Banca de QUALIFICAÇÃO: GEAN CARLOS DA SILVA SANTOS

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : GEAN CARLOS DA SILVA SANTOS
DATE: 15/04/2026
TIME: 08:00
LOCAL: Google meet - https://meet.google.com/ygk-fkqq-xay
TITLE:

Automated seed vigor assessment using seedling image analysis: a software applied to Cenostigma pyramidale

KEY WORDS:

Cenostigma pyramidale; automated image analysis; seed vigor; software; digital phenotyping.

PAGES: 29
BIG AREA: Ciências Agrárias
AREA: Recursos Florestais e Engenharia Florestal
SUBÁREA: Silvicultura
SPECIALTY: Sementes Florestais
SUMMARY:

Cenostigma pyramidale is a tree species native to the Caatinga biome with high ecological relevance and potential for forage, apiculture, and restoration of degraded areas. However, physiological seed quality assessment in native species is often constrained by limited seed availability and by the destructive and time-consuming nature of conventional vigor tests. This study aimed to develop and validate a software tool for automated seed vigor assessment based on seedling image analysis. Eight seed lots obtained from the Núcleo de Ecologia e Monitoramento Ambiental, Universidade Federal do Vale do São Francisco (Petrolina, PE, Brazil), were initially characterized using standard germination and vigor tests. Seed vigor was subsequently estimated from morphological traits automatically extracted from seedling images acquired at 5, 7, and 10 days after sowing, including total seedling length, shoot length, and primary root length, as well as growth, uniformity, vigor, and corrected vigor indices. The experiment followed a completely randomized design with four replicates of 50 seeds. Data were subjected to analysis of variance, Scott–Knott grouping test (p ≤ 0.05), Pearson correlation, and principal component analysis. Image-derived variables, particularly those obtained at 10 days, showed significant correlations with conventional vigor tests (r > 0.50; p < 0.05), indicating strong predictive ability for seed lot physiological performance. The proposed software enables rapid, objective, automated, and non-destructive evaluation of physiological seed quality in C. pyramidale, reducing the time required for vigor estimation compared with traditional methods.

COMMITTEE MEMBERS:
Presidente - 1880265 - MARCIO DIAS PEREIRA
Externo ao Programa - 1204363 - ANTONINO ALVES FEITOSA NETO - UFRNExterno à Instituição - ANDRÉ DANTAS DE MEDEIROS - UFV
Externo à Instituição - CHARLINE ZARATIN ALVES - UFMS