Automatic Leaf Structure Biometry: Computer Vision Techniques and Their Applications in Plant Taxonomy

Abstract

This paper proposes a new methodology to extract biometric features of plant leaf structures. Combining computer vision techniques and plant taxonomy protocols, these methods are capable of identifying plant species. The biometric measurements are concentrated in leaf internal forms, specifically in the veination system. The methodology was validated with real cases of plant taxonomy, and eleven species of passion fruit of the genus Passiflora were used. The features extracted from the leaves were applied to the neural network system to perform the classification of species. The results showed to be very accurate in correctly differentiating among species with 97% of success. The computer vision methods developed can be used to assist taxonomists to perform biometric measurements in plant leaf structures.

Geometry Feature Extraction of Shorea Leaf Venation Based on Digital Image and Classification Using Random Forest

“…Table III IV, the values employed were mean, standard deviation, and variant measurements [15], [16].…”

“…venation feature. Properties applied to differentiate fruit trees are growth, include angle (Ang), length (Len) and distance (Dis) [16] while to classify medicinal plants are densities of leaf vein, branching and ending points were employed [15]. However, to distinguish Shorea species, four attributes including, scale projection (Spr), angle difference (Adi), straightness (Str), and length ratio (Lra), were involved [9].…”

A New Leaf Venation Detection Technique for Plant Species Classification

“…For venation, Amlekar et al [4], Casanova et al [5], Kadir et al [6] [7] [8], Fern et al, [9], Mishra et al [10], Pornpanomchai et al [11], Rolland-Lagan et al [12] and De Oliveira Plotze and Bruno [13] measured the area, diameter, density, length and width for the leaf venation. However, this measure may influence by many factors such as the methods used to detect the leaf venation, the scaling, and the solution of the image.…”

The Cinderella discipline: morphometrics and their use in botanical classification

“…In botanical taxonomy, morphological characterization is the foundation of taxon description and identification, albeit often found in the formal and stylised format present in Floras and monographs. There is an opportunity for modern botanical taxonomy to explore the rapidly advancing field of morphometrics which already has some notable examples ranging from automatic leaf outline identification of Passiflora L. species (De Oliveira Plotze & Martinez Bruno, 2009), the tooth margin algorithm for Tilia L. leaf identification (Corney et al, 2012), and the use of leaf venation architecture for major angiosperm clade recognition (Wilf et al, 2016). Some computerised systems, starting with an existing classification of taxa, can use machine learning to handle the routine identification work, and then refer intransigent problems to a human expert (Clark, Corney, & Wilkin, 2017).…”