Automated grapevine cultivar classification based on machine learning using leaf morpho-colorimetry, fractal dimension and near-infrared spectroscopy parameters

Abstract The application of computer vision algorithms and chemometric fingerprinting using near-infrared spectrometry (NIR) of plant leaves, offers enhanced capabilities for ampelography by providing more accurate methods to discriminate leaves based on morphological parameters, and chemometrics, respectively. This paper showed that machine learning algorithms based on morpho-colorimetric parameters and NIR analysis separately, were able to automatically classify leaves of 16 grapevine cultivars. The artificial neural network (ANN) model developed with morpho-colorimetric parameters as inputs (Model 1), and 16 cultivars as targets, rendered an accuracy of 94% to classify leaves for all cultivars studied. The ANN model obtained with the NIR spectra per leaf as inputs (Model 2), and the real classification as targets, rendered 92% accuracy. The automatic extraction of morpho-colorimetric data, NIR chemical fingerprinting and machine learning modelling rendered rapid, accurate and non-destructive methods for cultivar classification, which can aid management practices.

[1]  Michael F. Covington,et al.  A Modern Ampelography: A Genetic Basis for Leaf Shape and Venation Patterning in Grape1[C][W][OPEN] , 2013, Plant Physiology.

[2]  Grape varieties (Vitis vinifera L.) from the Balearic Islands: genetic characterization and relationship with Iberian Peninsula and Mediterranean Basin , 2012, Genetic Resources and Crop Evolution.

[3]  Stefano Mancuso,et al.  Identifying Olive (Olea europaea ) Cultivars Using Artificial Neural Networks , 1999 .

[4]  Seishi Ninomiya,et al.  Discrimination of soybean leaflet shape by neural networks with image input , 2000 .

[5]  Karl J. Niklas,et al.  The evolution and functional significance of leaf shape in the angiosperms. , 2011, Functional plant biology : FPB.

[6]  Odemir Martinez Bruno,et al.  Fractal analysis of leaf-texture properties as a tool for taxonomic and identification purposes: a case study with species from Neotropical Melastomataceae (Miconieae tribe) , 2010, Plant Systematics and Evolution.

[7]  Giuseppe Messina,et al.  Discrimination and identification of morphotypes of Banksia integrifolia (Proteaceae) by an Artificial Neural Network (ANN), based on morphological and fractal parameters of leaves and flowers. , 2009 .

[8]  Sigfredo Fuentes,et al.  Assessment of beer quality based on foamability and chemical composition using computer vision algorithms, near infrared spectroscopy and machine learning algorithms. , 2018, Journal of the science of food and agriculture.

[9]  Paolo Remagnino,et al.  Plant species identification using digital morphometrics: A review , 2012, Expert Syst. Appl..

[10]  Pierre Dutilleul,et al.  Advances in the implementation of the box-counting method of fractal dimension estimation , 1999, Appl. Math. Comput..

[11]  P. Bodor,et al.  Stability of Ampelometric Characteristics of Vitis vinifera L. cv. ‘Syrah’ and ‘Sauvignon blanc’ Leaves: Impact of Within-vineyard Variability and Pruning Method/Bud Load , 2016 .

[12]  Emil W. Ciurczak,et al.  Handbook of Near-Infrared Analysis , 1992 .

[13]  F. Vuolo,et al.  Leaf Shape Evolution Through Duplication, Regulatory Diversification, and Loss of a Homeobox Gene , 2014, Science.

[14]  D. Cozzolino,et al.  Non-destructive measurement of grapevine water potential using near infrared spectroscopy , 2011 .

[15]  Juan Fernández-Novales,et al.  Assessment of quality parameters in grapes during ripening using a miniature fiber-optic near-infrared spectrometer , 2009, International journal of food sciences and nutrition.

[16]  S. Mancuso.  Clustering of grapevine ( Vitis vinifera L.) genotypes with Kohonen neural networks , 2015 .

[17]  H. Boshoff,et al.  Non-destructive assessment of leaf composition as related to growth of the grapevine ( Vitis vinifera L. cv. Shiraz). , 2011 .

[18]  S. Mancuso.  Discrimination of grapevine (Vitis vinifera L.) leaf shape by fractal spectrum , 2015 .

[19]  Antonio Odair Santos,et al.  Grapevine leaf water potential based upon near infrared spectroscopy , 2009 .

[20]  Sigfredo Fuentes,et al.  Development of a robotic pourer constructed with ubiquitous materials, open hardware and sensors to assess beer foam quality using computer vision and pattern recognition algorithms: RoboBEER. , 2016, Food research international.

[21]  Computer vision as a method complementary to molecular analysis: grapevine cultivar seeds case study. , 2012, Comptes rendus biologies.

[22]  Anatoly A. Gitelson,et al.  Nondestructive Estimation of Anthocyanin Content in Grapevine Leaves , 2009, American Journal of Enology and Viticulture.

[23]  L. Liebovitch,et al.  A fast algorithm to determine fractal dimensions by box counting , 1989 .

[24]  Wojciech Borkowski,et al.  Fractal dimension based features are useful descriptors of leaf complexity and shape , 1999 .

[25]  Daniel Cozzolino,et al.  Rapid measurement of total non-structural carbohydrate concentration in grapevine trunk and leaf tissues using near infrared spectroscopy , 2017, Comput. Electron. Agric..

[26]  Odemir Martinez Bruno,et al.  Fractal dimension applied to plant identification , 2008, Inf. Sci..

[27]  J. Fourie.  Soil management in the Breede River Valley wine grape region, South Africa. 4. Organic matter and macro-nutrient content of a medium-textured soil. , 2016 .

[28]  Anatoly A. Gitelson,et al.  A Comparison of Two Techniques for Nondestructive Measurement of Chlorophyll Content in Grapevine Leaves , 2008 .

[29]  R. Bandinelli,et al.  Application of an artificial neural network (ANN) for the identification of grapevine genotypes , 2015 .

[30]  S. Alessandri,et al.  AmpeloCADs(Ampelographic Computer-Aided Digitizing System): An Integrated System to Digitize, File, and Process Biometrical Data FromVitisspp. Leaves , 1996, American Journal of Enology and Viticulture.

[31]  S. Mancuso.  Fractal geometry-based image analysis of grapevine leaves using the box counting algorithm , 2015 .

[32]  Samuel Ortega-Farías,et al.  Effects of grapevine (Vitis vinifera L.) water status on water consumption, vegetative growth and grape quality: An irrigation scheduling application to achieve regulated deficit irrigation , 2010 .

[33]  G. Batten,et al.  Measuring Potassium in Plant Tissues Using near Infrared Spectroscopy , 1998 .

[34]  Sergio Mugnai,et al.  Artificial neural networks as a tool for plant identification: a case study on Vietnamese tea accessions , 2009, Euphytica.

[35]  Guido Sanguinetti,et al.  LEAFPROCESSOR: a new leaf phenotyping tool using contour bending energy and shape cluster analysis. , 2010, The New phytologist.

[36]  André Ricardo Backes,et al.  Color texture analysis based on fractal descriptors , 2012, Pattern Recognit..