Digital images and in-person evaluation of Anthurium ‘Tropical’ postharvest quality

Abstract

Accurate quality classification is one of the requirements of the flower market, which is becoming increasing more sophisticated and demanding. However, there is a lack of information regarding how such classification is performed, and it is often conducted in a less empirical and more subjective way. Therefore, this work aimed to compare the efficiency of personal classification (in-person) and using digital images to determine the quality of anthurium cv. Tropical (Anthurium andraeanum Lind.). Three experiments were performed regarding the recording of the images and the display mode of the images and comparing three methods of classifying postharvest quality of anthurium. It was concluded that measurements of colorimetric parameters can be performed anywhere on the spathe, but the period in which the parameters are measured does affect. Evaluators differ in the thoroughness with which they assign grades, and there is no need for prior knowledge to assess quality using images. The L* parameter measured by the colorimeter presented higher correlation with the evaluation of postharvest quality of Anthurium andraeanum Lind. ‘Tropical’. Classification by images can be performed using different electronic devices, and in-person classification is more rigorous than classification conducted using digital images.

Digital Image Analysis Using FloCIA Software for Ornamental Sunflower Ray Floret Color Evaluation

“…Accurate phenotype classification is one of the requirements of the ornamental plant market, which is becoming increasingly more sophisticated and demanding (Guerra Mattos et al, 2020). The same stands for ornamental sunflower breeding and marketing which is expected to adopt automatic classification, in order to improve breeding efficiency and quality of products offered to consumers (Lino et al, 2011).…”

“…Based on image analysis algorithms, in-house created software, FloCIA automatically detected dominant colors of ray florets in six ornamental sunflower genotypes, thus offering novel methodology for objective and more precise ornamental sunflower phenotyping. In ornamental plants, digital image technology has been used for gerbera flower classification (Lino et al, 2011), rose shape analysis (Miao et al, 2006), bedding plant species quality assessment (Parsons et al, 2009), flower color pattern determination in Primula sieboldii E. Morren (Yoshioka et al, 2004), as well as evaluation of Anthurium 'Tropical' postharvest quality (Guerra Mattos et al, 2020). In sunflower, image analysis so far has been used for early detection of broomrape infection (Cochavi et al, 2017;Ortiz-Bustos et al, 2017;Lati et al, 2019), weed mapping (López-Granados et al, 2016), architecture-based organ segmentation (Gélard et al, 2016) and floral dimension determination (Sunoj et al, 2018).…”