Walnut Recognition Method for UAV Remote Sensing Images

被引:6
作者
Wu, Mingjie [1 ,2 ]
Yun, Lijun [1 ,2 ]
Xue, Chen [1 ,2 ]
Chen, Zaiqing [1 ,2 ]
Xia, Yuelong [1 ,2 ]
机构
[1] Yunnan Normal Univ, Sch Informat, Kunming 650500, Peoples R China
[2] Engn Res Ctr Comp Vis & Intelligent Control Techno, Dept Educ Yunnan Prov, Kunming 650500, Peoples R China
来源
AGRICULTURE-BASEL | 2024年 / 14卷 / 04期
关键词
low-altitude remote sensing; walnut; small object detection; YOLOv8s; IDENTIFICATION; ALGORITHM;
D O I
10.3390/agriculture14040646
中图分类号
S3 [农学(农艺学)];
学科分类号
0901 ;
摘要
During the process of walnut identification and counting using UAVs in hilly areas, the complex lighting conditions on the surface of walnuts somewhat affect the detection effectiveness of deep learning models. To address this issue, we proposed a lightweight walnut small object recognition method called w-YOLO. We reconstructed the feature extraction network and feature fusion network of the model to reduce the volume and complexity of the model. Additionally, to improve the recognition accuracy of walnut objects under complex lighting conditions, we adopted an attention mechanism detection layer and redesigned a set of detection heads more suitable for walnut small objects. A series of experiments showed that when identifying walnut objects in UAV remote sensing images, w-YOLO outperforms other mainstream object detection models, achieving a mean Average Precision (mAP0.5) of 97% and an F1-score of 92%, with parameters reduced by 52.3% compared to the YOLOv8s model. Effectively addressed the identification of walnut targets in Yunnan, China, under the influence of complex lighting conditions.
引用
收藏
页数:19
相关论文
共 49 条
[1]   Molecular phylogeny of Juglans (Juglandaceae):: a biogeographic perspective [J].
Aradhya, Mallikarjuna K. ;
Potter, Daniel ;
Gao, Fangyou ;
Simon, Charles J. .
TREE GENETICS & GENOMES, 2007, 3 (04) :363-378
[2]   A robust algorithm based on color features for grape cluster segmentation [J].
Behroozi-Khazaei, Nasser ;
Maleki, Mohammad Reza .
COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2017, 142 :41-49
[3]   Run, Don't Walk: Chasing Higher FLOPS for Faster Neural Networks [J].
Chen, Jierun ;
Kao, Shiu-Hong ;
He, Hao ;
Zhuo, Weipeng ;
Wen, Song ;
Lee, Chul-Ho ;
Chan, S. -H. Gary .
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2023, :12021-12031
[4]   Mobile-Former: Bridging MobileNet and Transformer [J].
Chen, Yinpeng ;
Dai, Xiyang ;
Chen, Dongdong ;
Liu, Mengchen ;
Dong, Xiaoyi ;
Yuan, Lu ;
Liu, Zicheng .
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022), 2022, :5260-5269
[5]   An object detection method for bayberry trees based on an improved YOLO algorithm [J].
Chen, Youliang ;
Xu, Hanli ;
Zhang, Xiangjun ;
Gao, Peng ;
Xu, Zhigang ;
Huang, Xiaobin .
INTERNATIONAL JOURNAL OF DIGITAL EARTH, 2023, 16 (01) :781-805
[6]   Pest identification via deep residual learning in complex background [J].
Cheng, Xi ;
Zhang, Youhua ;
Chen, Yiqiong ;
Wu, Yunzhi ;
Yue, Yi .
COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2017, 141 :351-356
[7]   Dynamic Head: Unifying Object Detection Heads with Attentions [J].
Dai, Xiyang ;
Chen, Yinpeng ;
Xiao, Bin ;
Chen, Dongdong ;
Liu, Mengchen ;
Yuan, Lu ;
Zhang, Lei .
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021, 2021, :7369-7378
[8]   CenterNet: Keypoint Triplets for Object Detection [J].
Duan, Kaiwen ;
Bai, Song ;
Xie, Lingxi ;
Qi, Honggang ;
Huang, Qingming ;
Tian, Qi .
2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), 2019, :6568-6577
[9]  
Ge Z, 2021, Arxiv, DOI [arXiv:2107.08430, DOI 10.48550/ARXIV.2107.08430]
[10]   Fast R-CNN [J].
Girshick, Ross .
2015 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2015, :1440-1448