Background and objectives
Most puffed snacks in the market are made from refined cereal flours which allow greater expansion and better texture but are nutritionally inferior as they lack protein and dietary fiber. Whole barley and green lentil flours at several blending ratios were extruded as a function of temperature and moisture content to optimize the physical and microstructural quality of fiber and protein‐enriched snacks.
Findings
High extrusion temperature significantly improved overall expansion and texture. The effects of feed moisture depended on the blending ratio, parallel with the total protein and dietary fiber content of extrudates. Barley:green lentil of 45:55 showed the highest extrudate expansion (~1.9 mm/mm on average) and lowest hardness (~29 N on average), followed by the blend 60:40. X‐ray microtomography showed that this blending ratio also produced a larger mean cell size (~1.6 mm), lower mean wall thickness (~0.3 mm), and higher overall connectivity between cells.
Conclusion
Barley and green lentil when blended at the ratios of 45:55 and 60:40, and extruded at higher temperature resulted in optimal extrudate physical and microstructural properties including higher expansion and crispness, thinner cell walls, reduced hardness, and crunchiness.
Significance and novelty
Blending cereal and pulse flours in snack food applications will allow development of fiber and protein‐enriched options that are also texturally and structurally appealing.
Applications of deep-learning models in machine visions for crop/weed identification have remarkably upgraded the authenticity of precise weed management. However, compelling data are required to obtain the desired result from this highly data-driven operation. This study aims to curtail the effort needed to prepare very large image datasets by creating artificial images of maize (Zea mays) and four common weeds (i.e., Charlock, Fat Hen, Shepherd’s Purse, and small-flowered Cranesbill) through conditional Generative Adversarial Networks (cGANs). The fidelity of these synthetic images was tested through t-distributed stochastic neighbor embedding (t-SNE) visualization plots of real and artificial images of each class. The reliability of this method as a data augmentation technique was validated through classification results based on the transfer learning of a pre-defined convolutional neural network (CNN) architecture—the AlexNet; the feature extraction method came from the deepest pooling layer of the same network. Machine learning models based on a support vector machine (SVM) and linear discriminant analysis (LDA) were trained using these feature vectors. The F1 scores of the transfer learning model increased from 0.97 to 0.99, when additionally supported by an artificial dataset. Similarly, in the case of the feature extraction technique, the classification F1-scores increased from 0.93 to 0.96 for SVM and from 0.94 to 0.96 for the LDA model. The results show that image augmentation using generative adversarial networks (GANs) can improve the performance of crop/weed classification models with the added advantage of reduced time and manpower. Furthermore, it has demonstrated that generative networks could be a great tool for deep-learning applications in agriculture.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.