No abstract

Accurate disease classification in plants is important for a profound understanding of their growth and health. Recognizing diseases in plants from images is one of the critical and challenging problem in agriculture. In this research, a deep learning architecture model (CapPlant) is proposed that utilizes plant images to predict whether it is healthy or contain some disease. The prediction process does not require handcrafted features; rather, the representations are automatically extracted from input data sequence by architecture. Several convolutional layers are applied to extract and classify features accordingly. The last convolutional layer in CapPlant is replaced by state-of-the-art capsule layer to incorporate orientational and relative spatial relationship between different entities of a plant in an image to predict diseases more precisely. The proposed architecture is tested on the PlantVillage dataset, which contains more than 50,000 images of infected and healthy plants. Significant improvements in terms of prediction accuracy has been observed using the CapPlant model when compared with other plant disease classification models. The experimental results on the developed model have achieved an overall test accuracy of 93.01%, with F1 score of 93.07%.

Proteins are the core of all functions pertaining to living things. They consist of an extended amino acid chain folding into a three-dimensional shape that dictates their behavior. Currently, convolutional neural networks (CNNs) have been pivotal in predicting protein functions based on protein sequences. While it is a technology crucial to the niche, the computation cost and translational invariance associated with CNN make it impossible to detect spatial hierarchies between complex and simpler objects. Therefore, this research utilizes capsule networks to capture spatial information as opposed to CNNs. Since capsule networks focus on hierarchical links, they have a lot of potential for solving structural biology challenges. In comparison to the standard CNNs, our results exhibit an improvement in accuracy. Gene Ontology Capsule GAN (GOCAPGAN) achieved an F1 score of 82.6%, a precision score of 90.4% and recall score of 76.1%.