Tobacco Leaf Vein Segmentation Method Based on Hyperspectral Imaging and GAN-SA-UNet Algorithm
Abstract
As an important feature of plants, leaf veins contain physiological and genetic information. Aiming at the problems of blurred edge segmentation and low segmentation accuracy of small veins in complex leaf texture state, a GAN-SA-UNet vein segmentation algorithm was proposed with tobacco leaves as the research object. The spectral information of veins and leaves was obtained by hyperspectral imaging technology, and the principal component analysis ( PCA ) was used to reduce the dimension and obtain the composite map. On this basis, the spatial attention mechanismwas introduced to capture the key spatial features and improve the segmentation accuracy. At the same time, the adversarial network was introduced to optimize the generated results and improve the robustness of vein segmentation. The results showed that the interpretation rate of the first three principal components of PCA of leaf vein and leaf surface spectrum was 95.71%, and the spectral characteristics of the two after dimension reduction showed obvious separability. The first three principal components composite map could highlight the difference between leaf surface and leaf vein, and highlight the characteristics of leaf vein. The GAN-SA-UNet segmentation algorithm can capture the vein features in complex leaf texture images. The segmentation accuracy and intersection over union were 98.93% and 66.23%, respectively. Compared with the original model, they were increased by 0.18 percentage points and 4.21 percentage points, respectively. The inference time of single image was 4ms. It showed strong generalization ability and efficient and accurate recognition ability in the verification test of different producing areas, parts, grades and types of tobacco leaves.
Keyword: tobacco leaf vein segmentation ; hyperspectral imaging technology ; U-Net ; spatial attention mechanism ; generative adversarial networks (GANs)
Download Full Text:
PDFReferences
LAVOREL S, GRIGULIS K. How fundamental plant functional trait relationships scale-up to trade-offs and synergies in ecosystem services [j]. Journal of Ecology, 2012, 100( 1 ) ; 128 - 140.
LU Mengvao, ZHOU Qiang, JIANG Shuwen, et al. Classification method of flue-cured tobacco leaves based on deep learning and multi-scale feature fusion [J ]. Journal of Chinese Agricultural Mechanization, 2022, 43 ( 1 ) : 158 - 166. (in Chinese)
LU Mengvao, CHEN Dong, ZHOU Qiang, et al. Tobacco grade classification and feature visualization based on deep learning [ J ]. Tobacco Science and Technology, 2023, 56(6) ; 92 - 100. (in Chinese)
LIANG Miao, LIU Maolin, LIU Xiangzhen, et al. Directional de-stemming process based on the relationship between tobacco leaf size and tobacco stem [ J ]. Tobacco Science and Technology, 2020,53(5) ; 77 -82. ( in Chinese)
ZHU Peng. Leaf vein extraction method based on Canny operator and hue information [ D], Wuhan; Wuhan University of Technology, 2016. (in Chinese)
TIAN Tian, LIU Qiang, YIN Shiwei, et al. Leaf vein extraction algorithm based on total variation denoising and eight-direction sobel operator [J] . Acta Agriculturae Zhejiangensis, 2015,27(4) : 678 -683. (in Chinese)
ZHU Jiyou, YU Qiang, YANG Di, et al. Extraction and optimization of microscopic image leaf vein network based on eCognition software [J] . Transactions of the Chinese Society for Agricultural Machinery, 2019, 50( 1 ) : 51 -57. (in Chinese)
LI L, HU W, LU J, et al. Leaf vein segmentation with self-supervision [ J ]. Computers and Electronics in Agriculture, 2022, 203: 107352.
LIU X, XU B, GU W, et al. Plant leaf veins coupling feature representation and measurement method based on DeepLabV3 + J . Frontiers in Plant Science, 2022, 13; 1043884.
SONG P, YU Y, ZHANG Y. TTH — Net; two-stage Transformer-CNN hybrid network for leaf vein segmentation[ J] . Applied Sciences, 2023, 13( 19) : 11019.
WENG Haiyong, LI Xiaobin, XIAO Kangsong, et al. Instance segmentation model for citrus microscopic image of main leaf vein based on Mask R -CNN [J]. Transactions of the Chinese Society for Agricultural Machinery, 2023,54(7) ; 252 - 258, 271. (in Chinese)
XU II, BLONDER B, JODRA M , et al. Automated and accurate segmentation of leaf venation networks via deep learning [J] . New Phytologist, 2020, 229 ( 1 ) : 631 -648.
CAI W, WANG B, ZENG F. CUDU-Net: collaborative up-sampling decoder U — Net for leaf vein segmentation[ J]. Digital Signal Processing, 2024, 144: 104287.
GOODFELLOW I, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial nets [J]. Advances in Neural Information Processing Systems, 2014; 2672 -2680.
ODENA A, OLAH C, SHLENS J. Conditional image synthesis with auxiliary classifier gans [С] // International Conference on Machine Learning. PMLR, 2017 : 2642 -2651.
LUC P, COUPRIE C, CHINTALA S, et al. Semantic segmentation using adversarial networks[J]. arXiv Preprint, arXiv: 1611.08408, 2016.
WEI Shuo, LUO Dingqi, GU Yong,et al. Analysis of water status and morphological changes of upper leaves of flue-cured tobacco with stems [J] . Journal of Southern Agricultural, 2017,48(9) : 1659 - 1664. ( in Chinese)
HUANG Yiqi, LIU Xianghuan, HUANG Zhenyu, et al. Identification technology of early leaf spot and rust of sugarcane based on hyperspectral imaging[ J]. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(4) ; 259 -267. (in Chinese)
LONG J, SHELHAMER E, DARRELL T. Fully convolutional networks for semantic segmentation [ С ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015; 3431 -3440.
CHEN L C, PAPANDREOU G, KOKKINOS I, et al. Deeplab; semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 40(4) : 834 -848.
GONG D, YANG J, GAN X, et al. Fully automated construction of a deep U - Net network model for medical image segmentation[ J]. Sensors & Materials, 2023, 35( 10) :4633 -4652.
HAN Zhenhao, LI Jia, YUAN Yanwei, et al. Orchard visual navigation path recognition method based on U — Net network [J] . Transactions of the Chinese Society for Agricultural Machinery, 2021 , 52( 1 ) ; 30 -39. (in Chinese)
GUO C, SZEMENYEI M, YI Y, et al. Sa-unet; spatial attention U-Net for retinal vessel segmentation [ С ] //2020 25th International Conference on Pattern Recognition ( ICPR). IEEE, 2021 ; 1236 - 1242.
Refbacks
- There are currently no refbacks.
