VALIDAÇÃO DE MÉTODO AUTOMATIZADO PARA MENSURAÇÃO DO COMPRIMENTO DE PLÂNTULAS DE HÍBRIDOS DE FORRAGEIRAS

Abstract

The measurement of seedling length is an important tool for evaluating seed vigor and initial performance, traditionally carried out through manual measurements, which may require more time and present operational variability. In this context, automated methods based on image analysis have emerged as alternatives capable of providing greater speed, standardization, and precision in evaluations. Thus, the present study aimed to validate an automated method for measuring the length of forage hybrid seedlings, comparing it with the traditional method for the variables shoot length, root system length, and total length. The experiment was conducted using a completely randomized design in a factorial scheme involving forage hybrids, measurement methods, and replications. Data were subjected to descriptive statistical analyses, tests of normality and homogeneity of variances, t-tests for mean comparison, and evaluation of error indicators, including standard deviation of the error, mean absolute error, and root mean square error. The results indicated no statistically significant differences between the evaluated methods, with homogeneous variances, similar means, and overlapping confidence intervals, demonstrating agreement between the methodologies. Furthermore, the error indicators presented comparable values, showing similar levels of precision in the estimates. It is concluded that the automated image analysis method is efficient and reliable for measuring the length of forage hybrid seedlings and can be used as an alternative to the conventional method, contributing to the optimization of evaluations in agronomic studies.