Accurate classification of fresh and charred grape seeds to the varietal level, using machine learning based classification method

Abstract

Grapevine (Vitis vinifera L.) currently includes thousands of cultivars. Discrimination between these varieties, historically done by ampelography, is done in recent decades mostly by genetic analysis. However, when aiming to identify archaeobotanical remains, which are mostly charred with extremely low genomic preservation, the application of the genomic approach is rarely successful. As a result, variety-level identification of most grape remains is currently prevented. Because grape pips are highly polymorphic, several attempts were made to utilize their morphological diversity as a classification tool, mostly using 2D image analysis technics. Here, we present a highly accurate varietal classification tool using an innovative and accessible 3D seed scanning approach. The suggested classification methodology is machine-learning-based, applied with the Iterative Closest Point (ICP) registration algorithm and the Linear Discriminant Analysis (LDA) technique. This methodology achieved classification results of 91% to 93% accuracy in average when trained by fresh or charred seeds to test fresh or charred seeds, respectively. We show that when classifying 8 groups, enhanced accuracy levels can be achieved using a "tournament" approach. Future development of this new methodology can lead to an effective seed classification tool, significantly improving the fields of archaeobotany, as well as general taxonomy.

A Review of Ampelometry: Morphometric Characterization of the Grape (Vitis spp.) Leaf

“…With the spread of new phenotyping and statistical evaluation methods, not only the grapevine leaf, but also, for example, the bunch [ 131 ], berry [ 132 ] and seed [ 133 ] would provide further possibilities to evaluate the diversity of the cultivars, help identification and explore the effect of the different environmental factors.…”

Rapid recognition of different sources of methamphetamine drugs based on hand-held near infrared spectroscopy and multi-layer-extreme learning machine algorithms

“…Linear discriminant analysis (LDA) is a well-known dimension reduction and classification method. 12,13 In the algorithm, the data is projected into a low dimension space so that the different classes can be well separated. When the method is used for binary classification, a set of n samples can belong to two classes C1 with n1 samples and C2 with n2 samples.…”

Determination of charring conditions of archaeological grape seeds, towards machine-learning based classification

“…1. The measurements were conducted using the NIS-Elements D software, whereas the pips were scanned using a Nikon SMZ25 stereomicroscope (Nikon, Tokyo, Japan) equipped with a Nikon DS-Ri2 microscope camera, as described previously 17 . Such quanti cation constitutes an important step toward the sortation of archeological seeds in terms of suitability for classi cation under the proposed methodology.…”

“…Previously, we have shown that grape seeds can be classi ed to the varietal level by a morphological classi cation methodology based on 3D scanning 16 . Later, we have shown that a machine-learning-based 3D image processing methodology can accurately classify fresh and charred grape seeds to the variety level 17 . The latter constitutes the foundation for the construction of massive charred seed reference libraries, using traditional varieties' collections, towards a full-scale seed-based variety classi cation platform for archaeo-botanic grape seed remains.…”