This paper focuses on leaf cultivar classification, which is a long-established challenge in agricultural artificial intelligence. The difficulties in this task come from the fact that there is large amount of intra-class variability, arising from form changes during the growth of leaves and different physical development, and subtle inter-class differences, originating by belonging to the same species. To cope with this challenging task, we study the possibility of using deep learning techniques for distinguishing leaf cultivars. We employed a soybean leaf cultivar dataset and conducted extensive experiments on it for a comparison study of handcrafted methods and deep learning methods on leaves cultivar recognition tasks. The experimental results indicate the supervisor performance of the deep learning methods over the traditional methods.

Jiawei You. Leaf Image Classification Using Deep Learning Network. Academic Journal of Computing & Information Science (2021), Vol. 4, Issue 3: 109-115. https://doi.org/10.25236/AJCIS.2021.040317.

[1] Ale L, Sheta A, Li L, et al. Deep learning-based plant disease detection for smart agriculture[C]. 2019 IEEE Globecom Workshops (GC Wkshps). IEEE, 2019: 1-6.

[2] Magomadov V S. Deep learning and its role in smart agriculture[C]. Journal of Physics: Conference Series. IOP Publishing, 2019, 1399(4): 044109.

[3] Setiyono T D, Weiss A, Specht J E, et al. Leaf area index simulation in soybean grown under near-optimal conditions[J]. Field Crops Research, 2008, 108(1): 82-92.

[4] Cavassim J E, Bespalhok Filho J C, Alliprandini L F, et al. AMMI analysis to determine relative maturity groups for the classification of soybean genotypes[J]. Journal of Agronomy, 2013, 12(4): 168-178.

[5] Lee S H, Chan C S, Mayo S J, et al. How deep learning extracts and learns leaf features for plant classification. Pattern Recognition, 2017, 71: 1-13.

[6] Wang Y, Morariu V I, Davis L S. Learning a discriminative filter bank within a cnn for fine-grained recognition[C]. Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 4148-4157.

[7] Yang Z, Luo T, Wang D, et al. Learning to navigate for fine-grained classification[C]. Proceedings of the European Conference on Computer Vision (ECCV). 2018: 420-435.

[8] Hearn D J. Shape analysis for the automated identification of plants from images of leaves[J]. Taxon, 2009, 58(3): 934-954.

[9] Meade C, Parnell J. Multivariate analysis of leaf shape patterns in Asian species of the Uvaria group (Annonaceae)[J]. Botanical Journal of the Linnean Society, 2003, 143(3): 231-242.

[10] Plotze R O, Falvo M, Pádua J G, et al. Leaf shape analysis using the multiscale Minkowski fractal dimension, a new morphometric method: a study with Passiflora (Passifloraceae)[J]. Canadian Journal of Botany, 2005, 83(3): 287-301.

[11] Wang B, Gao Y. Hierarchical string cuts: a translation, rotation, scale, and mirror invariant descriptor for fast shape retrieval[J]. IEEE Transactions on Image Processing, 2014, 23(9): 4101-4111.

[12] Larese M G, Namías R, Craviotto R M, et al. Automatic classification of legumes using leaf vein image features [J]. Pattern Recognition, 2014, 47(1): 158-168.

[13] Park J K, Hwang E J, Nam Y. Utilizing venation features for efficient leaf image retrieval[J]. Journal of Systems and Software, 2008, 81(1): 71-82.

[14] McKown A D, Dengler N G. Shifts in leaf vein density through accelerated vein formation in C4 Flaveria (Asteraceae)[J]. Annals of Botany, 2009, 104(6): 1085-1098.

[15] Wang B, Gao Y, Yuan X, et al. From species to cultivar: Soybean cultivar recognition using joint leaf image patterns by multiscale sliding chord matching[J]. biosystems engineering, 2020, 194: 99-111.

[16] Tan X, Triggs B. Fusing Gabor and LBP feature sets for kernel-based face recognition[C]. International workshop on analysis and modeling of faces and gestures. Springer, Berlin, Heidelberg, 2007: 235-249.

[17] Ojala T, Pietikäinen M, Harwood D. A comparative study of texture measures with classification based on featured distributions[J]. Pattern recognition, 1996, 29(1): 51-59.

[18] Ojala T, Pietikainen M, Harwood D. Performance evaluation of texture measures with classification based on Kullback discrimination of distributions[C]. Proceedings of 12th international conference on pattern recognition. IEEE, 1994, 1: 582-585.

[19] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]. Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 770-778.

[20] Lowe D G. Distinctive image features from scale-invariant keypoints[J]. International journal of computer vision, 2004, 60(2): 91-110.

[21] Krizhevsky A, Sutskever I, Hinton G E. Imagenet classification with deep convolutional neural networks [J]. Advances in neural information processing systems, 2012, 25: 1097-1105.

[22] Chen J, Deng L. A primal-dual method for training recurrent neural networks constrained by the echo-state property[J]. arXiv preprint arXiv:1311.6091, 2013.

[23] Graves A, Jaitly N. Towards end-to-end speech recognition with recurrent neural networks[C]. International conference on machine learning. PMLR, 2014: 1764-1772.

[24] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[J]. arXiv preprint arXiv:1409.1556, 2014.

[25] Howard A G, Zhu M, Chen B, et al. Mobilenets: Efficient convolutional neural networks for mobile vision applications[J]. arXiv preprint arXiv:1704.04861, 2017.

[26] Hu J, Shen L, Sun G. Squeeze-and-excitation networks[C]. Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 7132-7141.

[27] Tan M, Le Q. Efficientnet: Rethinking model scaling for convolutional neural networks[C]. International Conference on Machine Learning. PMLR, 2019: 6105-6114.

[28] Novotný P, Suk T. Leaf recognition of woody species in Central Europe[J]. Biosystems Engineering, 2013, 115(4): 444-452.

[29] Hu R, Jia W, Ling H, et al. Multiscale distance matrix for fast plant leaf recognition[J]. IEEE transactions on image processing, 2012, 21(11): 4667-4672.