No abstract

One of the essential factors contributing to a plant's growth is identifying and preventing diseases in the early stages. Healthy plants are essential for a rich production. Recent advances in Deep learning - a subset of Artificial Intelligence and Machine Learning are playing a pivotal role in solving image classification problems and can be applied to the agricultural sector for crop surveillance and early anomaly identification. For this research, we used an open-source dataset of leaf images divided into three classes, two of which are the most common disease types found on many crops; the graphical characterizations for the three classes are images of leaves with Powdery Residue, images of leaves with Rusty Spots, and images of Healthy leaves. The primary objective of this research is to present a pre-trained ImageNet network architecture that is well suited for dealing with plant-based data, even when sample sizes collected are limited. We used different convolutional neural network-based architectures such as InceptionV3, MobileNetV2, Xception, VGG16, and VGG19 to classify plant leaf images with visually different representations of each disease. Xception, MobileNetV2, and DenseNet had a considerable advantage over all the performance metrics recorded among the other networks trained.

As the size and dimensionality of microarray datasets increase, it is vital to select essential features for data classification. Traditionally, ranking and selection measures are used to select the essential features from the high dimensional feature space. However, these measures are used to improve the data classification rate with limited number of instances and features space. Feature selection is one of the challenging issues for microarray datasets due to noise, sparsity and missing values. Traditional feature selection models such as Artificial Bee Colony (ABC), Particle Swarm Optimization (PSO) and Genetic Algorithms (GA) are used to select the highly weighted features for data classification. But these models require high computational memory and time for data classification. In this paper, a hybrid PSO+ABC based feature selection model is designed and implemented on microarray disease datasets. Proposed hybrid feature selection model is applied on multiple classification models to improve the true positive rate and error rate for different dimensional datasets. Experimental results are simulated on different microarray disease dataset and these results proved that the hybrid feature selection model has high true positive rate and minimal mean squared error rate compared to the traditional models.

Even though several methods are introduced to perform the heart beat classification by using the ECG signals, but classifying the heartbeat by analyzing the cardiac arrhythmia is still a challenging task. Hence, an effective RNN-based SM-BS optimization algorithm is introduced in this work, which is the combination of the SMO and the BSA algorithm. As the SMO and BSA are integrated together, the features from both the optimization will be utilized in the proposed work. As SMO uses the intelligent behavior and BSA mimics the foraging behavior, flight behavior, and vigilance behavior. Both these algorithms are effectively used to optimize the problem. Hence, this can be fused with the RNN to perform the beat classification in the proposed work.

No abstract