[1]吕继东,王艺洁,夏正旺,等.基于改进的MaskR-CNN自然场景下苹果识别研究[J].常州大学学报(自然科学版),2022,34(01):68-77.[doi:10.3969/j.issn.2095-0411.2022.01.008]
　LYU Jidong,WANG Yijie,XIA Zhengwang,et al.Research on Natural Scene Apple Recongnition Based on Improved Mask R-CNN[J].Journal of Changzhou University(Natural Science Edition),2022,34(01):68-77.[doi:10.3969/j.issn.2095-0411.2022.01.008]
点击复制

基于改进的MaskR-CNN自然场景下苹果识别研究()
分享到：

常州大学学报(自然科学版)[ISSN:2095-0411/CN:32-1822/N]

卷:
第34卷
期数:
2022年01期
页码:
68-77
栏目:
计算机与信息工程
出版日期:
2022-01-28

文章信息/Info

Title:
Research on Natural Scene Apple Recongnition Based on Improved Mask R-CNN
文章编号:
2095-0411(2022)01-0068-10
作者:
吕继东; 王艺洁; 夏正旺; 马正华
(常州大学微电子与控制工程学院,江苏常州213164)
Author(s):
LYU Jidong;  WANG Yijie;  XIA Zhengwang;  MA Zhenghua
(School of Microelectronics and Control Engineering, Changzhou University, Changzhou 213164, China)
关键词:
神经网络;  目标检测;  苹果采摘;  Mask R-CNN;  实例分割
Keywords:
neural network;  target detection;  apple picking;  Mask R-CNN;  instance segmentation
分类号:
TP 391.4
DOI:
10.3969/j.issn.2095-0411.2022.01.008
文献标志码:
A
摘要:
在复杂自然场景下,苹果目标因具有成簇生长、重叠果实和光线变化大等特点,应用深度学习方法相比传统方法来实现果实的识别优势明显。提出基于Mask R-CNN网络检测分割架构,采用膨胀卷积的优化策略,通过候选框与像素分割相结合的思路,同时对输入苹果图像进行目标果实的识别。实验结果表明,基于Mask R-CNN框架改进的网络模型的识别性能较原始Mask R-CNN网络有较大提升。针对不同光照角度、不同颜色和不同大小的苹果,改进Mask R-CNN网络的F1值分别提升了2.17%,1.87%和4.93%。
Abstract:
In the field environment, fruit images are easily affected by many external environmental factors such as light changes, fruit size difference, complicated background noise, the application of deep learning method has obvious advantages over traditional methods to realize fruit recognition. To address these problems, this paper proposes a detection and recognition framework based on Mask R-CNN network, which uses the dilated convolution optimization strategy and combines bounding box and pixel segmentation, simultaneously recognizes fruits from the input apple image. The experimental results show that the recognition performance of the improved network model based on Mask R-CNN framework is better than that of the original Mask R-CNN network. The F1 score of improved Mask R-CNN network was improved by 2.17%, 1.87% and 4.93% for apples of different illumination angles, colors and sizes, respectively.

参考文献/References:

[1]熊俊涛, 叶敏, 邹湘军, 等. 多类型水果采摘机器人系统设计与性能分析[J]. 农业机械学报, 2013, 44(S1): 230-235.
[2]王丹丹, 宋怀波, 何东健. 苹果采摘机器人视觉系统研究进展[J]. 农业工程学报, 2017, 33(10): 59-69.
[3]WEI X, JIA K, LAN J, et al. Automatic method of fruit object extraction under complex agricultural background for vision system of fruit picking robot[J]. Optik-International Journal for Light and Electron Optics, 2014, 125(19): 5684-5689.
[4]刘继展. 温室采摘机器人技术研究进展分析[J]. 农业机械学报, 2017, 48(12): 1-18.
[5]项荣, 应义斌, 蒋焕煜. 田间环境下果蔬采摘快速识别与定位方法研究进展[J]. 农业机械学报, 2013, 44(11): 208-223.
[6]傅隆生, 冯亚利, ELKAMIL T, 等. 基于卷积神经网络的田间多簇猕猴桃图像识别方法[J]. 农业工程学报, 2018, 34(2): 205-211.
[7]薛月菊, 黄宁, 涂淑琴, 等. 未成熟芒果的改进YOLOv2识别方法[J]. 农业工程学报, 2018, 34(7): 173-179.
[8]熊俊涛, 刘振, 林睿, 等. 自然环境下树上绿色芒果的无人机视觉检测技术[J]. 农业机械学报, 2018, 49(11): 23-29.
[9]赵德安, 吴任迪, 刘晓洋, 等. 基于YOLO深度卷积神经网络的复杂背景下机器人采摘苹果定位[J]. 农业工程学报, 2019, 35(3): 164-173.
[10]HE K M, GKIOXARI G, DOLLAR P, et al. Mask R-CNN[C]// 2017 IEEE International Conference on Computer Vision(ICCV). Venice: IEEE, 2017: 2980-2988.
[11]HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition [C] // 2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Las Vegas: IEEE, 2016: 770-778.
[12]LECUN Y L, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.
[13]JI S, XU W, YANG M, et al. 3D convolutional neural networks for human action recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(1): 221-231.
[14]YU F, KOLTUN V. Multi-scale context aggregation by dilated convolutions[C]. Puerto Rico: ICLR, 2016.
[15]CHEN L C, PAPANDREOU G, SCHROFF F, et al. Rethinking atrous convolution for semantic image segmentation [EB/OL]. [2017-06-17]. https://arxiv.org/abs/1706.05587.

相似文献/References:

[1]王正洪,蒋建明,张小鸣.基于神经网络的瓦斯传感器非线性校正[J].常州大学学报(自然科学版),2005,(01):37.
　WANG Zheng -hong,JIANG Jian -ming,ZHANG Xiao -ming.Nonlinear Compensation of Methane Sensor Based on Neural Network[J].Journal of Changzhou University(Natural Science Edition),2005,(01):37.
[2]倪重文,江兴方,王叶荟,等.神经网络模型应用于少数投影图像重建及数值模拟[J].常州大学学报(自然科学版),2005,(02):1.
　NI Zhong -wen,JIANG Xing -fang,WANG Ye -hui,et al.Neural Network Model Used in Image Reconstruction from a Few Projections and Numeric Simulation[J].Journal of Changzhou University(Natural Science Edition),2005,(01):1.
[3]周序洋,查利权.基于前馈多层神经网络的振动模态参数计算[J].常州大学学报(自然科学版),2002,(04):50.
　ZHOU Xu -yang,ZHA Li-quan.Computing Vibrational Modal Parameters by Neural Network[J].Journal of Changzhou University(Natural Science Edition),2002,(01):50.
[4]朱家群,王东阳,顾玉宛,等.基于多尺度融合CNNs特征和Transformer特征的显著目标检测[J].常州大学学报(自然科学版),2023,35(06):35.[doi:10.3969/j.issn.2095-0411.2023.06.005]
　ZHU Jiaqun,WANG Dongyang,GU Yuwan,et al.Salient object detection based on multi-scale fusion of CNNs features and Transformer features[J].Journal of Changzhou University(Natural Science Edition),2023,35(01):35.[doi:10.3969/j.issn.2095-0411.2023.06.005]

备注/Memo

备注/Memo:
收稿日期: 2021-05-06。
基金项目: 江苏省自然科学青年基金资助项目(BK20140266); 江苏省高等学校自然科学研究面上资助项目(17KJB416002); 常州市科技计划资助项目(CJ20180021)。
作者简介: 吕继东(1980—), 男, 河南驻马店人, 博士, 副教授。通信联系人: 马正华(1962—), E-mail: lmcczu@163.com
更新日期/Last Update: 1900-01-01