Abstract
Morphological characteristics are still the most used tools for the identification of plant species. In this context, leaves are the most available plant organ used, given their perenniality and diversity. Computer-based image analysis help extract morphological features for botanical identification and maybe a solution to taxonomic problems requiring extensively trained specialists that use visual identification as the primary method for this approach. In this study, were collected 40 leaves from 30 trees and shrub species from 19 different families. Here, we compared two popular image capture devices: a scanner and a mobile phone. Features analyzed comprised color, shape, and texture. The performance of both devices was compared through three machine learning algorithms (adaptive boosting—AdaBoost, random forest, support vector machine—SVM) and an artificial neural network model (deep learning). Computer vision showed to be efficient in the identification of species (higher than 93%), with similar results obtained for both mobile phones and scanners. The algorithms SVM, random forest and deep learning performed more efficiently than AdaBoost. Based on the results, we present the Inovtaxon Plant Species Identification Software, available at https://github.com/DeborahBambil/Inovtaxon.
This is a preview of subscription content, log in via an institution to check access.
References
Arafat SY, Saghir MI, Ishtiaq M, Bashir U (2016) Comparison of techniques for leaf classification. Digit Inf Commun Technol Appl. https://doi.org/10.1109/DICTAP.2016.7544015
Chase MW, Christenhusz MJM, Fay MF, Byng JW, Judd WS, Soltis DE et al (2016) An update of the angiosperm phylogeny group classification for the orders and families of flowering plants: APG IV. Bot J Linn Soc 181:1–20. https://doi.org/10.1111/boj.12385
Filho PL, Oliveira LS, Nisgoski S, Britto AS (2014) Forest species recognition using macroscopic images. Mach Vis Appl 25:1019–1031. https://doi.org/10.1007/s00138-014-0592-7
Geman D, Geman S, Hallonquist N, Younes L (2015) Visual turing test for computer vision systems. Proc Natl Acad Sci USA 112:3618–3623. https://doi.org/10.1073/pnas.1422953112
Hasinoff SW, Kutulakos KN (2011) Light-efficient photography. IEEE Trans Pattern Anal Mach Intell 33:2203–2214. https://doi.org/10.1109/TPAMI.2011.62
Johnson RA, Wichern DW (1999) Applied multivariate statistical analysis. Prentice-Hall, Upper Saddle River
Kan HX, Jin L, Zhou FL (2017) Classification of medicinal plant leaf image based on multi-feature extraction. Pattern Recogn Image Anal 27:581–587. https://doi.org/10.1134/S105466181703018X
Kremic E, Subasi A (2016) Performance of random forest and SVM in face recognition. Arab J Inf Technol 13:287–293
Lang S, Bravo-Marquez F, Beckham C, Hall M, Frank E (2019) WekaDeeplearning4j: a deep learning package for Weka based on Deeplearning4j. Knowl Based Syst 178:48–50. https://doi.org/10.1016/j.knosys.2019.04.013
Larese MG, Namías R, Craviotto RM, Arango MR, Gallo C, Granitto PM (2014) Automatic classification of legumes using leaf vein image features. Pattern Recogn 47:158–168. https://doi.org/10.1016/j.patcog.2013.06.012
Lukic M, Tuba E, Tuba M (2017) Leaf recognition algorithm using support vector machine with Hu moments and local binary patterns. In: 2017 IEEE 15th international symposium on applied machine intelligence and informatics (SAMI) IEEE 000485-000490
Mattila H, Valli P, Pahikkala T, Teuhola J, Nevalainen OS, Tyystjärvi E (2013) Comparison of chlorophyll fluorescence curves and texture analysis for automatic plant identification. Precision Agric 14:621–636. https://doi.org/10.1007/s11119-013-9320-y
Olsen A, Han S, Calvert B, Ridd P, Kenny O (2015) In situ leaf classification using histograms of oriented gradients. Int Conf Austral Digit Image Comput Tech Appl. https://doi.org/10.1109/DICTA.2015.7371274
Patil JK, Kumar R (2017) Analysis of content based image retrieval for plant leaf diseases using color, shape and texture features. Eng Agric Environ Food 10:69–78. https://doi.org/10.1016/j.eaef.2016.11.004
Pham NH, Le TL, Grard P, Nguyen VN (2013) Computer aided plant identification system. Comput Manage Telecommun Int Conf IEEE. https://doi.org/10.1109/ComManTel.2013.6482379
Seager TP, Hinrichs MM (2017) Technology and science: innovation at the International Symposium on Sustainable Systems and Technology, pp 1–5. https://doi.org/10.1007/s10669-017-9630-0
Şekeroğlu B, İnan Y (2016) Leaves recognition system using a neural network. Procedia Comput Sci 102:578–582. https://doi.org/10.1016/j.procs.2016.09.445
Soleimanizadeh S, Mohamad D, Saba T, Rehman A (2015) Recognition of partially occluded objects based on the three different color spaces (RGB, YCbCr, HSV). 3D Res 6:1–10. https://doi.org/10.1007/s13319-015-0052-9
Wäldchen J, Mäder P (2018) Plant species identification using computer vision techniques: a systematic literature review. Arch Comput Methods Eng 25:507–543. https://doi.org/10.1007/s11831-016-9206-z
Wen X, Shao L, Xue Y, Fang W (2015) A rapid learning algorithm for vehicle classification. Inf Sci 295:395–406. https://doi.org/10.1016/j.ins.2014.10.040
Funding
Funding was provided by Fundação de Apoio ao Desenvolvimento do Ensino, Ciência e Tecnologia do Estado de Mato Grosso do Sul.
Author information
Authors and Affiliations
Corresponding author
Electronic supplementary material
Below is the link to the electronic supplementary material.
10669_2020_9769_MOESM1_ESM.eps
Supplementary file1 (EPS 1777 kb) Fig A1. Summary of the flowchart. Forty leaves were taken from each of the 30 trees and shrub species from Cerrado. Pictures were taken with a popular mobile phone and scanner. The leaf images were from adaxial and abaxial surfaces, generating a bank with 4800 images. The leaf image features were color, shape, and texture. These features were extracted and tested with machine learning algorithms: AdaBoost, random forest, SVM, and deep learning.
10669_2020_9769_MOESM2_ESM.eps
Supplementary file2 (EPS 3656 kb) Fig A2. Dendrogram of cluster analysis of the four algorithms using the features of the leaf images acquired with the scanner and the mobile phone. AdaMp: AdaBoost with mobile phone, AdaSc: AdaBoost with scanner, DelMp: Deep learning with mobile phone, DelSc: Deep learning with scanner, RafMp: Random forest wit mobile phone, RafSc: Random forest with scanner, SVMMp: SVM with mobile phone, SVMSc: SVM with scanner.
10669_2020_9769_MOESM3_ESM.docx
Supplementary file3 (DOCX 17 kb) Table A1 Botanic characteristics and countries of occurrence of the 30 studied species.
10669_2020_9769_MOESM4_ESM.docx
Supplementary file4 (DOCX 15 kb) Table A2 Correctly species classification (%) with images acquired from the scanner and the cell phone analyzed through the four algorithms using the leaf surfaces individually (adaxial and abaxial) and combined.
Supplementary file5 (WMV 28303 kb)
Rights and permissions
About this article
Cite this article
Bambil, D., Pistori, H., Bao, F. et al. Plant species identification using color learning resources, shape, texture, through machine learning and artificial neural networks. Environ Syst Decis 40, 480–484 (2020). https://doi.org/10.1007/s10669-020-09769-w
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10669-020-09769-w