Statistical seed classifiers of 10 plant families representative of the Mediterranean vascular flora

Abstract

The aim of this study was to implement a seed morphometric and colorimetric database in order to develop specific statistical classifiers for 10 families representative of the Mediterranean vascular flora. The 501 analysed accessions consisted of 274 taxa belonging to 161 genera and 52 families. Images were acquired by a flatbed scanner and then elaborated by a macro software program developed for the morpho-colorimetric measurements. The mean seed weight was also calculated for each accession. Statistical classifiers were elaborated for genera and specific and infraspecific taxa discrimination within the following families: Apiaceae, Asteraceae, Boraginaceae, Brassicaceae, Caryophyllaceae, Cistaceae, Fabaceae, Lamiaceae, Poaceae and Scrophulariaceae. Such classifiers, based on Linear Discriminant Analysis and checked by cross validation, showed a performance ranging from 90.4% to 98.5% and from 87.8% to 98.3% at genus and taxon levels respectively. A first application of this technique to gymnosperm seeds was also presented. The performance of correct classification (93.6%) obtained for the genus Juniperus suggests that this technology is also reliable for gymnosperms. Seed weight improved the accuracy of the classifier system, confirming that an extensive database of bio-morphological traits constituted by a large number of seed species may be useful for taxonomic identification.

Intraspecific Variability of Seed Morphology in Capparis Spinosa L.

“…Analysis of seed size based on scanning technology and image analysis algorithms (Herridge et al, 2011;Tanabata et al, 2012) gives automated measurements of seed dimensions which may be useful in plant taxonomy and evolution studies (Grillo et al, 2010). Other methods for morphology and carpology are based on a combination of image analysis with diverse morphological parameters (Borkowski, 2001;Toma, 2008).…”

Seed image analysis provides evidence of taxonomic differentiation within the Medicago L. sect. Dendrotelis (Fabaceae)

“…Taxonomical differentiation within the Medicago 9 executed statistical comparison. This achievement highlights the importance of the introduction of these descriptors, improving the image analysis system previously developed by Grillo et al (2010) in which morphometric features were the first discriminant parameters. Also in Bacchetta, Garc ıa, Grillo, Mascia and Venora (2011), regarding the Lavatera triloba aggregate, the first three parameters with the highest discriminatory power were of morphological type, although colour evaluation was very important in this work for correct seed identification.…”

“…Bacchetta, Grillo, Mattana, and Venora (2008) characterized seeds of wild vascular plants of the Mediterranean Basin, using digital images and implementing statistical classifiers able to discriminate seeds belonging to different genera and species. Then, Grillo, Mattana, Venora, and Bacchetta (2010) improved that classification system, developing 10 specific statistical classifiers at the family level for Angiosperms and testing the system on the genus Juniperus L. (Cupressaceae), demonstrating that the method is also reliable for Gymnosperms. Recently, Pinna, Grillo, Mattana, Cañadas, and Bacchetta (2014) focused this topic on Mediterranean taxa of Juniperus at interspecific, specific and intraspecific levels, and shortly before Orr u, Grillo, Venora, and Bacchetta (2012) confirmed the effectiveness of this identification method, studying seeds of Vitis vinifera L. varieties.…”

Seed image analysis provides evidence of taxonomical differentiation within the Lavatera triloba aggregate (Malvaceae)

“…Its utility in taxonomic studies is promising due to its efficiency in discriminating accessions at inter-population level . Furthermore, the availability of a broad database of morpho-colorimetric seed features for an adequate amount of families, genera and species Grillo et al, 2010) may enable the identification of species or other taxonomic groups (Bacchetta et al, 2011;Fagundez and Izco, 2009;Mattana et al, 2008), opening new perspectives in plant taxonomy, and also offering the opportunity for germplasm banks to identify their accessions in a standardized, speedy way.…”

“…A macro specifically developed for the characterization of wild seeds , and later modified to measure further 20 morpho-colorimetric seed features , was used to perform the measurements of seed size, shape and colour. In addition, mean seed weight of each accession was included to increase the discriminant power of the statistical analysis (Table 2), as reported in Grillo et al (2010). Mean seed weight was determined weighing 20 seeds, 10 times, on a four decimal places scale.…”