Discrimination of foreign bodies in quinoa (Chenopodium quinoa Willd.) grains using convolutional neural networks with a transfer learning approach.

The rising interest in quinoa (Chenopodium quinoa Willd.) is due to its high protein content and gluten-free condition; nonetheless, the presence of foreign bodies in quinoa processing facilities is an issue that must be addressed. As a result, convolutional neural networks have been adopted, mostly because of their data extraction capabilities, which had not been utilized before for this purpose. Consequently, the main objective of this work is to evaluate convolutional neural networks with a learning transfer for foreign bodies identification in quinoa samples. For experimentation, quinoa samples were collected and manually split into 17 classes: quinoa grains and 16 foreign bodies. Then, one thousand images were obtained from each class in RGB space and transformed into four different color spaces (L*a*b*, HSV, YCbCr, and Gray). Three convolutional neural networks (AlexNet, MobileNetv2, and DenseNet-201) were trained using the five color spaces, and the evaluation results were expressed in terms of accuracy and F-score. All the CNN approaches compared showed an F-score ranging from 98% to 99%; both color space and CNN structure were found to have significant effects on the F-score. Also, DenseNet-201 was the most robust architecture and, at the same time, the most time-consuming. These results evidence the capacity of CNN architectures to be used for the discrimination of foreign bodies in quinoa processing facilities.

Análisis cinemático de la masticación de una granola prototipo en contraste con la masticación de maní / Kinematic analysis of chewing by using a granola prototype and peanut

La masticación se ha estudiado desde diferentes puntos de vista, utilizando alimentos de prueba naturales y artificiales. La evidencia es escasa cuando se analizan alimentos a base de cereales, que van desde cereales para el desayuno hasta barras de granola. El investigar este tipo de alimentos, se vuelve importante para entender el comportamiento de la masticación frente a alimentos con diferentes composiciones y texturas, y como estas características pueden influir en el proceso masticatorio. Se analizó la masticación desde un punto de vista cinemático, en sujetos jóvenes dentados. El alimento de prueba utilizado fue granola prototipo y maní, este último se ha estudiado en sujetos con rehabilitación protésica y su consumo se recomienda en esta población. Se analizaron las características cinemáticas de la masticación como numero de ciclos, frecuencia masticatoria, velocidad de masticación de ascenso y descenso, y el área de masticación en los tres planos del espacio. Se relacionaron los movimientos masticatorios con los movimientos mandibulares bordeantes que conformaron el polígono de Posselt, este también se analizó en los tres planos espaciales. En todas las variables analizadas la granola presento valores mayores, excepto en el número de ciclos masticatorios, sólo se encontraron diferencias estadísticamente significativas (p= 0,03) al comparar la velocidad (ascenso y descenso) y el área de masticación en el plano horizontal.

Chewing has been studied from different points of view, using natural and artificial foods test. When analyzing cereal-based foods, from breakfast cereals to granola bars, the evidence is scarce. Investigate this type of food is important to understand the behavior of chewing, with foods of different compositions and textures, and how these characteristics can influence the chewing process. Chewing was analyzed from a cinematic point of view, in young subjects complete dental. The test food used was prototype granola and peanuts, last one has been studied in subjects with prosthetic rehabilitation and its consumption is recommended in this population. The kinematic characteristics of chewing were analyzed: number of cycles, chewing frequency, ascent and descent chewing speed, and the chewing area in the three planes of space. The masticatory movements were related to the bordering mandibular movements, that formed the Posselt polygon, which was also analyzed in the three spatial planes. In all the variables analyzed, granola showed higher values, except in the number of chewing cycles, only statistically significant differences (p = 0.03) were found when comparing speed (ascent and descent) and the chewing area in the horizontal plane.