To achieve rapid and accurate diagnosis and recognition of three major nutrient element deficiency types in rice,one late rice variety“Huanghuazan”was selected as the research object for field cultivation trials. Different treatments are set as follows: Four nitrogen application levels are N0 (0 kg/hm2), N1 (130 kg/hm2), N2 (260 kg/hm2) and N3 (390 kg/hm2); Four phosphorus application levels are P0(0 kg/hm2), P1(300 kg/hm2), P2(600 kg/hm2), P3 (780 kg/hm²); Four potassium application levels are K0 (0 kg/hm²), K1 (90 kg/hm2), K2 (180 kg/hm2) and K3 (270 kg/hm2). During the tillering and jointing stages of rice growth, the high-resolution image data of the top three fully expanded leaves from each tiller were scanned, and after generative adversarial network (GAN) reconstruction of rice image with super-resolution, the data underwent normalization and expansion through image preprocessing; Additionally, attention mechanisms and soft thresholding functions were integrated into the residual block while preserving the backbone structure of residual neural network Resnet34; Furthermore, pre-trained weights obtained from the ImageNet dataset were transferred to the model with an improved residual structure. The results indicate that: The improvement of Resnet34 network yielded extraordinary results, exhibiting a remarkable accuracy of 98.98% and 98.10% within the severe fertilizer deficiency gradients during these pivotal growth phases. Resnet34 network demonstrated robustness in moderate deficiency scenarios, and achieved the accuracies of 97.99% and 95.90%, respectively. Regarding excessive fertilizer application, the Resnet34 model maintained a commendable performance and achieved accuracies of 91.87% and 88.49% across the respective gradients. Compared with the pre-improvement network, the three-element stress classification accuracy of the rice image test set at the tillering stage and jointing stage increased by 29.58% and 29.75%; The analysis of the confusion matrix indicated that the identification accuracy of nitrogen stress was superior, and the training curve demonstrated faster convergence speed. In conclusion, the model established in this study exhibits exceptional proficiency in diagnosing nutrient deficiency stress during the pivotal tillering and jointing stages of rice growth, and can accurately predict the rice nutritional status. This study provides a scientific reference for the nutritional diagnosis of nitrogen, phosphorus and potassium in rice.
PengJ W, LiuQ, RongX M, ZhuH M, XieG X, TangG R. Effects of different rational ratio of N, P, K fertilizer and amount of N fertilizer on photosynthesis character and yield of rice [J]. Journal of Hunan Agricultural University, 2004, 30(2): 123-127 (in Chinese)
LiW G, RenY L. The effects of combined N P K Si fertilization on rice yield and component factors[J]. Journal of Shanxi Agricultural Sciences, 2001, 29(1): 53-58 (in Chinese)
[7]
何小亮, 祁秀平. 水稻缺肥或施肥过量各元素表现症状分析[J]. 吉林农业, 2013(8): 8
[8]
HeX L, QiX P. Analysis on the symptoms of various elements in rice with fertilizer deficiency or excessive fertilization[J]. Jilin Agriculture, 2013(8): 8 (in Chinese)
AoH J, WangS H, ZouY B, PengS B, ChengZ W, LiuW, TangQ Y. Characteristics of nutrient uptake and utilization of super hybrid rice under different fertilizer application rates[J]. Scientia Agricultura Sinica, 2008, 41(10): 3123-3132 (in Chinese)
HouY P, HanL G, KongL L, YinC X, QinY B, LiQ, XieJ G. Nutrient absorption, translocation in rice and soil nitrogen equilibrium under different nitrogen application doses[J]. Journal of Plant Nutrition and Fertilizer, 2015,21(4): 836-845 (in Chinese)
[19]
DuanG L, LiuW J, ChenX P, HuY, ZhuY G. Association of arsenic with nutrient elements in rice plants[J]. Metallomics,2013, 5(7): 784-792
[20]
ChenL S, LinL, CaiG Z, SunY Y, HuangT, WangK, DengJ S. Identification of nitrogen, phosphorus, and potassium deficiencies in rice based on static scanning technology and hierarchical identification method [J]. PLoS One, 2014, 9(11): e113200
HanS, WuJ, HuX R, SangY S, HuR, HuP, GuoX S. Responses of different nitrogen fertilizer single basal application methods on rapeseed( Brassica napus L) in different regions of Anhui Province [J]. Soil and Fertilizer Sciences in China, 2018(2): 82-88 (in Chinese)
LvS P, LiD H, XianR H. Research status of deep learning in agriculture of China[J]. Computer Enginnering and Applications, 2019, 55(20): 24-33, 51 (in Chinese)
ZhouQ, YangH Y, YangJ, SunY T, SunA Z, YangW J. Feasibility study of BP neural network and probabilistic neural network for nitrogen nutrition diagnosis of rice images[J]. Journal of Plant Nutrition and Fertilizers, 2019, 25(1): 134-141 (in Chinese)
ZhangL P, YangH Y, QianZ, GuoZ W. Nitrogen nutrition diagnosis in rice based on improved VGG16 network and transfer learning[J]. Journal of China Agricultural University, 2023, 28(6): 219-229 (in Chinese)
YangH Y, GuoZ W, GuoG F, HuangJ L, QianZ, ZhangL P, LiuJ J. Rice nitrogen nutrition diagnosis based on stacking integrated convolutional neural network[J]. Journal of Plant Nutrition and Fertilizer, 2023, 29(3): 573-581 (in Chinese)
[35]
SethyP K, BarpandaN K, RathA K, BeheraS K. Nitrogen deficiency prediction of rice crop based on convolutional neural network[J]. Journal of Ambient Intelligence and Humanized Computing, 2020, 11(11): 5703-5711
YeC, LiuY, LiuJ Z, ShuS F, LiY D, WuL F. A method for monitoring nitrogen nutrition in rice canopy based on LBPHSV+ResNet50 fusion[J]. Journal of China Agricultural University,2023, 28(1): 39-47 (in Chinese)
XuX P, ZhouW, LiangG Q, SunJ W, WangX B, HeP, XuF S, YuX C. Effects of nitrogen and density interactions on grain yield and nitrogen use efficiency of double-rice systems[J]. Journal of Plant Nutrition and Fertilizers, 2015, 21(3): 763-772
KangY Y, FengX F, YangX. Effect of excess phosphorus on growth, yield and quality of Chinese flowering cabbage[J]. Northern Horticulture, 2011(15): 70-72 (in Chinese)
[42]
王强盛. 水稻钾素营养的积累特征及生理效应[D]. 南京: 南京农业大学, 2009
[43]
WangQ S. Accumulation characteristics and physiological effects of rice potassium nutrition[D]. Nanjing: Nanjing Agricultural University,2009 (in Chinese)
[44]
GoodfellowI, Pouget-AbadieJ, MirzaM, XuB, Warde-FarleyD, OzairS, CourvilleA, BengioY. Generative adversarial networks[J]. Communications of the ACM, 2020, 63(11): 139-144
[45]
LedigC, TheisL, HuszárF, CaballeroJ, CunninghamA, AcostaA, AitkenA, TejaniA, TotzJ, WangZ H, ShiW Z. Photo-realistic single image super-resolution using a generative adversarial network [C]//2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). ,, Honolulu,: IEEE, 2017: 105-114
[46]
KingmaD, BaJ.ADAM:A method for stochastic optimization[J].Computer Science, 2014:6980
[47]
DonohoD L. De-noising by soft-thresholding [J]. IEEE Transactions on Information Theory, 1995, 41(3): 613-627
[48]
HuJ, ShenL, SunG. Squeeze-and-excitation networks [C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. ,, Salt Lake City: IEEE, 2018: 7132-7141
[49]
ZhaoM H, ZhongS S, FuX Y, TangB P, PechtM. Deep residual shrinkage networks for fault diagnosis[J]. IEEE Transactions on Industrial Informatics, 2020, 16(7): 4681-4690
[50]
孙棋. 基于数字图像处理技术的水稻氮素营养诊断研究[D]. 杭州: 浙江大学, 2008
[51]
SunQ. Study on nitrogen nutrition diagnosis of rice based on digital image processing technology[D]. Hangzhou: Zhejiang University, 2008 (in Chinese)