霞, 等. 基于改进BOF算法的图像识别和分类[J]. 中南大学学报(自然科学版), 2016,47(5):1599-1605.
	Li Kangshun, Wang Fubin, Zhang Lixia, et al. Image recognition and classification based on improved BOF algorithm[J]. Journal of Central South University(Science and Technology), 2016,47(5):1599-1605.
[2]	阮嘉. 基于BoF的交通标志分类算法研究[D]. 安庆:安庆师范大学, 2018.
	Ruan Jia. Research on traffic sign classification algorithm based on BoF[D]. Anqing:Anqing Normal University, 2018.
[3]	张泽晨, 巨志勇. 基于BoF模型的多特征融合果蔬图像分类方法[J]. 电子科技, 2020,33(7):41-45,56.
	Zhang Zechen, Ju Zhiyong. Multi-feature fusion fruit and vegetable image classification method based on BoF model[J]. Electronic Science and Technology, 2020,33(7):41-45,56.
[4]	赵春晖, 王莹, Masahide Kaneko. 一种基于词袋模型的图像优化分类方法[J]. 电子与信息学报, 2012,34(9):2064-2070.
	Zhao Chunhui, Wang Ying, Masahide Kaneko. An optimized method for image classifcation based on bag of words model[J]. Journal of Electronics and Information Technology, 2012,34(9):2064-2070.
[5]	汪宇玲, 黎明, 李军华, 等. 基于BoF模型的多特征融合纹理图像分类[J]. 北京航空航天大学学报, 2018,44(9):1869-1877.
	Wang Yuling, Li Ming, Li Junhua, et al. Texture image classification based on BoF model with multi-feature fusion[J]. Journal of Beijing University of Aeronautics and Astronautics, 2018,44(9):1869-1877.
[6]	权稳稳, 林明星. CNN特征与BoF相融合的水下目标识别算法[J]. 山东大学学报(工学版), 2019,49(1):107-113.
	Quan Wenwen, Lin Mingxing. Algorithm of underwater target recognition based on CNN features with BoF[J]. Journal of Shandong University (Engineering Science), 2019,49(1):107-113.
[7]	戴天虹, 胡学晶, 黄建平. 基于词袋模型的单板分类的研究[J]. 电脑知识与技术, 2017,13(9):152-154,169.
	Dai Tianhong, Hu Xuejing, Huang Jianping. Based on the bag-of-word model veneer image classification[J]. Computer Knowledge and Technology, 2017,13(9):152-154,169.
[8]	杨晨欣, 邓燕妮. 基于改进BoF算法的导线绝缘破损识别[J]. 科学技术创新, 2019(23):77-79.
	Yang Chenxin, Deng Yanni. Identification of conductor insulation damage based on improved BoF algorithm[J]. Scientific and Technological Innovation, 2019(23):77-79.
[9]	Yang Cao, Wang Changhu, Li Zhiwei. Spatial-bag-of-features[C]. San Francisco:IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2010.
[10]	Jegou H, Douze M, Schmid C, et al. Packing bag-of-features[C]. Shanghai:IEEE the Twelfth International Conference, 2009.
[11]	李康, 尚鹏, 耿国华. 基于深度图像的人脸模型特征点自动标定[J]. 计算机科学, 2014,41(4):287-291.
	Li Kang, Shang Peng, Geng Guohua. Automatic location of feature points on Three-dimensional facial model based on depth image[J]. Computer Science, 2014,41(4):287-291.
[12]	祁曦. 基于改进SIFT和改进K-means的遥感图像配准算法[D]. 上海:华东师范大学, 2018.
	Qi Xi. Remote sensing image registration algorithm based on improved SIFT and improved K-means[D]. Shanghai:East China normal university, 2018.
[13]	王娜娜, 李晓旭, 曹洁. 提取目标区域词袋特征的图像分类方法[J]. 计算机工程与应用, 2018,54(20):208-212,218.
	Wang Nana, Li Xiaoxu, Cao Jie. Image classification method based on extraction bag of words features of objeact area[J]. Computer Engineering and Application, 2018,54(20):208-212,218.
[14]	裴勇. 基于数字图像的花卉种类识别技术研究[D]. 北京: 北京林业大学, 2011.
	Pei Yong. Research on flower species identification based on digital image[D]. Beijing:Beijing Forestry University, 2011.
[15]	Lowe D G. Distinctive image features from scale-invariant key point[J]. Internal Journal of Computer Vision, 2004,60(2):91-110.
[16]	Lazebnik S, Schmid C, Ponce J, et al. Beyond bags of features: spatial pyramid matching for recognizing natural scene categories[C]. New York:IEEE Computer Society Conference, 2006.
[17]	冯进丽, 杨红菊. 基于BoC-BoF特征的图像检索方法研究[J]. 计算机科学, 2015,42(4):297-301.
	Feng Jinli, Yang Hongju. Image retrieval method research based on BoC-BoF feature[J]. Computer Science, 2015,42(4):297-301.
[18]	林君宇, 李奕萱, 郑聪尉, 等. 应用卷积神经网络识别花卉及其病症[J]. 小型微型计算机系统, 2019,40(6):1330-1335.
	Lin Junyu, Li Yixuan, Zheng Congwei, et al. Classifying flowers and their diseases by convolution neural network[J]. Journal of Chinese Computer System, 2019,40(6):1330-1335.

检测方法	特征点数/个	运行时间/s	识别率/%
BoF+SIFT	4 096	-	80.60%
BoF+SURF^[17]	8 192	5.853	91.00%
CNN^[18]	-	-	90.45%
本文方法	4 352	4.445	90.60%

杜鹃花各生长期识别与监测研究

Study on the Identification of Various Growth and Monitoring of Pest and Disease of Rhododendron

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 18

相关文章 15

Metrics

本文评价

推荐阅读 10

[1]	刘建龙,郝正航. 基于背靠背变流器的风电系统仿真对比研究[J]. 电子科技, 2022, 35(2): 67-73.
[2]	刘晨旻,王亚刚. 基于连续空间的萤火虫算法改进[J]. 电子科技, 2022, 35(2): 40-45.
[3]	牛帅臣,王福忠,韩耀飞,何国锋,牛叶克. 电网不平衡条件下模块化多电平变换器控制策略[J]. 电子科技, 2022, 35(1): 80-86.
[4]	王阳,王亚刚. 基于阶跃响应和遗传算法优化高阶加时滞模型的辨识方法[J]. 电子科技, 2021, 34(9): 41-46.
[5]	李玉东,连海山,胡晓丹. 三相输入型阻塞式交交变频电源[J]. 电子科技, 2021, 34(9): 47-53.
[6]	赵崇,迟蒙蒙,储聪,张鹏. 导盲犬行走机构运动仿真及其视觉识别算法研究[J]. 电子科技, 2021, 34(9): 66-72.
[7]	闫振彬,郑柏超,周志勇. 基于新型鲁棒下垂控制方法的光伏发电系统并网研究[J]. 电子科技, 2021, 34(8): 79-86.
[8]	马立新,豆晨飞,宋晨灿,杨天笑. 基于特征融合卷积神经网络的绝缘子无损检测[J]. 电子科技, 2021, 34(7): 26-30.
[9]	仝兆景,秦紫霓,赵运星,芦彤,郑权. 基于贝叶斯网络的变压器故障诊断研究[J]. 电子科技, 2021, 34(3): 43-47.
[10]	孙慧霞,周上楠,周玲,窦永梅. 基于MATLAB GUI的数字信号处理仿真平台开发[J]. 电子科技, 2021, 34(2): 74-78.
[11]	何强,杨青慧,张怀武. 一种射频放大器上电时序及偏置点控制方法[J]. 电子科技, 2021, 34(2): 1-6.
[12]	蒋凯文,金海,许参. 一种高功率因数无电解电容PMSM的设计与仿真[J]. 电子科技, 2021, 34(12): 13-18.
[13]	陈文佑,章伟,史晓帆,宋芳. 一种改进ORB特征匹配的半稠密三维重建ORB-SLAM算法[J]. 电子科技, 2021, 34(12): 62-67.
[14]	王佳人,王亚刚. 非最小相位系统的复杂控制算法研究[J]. 电子科技, 2021, 34(1): 55-59.
[15]	胡晓丹,李玉东,连海山. 基于dSPACE的阻塞式交交变频系统实现[J]. 电子科技, 2020, 33(7): 51-56.