| Popis: |
Traditional morphometrics incorporates manual measurements of quantitative characteristics and the calculation of variables as basic functions of the raw data, i.e. linear distances and angles, but this sometimes overlooks crucial biomathematical features of the initial measurements. Geometric morphometric methods, i.e. digital image processing, are based on the Cartesian coordinates of measurement points, also known as landmarks. Landmarks allow for effective visual representations of mathematical effects as real shape or form deformations - the geometry of the measured landmark arrangement is maintained by the collection of their coordinates. In order to test the feasibility of the digital protocol, we used both traditional and digital approach to calculating morphometric features to explore the distinctions between Podarcis sicula and Podarcis melisellensis. We used digital callipers for linear measurements of adult lizards: snout-vent length (SVL), tail length (TL), head length (HL) and head height (HH). For our digital method, we used a camera on a tripod and took photographs of lizards. Then we extracted the lizards from the pictures and analysed it with our newly developed Python 3.9 script. Anova results on 10 randomly selected lizards showed no real impact of the methodological approaches on the values in any of the measurement categories. Pearson correlation coefficients showed high correlation between different methodological approaches. Our preliminary findings, confirm our newly developed protocol as a valid replacement for the traditional method. Even though we observed minimal absolute differences, studies should not explicitly compare measurements taken with the two methods without first estimating a correction factor. |
