AUTHOR=Zorić Martina , Cvejić Sandra , Mladenović Emina , Jocić Siniša , Babić Zdenka , Marjanović Jeromela Ana , Miladinović Dragana TITLE=Digital Image Analysis Using FloCIA Software for Ornamental Sunflower Ray Floret Color Evaluation JOURNAL=Frontiers in Plant Science VOLUME=Volume 11 - 2020 YEAR=2020 URL=https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2020.584822 DOI=10.3389/fpls.2020.584822 ISSN=1664-462X ABSTRACT=As an aesthetic trait, ray floret color has a high value for the development of new sunflower genotypes and its position on the horticulture market. Standard methodology for the evaluation of sunflower ray florets is based on UPOV guidelines for sunflower. Major deficiency of this methodology is the necessity of high expertise from evaluators and its high subjectivity due to the human factor decision. To test the hypothesis that humans cannot distinguish colors equally, six commercial sunflower genotypes were evaluated by 100 agriculture experts, using UPOV guidelines. Moreover, the paper proposes a new digital UPOV (dUPOV), an assisted decision making methodology that relies on software image analysis but still leaves the final decision to the evaluator. For the purpose we created new Flower Color Image Analysis (FloCIA) software for sunflower digital image segmentation and automatic classification into one of the categories given by the UPOV guidelines for sunflower. To confirm the benefits and relevance of this method, accuracy of the newly-developed software was investigated using 153 photographs of sunflowers and expert evaluator answers which had been defined as the ground truth. The FloCIA enabled visualizations of segmentation of the examined sunflower genotype images, its two dominant color clusters, percentages of pixels belonging to each UPOV color category with graphical representation in ab plane of Lab color space, and the position of the examined sunflower mean vector in Lab color space in relation to the mean vectors of the UPOV category. Precision (repeatability) was greater between dUPOV based expert color evaluation and software evaluation than between two UPOV based evaluations performed by the same expert. Using expert evaluations obtained using dUPOV methodology as the ground truth, the accuracy of FloCIA software on the image dataset containing 153 photographs of sunflowers was 91.50%, indicated high accuracy of color category matching. This visual presentation can serve as a guideline for evaluators to determine the dominant color and to conclude if more than one significant color exists in the examined genotype.