Performance of deep learning vs machine learning in plant leaf disease detection

Hyperspectral imaging has been rapidly developing over the past decade, and modern sensor technologies can cover large areas with exceptional spatial, spectral, and temporal resolutions. Due to these features, hyperspectral imaging is used effectively in numerous remote sensing applications such as precision agriculture, environmental monitoring, food analysis, and military applications requiring estimation of physical parameters of many complex surfaces and identifying visually similar materials with acceptable spectral signatures. The scope of fusion of the two images, one with high spatial content and the other with high spectral content, is to estimate one image with high spatial and spectral content. This paper presents a brief review of recent image resolution fusion algorithms, including deep learning techniques, for hyperspectral images. The need of high resolution panchromatic (pan) and multispectral (MS) images, lossless registration of images from multiple sources and point spread function (PSF) impose limitations for performing pan sharpening process. Hence the fusion is achieved using multispectral images instead of panchromatic images on hyperspectral images. It is essential to identify and reduce uncertainties in the image processing chain to improve image fusion enhancement. This paper also presents the current practices, problems, and prospects of hyperspectral image fusion. In addition, some important issues affecting fusion performance are discussed.

Hyperspectral and multispectral image fusion techniques for high resolution applications: a review

Airlines operate in a competitive marketplace and must upgrade their services to meet customer safety and comfort. Post-pandemic, the government and airlines resumed flights with many restrictions, the impact which is unexplored. An increasing number of customers use social media to leave reviews and in this age of Machine Learning (ML), if a model is available to automatically polarize flyer sentiments, it can help airlines upscale. In this work, a custom dataset is scraped from Twitter by including online reviews of five Indian airlines. Multiclass sentiment analysis using three classifiers, support vector machine, K-nearest neighbor and random forest with word2vec and TF-IDF word embeddings is implemented. AirBERT, a fine-tuned deep learning attention model based on bidirectional encoder representation from transformers is proposed. From results, it is observed that on ML, Random Forest with TF-IDF performs the best but the graphical processing unit and domain corpora trained AirBERT outperforms all the other models with an accuracy of 91%. Indigo airlines and Jet Airways received the maximum percentage of positive and negative reviews respectively. In performance comparison with three existing models on the USA airlines tweets dataset, the proposed model outperforms others trained on general domain corpora and matches state-of-the-art TweetBERTv2 model accuracy. The model can be deployed by airlines and other service industries to implement a customer relationship management (CRM) system.

AirBERT: A fine-tuned language representation model for airlines tweet sentiment analysis

Cloud Cryptography for Cloud Data Analytics in IoT

Comparative analysis of CN2 rule induction with other classification algorithms for network security

Forests play a vital role in the regulation of climate, absorption of carbon dioxide, hydrological cycle, conservation of water, soil and biodiversity and help mitigate natural disasters. With the help of various remote sensors, high-resolution satellite images are being collected nowadays, which helps in tackling the global challenges of forest mapping in remote areas. Each landscape will grow different types of trees and in turn substantiate a part of the country’s economy. This paper uses visualization and machine learning (ML) processes to classify the forest land on the terrain dataset composed of the advanced spaceborne thermal emission and reflection radiometer (ASTER) imaging instrument to get the insight of the cumulated data by using Box Plot and Heat Map. The accuracy obtained by utilizing different machine learning techniques like Support Vector Machine (SVM) gives 95.4%, Logistic Regression (LR) gives 94.5%, K-Nearest Neighbor (K-NN) gives 93.7%, Decision Tree (DT) with 89.5%, Stochastic Gradient Descendent (SGD) with 92.4% and CN2 Rule Induction (RI) gives 85.3% are allied which gives appreciable results in forest mapping substantiated the same with confusion matrix and ROC. We also obtained the DT and rules for the considered dataset.

Sujatha Radhakrishnan

Papers

Forest data visualization and land mapping using support vector machines and decision trees