Segmentation of Buckwheat by UAV Based on Improved Lightweight DeepLabV3+ at Seedling Stage

WU Jinlong, WU Hongqi, LI Hao, LEI Xingpeng, SONG Haiyan

Abstract

In view of the problems of high computational complexity, large memory consumption, and difficulty in deployment on mobile platforms with limited computing power in DeepLabV3+ segmentation model, an improved lightweight DeepLabV3+ algorithm was proposed to realize the segmentation and recognition of buckwheat by UAV at seedling stage. The algorithm adopted the fusion of re-parameterization visual geometry group (RepVGG) and mobile vision transformer (MobileViT) modules to establish the backbone network for feature extraction. At the same time, the squeeze-and-excitation networks (SENet) attention mechanism was introduced into the RepVGG network structure to capture more global semantic information by using the correlation between channels, and ensure the performance of buckwheat segmentation. Experimental results showed that compared with fully convolutional networks (FCN), pyramid scene parsing network (PSPNet), dense atrous spatial pyramid pooling (DenseASPP), DeepLabV3, and DeepLabV3+ models, the improved algorithm proposed greatly reduced the model parameters, making it more suitable for deployment on mobile terminals. The mean pixel accuracy (mPA) and mean intersection over union (mIoU) on the selfbuilt buckwheat segmentation dataset were 97.02% and 91.45%, the overall parameters, floatingpoint operations (FLOPs) and inference speed were 9.01×106, 8.215×1010 and 37.83 f/s, respectively, with the best performance. In the full-size image segmentation, the mPA and mIoU for buckwheat segmentation can meet the requirements at different flight heights, which had good segmentation ability and inference speed. The algorithm can provide technical support for the later buckwheat seed replacement, fertilization maintenance, and growth monitoring, and promote the intelligent development of small and coarse grain industry.

 

Keywords: buckwheat at seedling stage; UAV remote sensing; image segmentation; DeepLabV3+; lightweight

 

Download Full Text:

PDF


References


CHEN Shanxiong, WU Sheng, YU Xianping, et al. Buckwheat disease recognition using convolution neural network combined with image processing[ J]. Transactions of the CSAE, 2021 , 37(3) ; 155 -163. (in Chinese)

LL Shaozhong, DU Wenliang, CHEN Zhen, et al. On-line measuring method of buckwheat hulling efficiency parameters based on machine vision [J]. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50( 10) : 35 -43. (in Chinese)

XU S, ZHANG Y, DONG W, et al. Early identification and localization algorithm for weak seedlings based on phenotype detection and machine learning [J ]. Agriculture, 2023, 13(1): 212.

ZOU K, CHEN X, WANG Y, et al. A modified U - Net with a specific data argumentation method for semantic segmentation of weed images in the field [J ]. Computers and Electronics in Agriculture, 2021 , 187; 106242.

KONG S, LI J, ZHA1 Y, et al. Real-time detection of crops with dense planting using deep learning at seedling stage [ J]. Agronomy, 2023, 13(6): 1503.

DIAO Z, GUO P, ZHANG B, et al. Maize crop row recognition algorithm based on improved UNet network[J]. Computers and Electronics in Agriculture, 2023, 210: 107940.

LIU X, QI J, ZHANG W, et al. Recognition method of maize crop rows at the seedling stage based on MS-ERFNet model [ J]. Computers and Electronics in Agriculture, 2023, 211 ; 107964.

SUN J, ZHOU J, HE Y, et al. RL-DeepLabv3 + : a lightweight rice lodging semantic segmentation model for unmanned rice harvester [ J]. Computers and Electronics in Agriculture, 2023, 209; 107823.

ZUO X, LIN H, WANG D, et al. A method of crop seedling plant segmentation on edge information fusion model [ J ]. IEEE Access, 2022, 10: 95281 -95293.

ZHANG P, SUN X, ZHANG D, et al. Lightweight deep learning models for high-precision rice seedling segmentation from UAV-based multispectral images [J] . Plant Phenomics, 2023, 5: 0123.

WAN H, ZENG X, FAN Z, et al. U2ESPNet - a lightweight and high-accuracy convolutional neural network for real-time semantic segmentation of visible branches [ J]. Computers and Electronics in Agriculture, 2023, 204: 107542.

ZHANG Y, LV C. Tinysegformer; a lightweight visual segmentation model for real-time agricultural pest detection [J]. Computers and Electronics in Agriculture, 2024, 218; 108740.

ZOU K, LIAO Q, ZHANG F, et al. A segmentation network for smart weed management in wheat fields J . Computers and Electronics in Agriculture, 2022, 202: 107303.

WANG C, DU P, WU H, et al. A cucumber leaf disease severity classification method based on the fusion of DeepLabY'3 + and U - Net [J ]. Computers and Electronics in Agriculture, 2021 , 189: 106373.

PENG H, ZHONG J, LIU H, et al. ResDense - focal - DeepLabV3 + enabled litchi branch semantic segmentation for robotic Harvesting[ J]. Computers and Electronics in Agriculture, 2023, 206: 107691.

НЕ X, XUAN S, WANG К, et al. DeepLabV3 + image segmentation algorithm fusing cumulative distribution function and channel attention mechanism J . Journal of Computer Applications, 2023, 43(3) ; 936.

GAO Yingli, ZHAO Yuwei, YANG Lulu, et al. Weed identification method in rice field based on improved DeepLabV3 [J]. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54( 12) : 242 -252. (in Chinese)

GU Wenjuan, WEI Jin, YIN Yanchao, et al. Multi class segmentation method for tomato images based on improved DeepLabv3 [ J]. Transactions of The Chinese Society for Agricultural Machinery, 2023, 54 ( 12); 242 -252. (in Chinese)

PEI Y, LIU Y, LING N. MobileViT - GAN; a generative model for low bitrate image coding[ C]//2023 IEEE International Conference on Visual Communications and Image Processing (VCIP) , 2023.

FANG M, LIANG X, FU F, et al. Attention mechanism based semi-supervised multi-gain image fusion[ J ]. Symmetry, 2020, 12(3): 451.

W ANG Z, DONG L, ZHANG M, et al. Keniu - Net: multi-branch net framework for 3D brain tumor segmentation [ J ]. Journal of Circuits, Systems and Computers, 2022,31 ( 12) ;2250215.

МАО Нао, LI Xinli, WANG Xiaowei, et al. Research on semantic segmentation of transformer substation image based on multi-category Focal Loss function [J] . Journal of North China Electric Power University (Natural Science Edition), 2022, 49(5):84 -92. (in Chinese)

ZHAO C. A novel semantic segmentation method based on improved FCN for unmanned driving [С] // 2023 IEEE 6th International Conference on Automation, Electronics and Electrical Engineering (AUTEEE) , 2023.

WU Hua, ZHANG Xinchang, SUN Ying, et al. Building extraction in complex scenes based on the fusion of multi-feature improved PSPNet model [ J ]. Bulletin of Surveying and Mapping, 2021 (6) ; 21 -27. (in Chinese)

WU K, ZHANG S, XIE Z. Monocular depth prediction with residual DenseASPP network [J ]. IEEE Access, 2020(99); 1.


Refbacks

  • There are currently no refbacks.