基于改进Faster R-CNN模型的丁岙杨梅成熟度检测方法

doi:10.14088/j.cnki.issn0439-8114.2025.04.002

摘要/Abstract

摘要： 为了在复杂的自然生长环境中快速、精准地实现丁岙杨梅（Myrica rubra）不同成熟度检测,提出基于改进Faster R-CNN模型（ConvNeXt-T+SE+FPN）的丁岙杨梅成熟度检测方法。采用ConvNeXt-T作为主干特征提取网络,提升复杂场景下的检测能力;引入了SE注意力机制和特征金字塔网络（FPN）,增强模型对丁岙杨梅不同成熟度特征敏感性以及小目标果实的检测能力。相较于ResNet50,ConvNeXt-T+SE、ConvNeXt-T+FPN、ConvNeXt-T+SE+FPN能够使模型的平均精度均值（mAP）分别提升14.75%、19.85%、21.86%,其中ConvNeXt-T+SE+FPN的mAP提升幅度最大,能够有效提高丁岙杨梅不同成熟度的检测性能。通过对丁岙杨梅图像数据集进行训练和测试,改进Faster R-CNN模型在不同成熟度果实的检测中表现出较高的准确性,对未成熟、半成熟、近成熟和全成熟果实识别的平均精度（AP）分别为96.90%、94.63%、95.91%、97.58%,mAP为96.26%;相比Faster R-CNN模型,改进Faster R-CNN模型的mAP提升了21.86%。改进Faster R-CNN模型能够有效提升丁岙杨梅成熟度的检测精度,给杨梅的智能化采摘提供有力支持。

关键词: 丁岙杨梅（Myrica rubra）, 改进Faster R-CNN模型, 成熟度, 检测

Abstract: To rapidly and accurately detect the maturity levels of Ding’ao bayberry (Myrica rubra) in complex natural growth environments, an improved Faster R-CNN model (ConvNeXt-T+SE+FPN)-based maturity detection method was proposed. ConvNeXt-T was adopted as the backbone feature extraction network to enhance detection capabilities in complex scenarios. The SE attention mechanism and Feature Pyramid Network (FPN) were introduced to improve the model’s sensitivity to maturity-related features and detection of small-target fruits. Compared to ResNet50, ConvNeXt-T+SE, ConvNeXt-T+FPN, and ConvNeXt-T+SE+FPN increased the mean average precision (mAP) by 14.75%, 19.85%, and 21.86%, respectively. The ConvNeXt-T+SE+FPN configuration achieved the largest mAP improvement, effectively enhancing detection performance for different maturity levels of Ding’ao bayberry. Through training and testing on the Ding’ao bayberry image dataset, the improved Faster R-CNN model demonstrated high accuracy in detecting different maturity levels. The average precision (AP) for unripe, semi-ripe, near-ripe, and fully ripe fruit recognition was 96.90%, 94.63%, 95.91%, and 97.58%, respectively, with an mAP of 96.26%. Compared to the original Faster R-CNN model, the improved model achieved a 21.86% increase in mAP. The improved Faster R-CNN model effectively enhanced the detection accuracy of Ding’ao bayberry maturity, providing strong support for intelligent harvesting of bayberry fruits.

Key words: Ding’ao bayberry (Myrica rubra), improved Faster R-CNN model, maturity, detection

中图分类号:

TP391.4

刘玉耀, 彭琼尹. 基于改进Faster R-CNN模型的丁岙杨梅成熟度检测方法[J]. 湖北农业科学, 2025, 64(4): 7-13.

LIU Yu-yao, PENG Qiong-yin. Improved Faster R-CNN model-based maturity detection method for Ding’ao bayberry[J]. HUBEI AGRICULTURAL SCIENCES, 2025, 64(4): 7-13.

参考文献

[1] 黄建珍,黄胜华,赵友淦. 地理标志产品——丁岙杨梅标准化生产技术[J]. 中国南方果树, 2009, 38(6): 49-50.
[2] 尹建军,王新忠,毛罕平,等. RGB与HSI颜色空间下番茄图像分割的对比研究[J]. 农机化研究, 2006(11): 171-174.
[3] AI HIARY H, BANI AHMAD S,REYALAT M, et al.Fast and accurate detection and classification of plant diseases[J].International journal of computer applications,2011,17(1):31-38.
[4] 宋怀波,何东健,潘景朋.基于凸壳理论的遮挡苹果目标识别与定位方法[J].农业工程学报,2012,28(22):174-180.
[5] JI W,ZHA0 D A, CHENG F Y, et al. Automatic recognition vision system guided for apple harvesting robot[J].Computers & electrical engineering,2012,38(5):1186-1195.
[6] 徐黎明,吕继东.SUSAN算子和Hough变换在杨梅果实识别中的应用[J].中国农机化学报, 2015, 36(6): 216-220.
[7] 雷欢,吴亮生,焦泽昱,等.果园环境中成熟杨梅自动检测方法研究[J].自动化与信息工程, 2021, 42(3):9-14,26.
[8] HUANG K, LEI H, JIAO Z Y, et al.Bayberry maturity estimation algorithm based on multi-feature fusion[A].2021 IEEE international conference on artificial intelligence and computer applications (ICAICA)[C]. Piscataway, NJ: IEEE, 2021.514-518.
[9] KAMILARIS A,PRENAFETA-BOLDU F X. Deep learning in agriculture: A survey[J]. Computers and electronics in agriculture,2018,147:70-90.
[10] ALSHEMALI B,KALITA J.Improving the reliability of deep neural networks in NLP: A review[J]. Knowledge-Based systems, 2020,191:105210.
[11] 朱旭,马淏,姬江涛,等.基于Faster R-CNN的蓝莓冠层果实检测识别分析[J].南方农业学报,2020,51(6):1493-1501.
[12] WIDIYANTO S, WARDANI D T, PRANATA S W.Image-Based tomato maturity classification and detection using Faster R-CNN method[A].2021 5th International symposium on multidisciplinary studies and innovative technologies (ISMSIT)[C]. Piscataway, NJ: IEEE, 2021.130-134.
[13] WANG Z, LING Y, WANG X, et al.An improved Faster R-CNN model for multi-object tomato maturity detection in complex scenarios[J]. Ecological informatics, 2022, 72: 101886.
[14] 王菁,范晓飞,赵智慧,等.基于YOLO算法的不同品种枣自然环境下成熟度识别[J].中国农机化学报,2022,43(11):165-171.
[15] LI P, ZHENG J, LI P, et al.Tomato maturity detection and counting model based on MHSA-YOLOv8[J]. Sensors,2023, 23(15): 6701.
[16] 熊俊涛,韩咏林,王潇,等.基于YOLOv5-Lite的自然环境木瓜成熟度检测方法[J].农业机械学报,2023,54(6):243-252.
[17] 项新建,周焜,费正顺,等.基于改进YOLOX算法的杨梅成熟度检测方法[J].中国农机化学报, 2023,44(10):201-208.
[18] REN S,HE K,GIRSHICK R, et al.Faster-RCNN: Towards real-time object detection with region proposal networks[A].International conference on neural information processing systems[C]. MIT press, 2015.91-99.
[19] LIU Z,MAO H Z,WU C Y, et al.A convnet for the 2020s[A].Proceedings of the IEEE/CVF conference on computer vision and pattern recognition[C]. Piscataway, NJ: IEEE,2022.11976-11986.
[20] 宋洋,赵泉,张培安,等. 设施栽培对丁岙杨梅成熟期及果实品质的影响分析[J]. 浙江农业科学, 2023, 64(11): 2677-2681.
[21] 黄继鹏,史颖欢,高阳. 面向小目标的多尺度Faster-RCNN检测算法[J]. 计算机研究与发展, 2019, 56(2): 319-327.