ABSTRACT
Sunflower seeds, one of the main oilseeds produced around the world, are widely used in the food industry. Mixtures of seed varieties can occur throughout the supply chain. Intermediaries and the food industry need to identify the varieties to produce high-quality products. Considering that high oleic oilseed varieties are similar, a computer-based system to classify varieties could be useful to the food industry. The objective of our study is to examine the capacity of deep learning (DL) algorithms to classify sunflower seeds. An image acquisition system, with controlled lighting and a Nikon camera in a fixed position, was constructed to take photos of 6000 seeds of six sunflower seed varieties. Images were used to create datasets for training, validation, and testing of the system. A CNN AlexNet model was implemented to perform variety classification, specifically classifying from two to six varieties. The classification model reached an accuracy value of 100% for two classes and 89.5% for the six classes. These values can be considered acceptable, because the varieties classified are very similar, and they can hardly be classified with the naked eye. This result proves that DL algorithms can be useful for classifying high oleic sunflower seeds.
Subject(s)
Deep Learning , Helianthus , Algorithms , SeedsABSTRACT
Lung surfactant (LS) stabilizes the respiratory surface by forming a film at the alveolar air-liquid interface that reduces surface tension and minimizes the work of breathing. Typically, this surface-active agent has been isolated from animal lungs both for research and biomedical applications. However, these materials are constituted by complex membranous architectures including surface-active and inactive lipid/protein assemblies. In this work, we describe the composition, structure and surface activity of discrete membranous entities that are part of a LS preparation isolated from bronchoalveolar lavages of porcine lungs. Seven different fractions could be resolved from whole surfactant subjected to sucrose density gradient centrifugation. Detailed compositional characterization revealed differences in protein and cholesterol content but no distinct saturated:unsaturated phosphatidylcholine ratios. Moreover, no significant differences were detected regarding apparent hydration at the headgroup region of membranes, as reported by the probe Laurdan, and lipid chain mobility analysed by electron spin resonance (ESR) in spite of the variety of membranous assemblies observed by transmission electron microscopy. In addition, six of the seven separated LS subfractions formed similar, essentially disordered-like, interfacial films and performed efficient surface activity, under physiologically relevant conditions. Altogether, our work show that a LS isolated from porcine lungs is comprised by a heterogenous population of membranous assemblies lacking freshly secreted unused LS complexes sustaining highly dehydrated and ordered membranous assemblies as previously reported. We propose that surfactant subfractions may illustrate intermediates in sequential structural steps within the structural transformations occurring along the respiratory compression-expansion cycles.
