Search

In hilly areas, landslides can occur due to natural factors such as heavy rainfall, earthquakes, moisture in soil, or man-made factors like unplanned constructions. Landslides can be disastrous leading to a huge loss of property and lives which can be avoided using automatic prediction. Recently, machine learning algorithms have been applied to automatically identify landslides. Numerous feature extraction and classification-based approaches have been implemented on satellite images for semiautomatic detection and prediction of landslides. However, limited research has been done on fully automatic detection with acceptable accuracy. The most challenging task in the classification and prediction of landslides from satellite images is to find an appropriate database for training and yield highly accurate testing results. The primary agenda of a comprehensive study of various techniques used for the detection and classification of landslides using satellite images is to identify the research gap. The secondary objective aims to propose a prototype of novel approach for the same task. Fifty papers based on machine learning and deep learning algorithms from reputed journals are considered for analysis. This article summarizes the performances of different classification techniques from recent literature followed by comparison and discussion with respect to accuracy. Based on the gap identified an effective prototype of the landslide classification approach is proposed. A slightly modified version of the deep learning model ResNet101 is proposed which yields an accuracy of 96.88% when tested on an augmented Beijing dataset of 770 satellite images. The article also offers the researchers the latest status, overview, and potential avenues of machine and deep learning algorithms for landslide detection. The techniques discussed will serve as a valuable resource for identifying research gaps, guiding new researchers, and fostering innovative exploration in the field of landslide classification using satellite images.

Active Noise Control (ANC) is an effective technique for removing undesirable disturbances based on destructive interference between two noises (i.e., the superposition principle). To reduce the Non-Gaussian distribution of impulsive noises, the ANC is implemented using a prominent Filtered Cross Least Mean Square (FxLMS) method that relies on reducing the unwanted noise. The standard FxLMS method fails to adapt to its specifications, resulting in poor convergence and instability in the presence of impulsive noises and a non-linear response from the ANC system’s components. The Least Square family of Recursive Least Square (RLS) increases ANC performance by offering superior convergence performance to traditional stochastic algorithms. This paper proposes a novel technique called the recursive non-linear active threshold-based and modified gain FXRLS (NAMGFxRLS) algorithm to overcome the inadequacies of impulsive noise and non-linearity issues in the ANC. The suggested technique aims to automatically modify weights by the various sample processes, i.e., employing the FxRLS algorithm’s updated gain to adjust the error and reference signals and appropriately deploying the threshold. The potential of the suggested strategy is proved by simulated results of convergence speed, stability, and excellent Mean Noise Reduction of roughly 52.8 % for various noises when compared to previous approaches, notably large impulsive noises.

Back pain is an issue affecting millions of people throughout the world. Research on back pain root cause detection is immense. The lumbar spine is a lower back region of the backbone that could also be responsible for back pain. Research on issues related to lumbar spine disease (LSD) is limited. The lumbar spine is critical for the human body as it supports the weight of the body. Many diseases can affect the lumbar spine adversely. This research is directed towards the detection of LSDs using optimized feature extraction and selection phases. Furthermore, the linearity-based model is used for feature selection, selecting only the best possible features to reduce the missed classification degree. The flow of the proposed research work is divided into phases. In the first phase, data collection is performed. Data collection in the proposed work includes both real-time and benchmark magnetic resonance imaging (MRI) datasets. The MRI dataset collected from the hospital is validated by medical experts. After data collection, pre-processing is applied. This step removes the noise from the collected dataset. The pre-processing mechanism is performed using histogram equalization, median filtering, validation, and normalization. After the pre-processing phase, background subtraction and region of interest (ROI) detection are performed using the region-cut mechanism. Optimal feature extraction is achieved using a differential spider monkey optimization (SMO), and feature selection is performed using a linearity-based convolutional neural network (CNN) model. In the end, ensemble-based classification is used for disease prediction. The validation of the result is conducted through classification accuracy, specificity, sensitivity, and F-Score. The high classification accuracy of 96% is achieved with multi support vector machine (MSVM), 94% with random forest (RF), 93.5% with a decision tree (DT), and 91% with the Naïve Bayes (NB) approach, proving the validity of the proposed approach.

Brain tumour (BT) detection involves the process of identifying the presence of a brain tumour in medical imaging, such as MRI scans. BT detection often relies on medical imaging techniques, such as MRI (Magnetic Resonance Imaging), CT (Computed Tomography), or PET (Positron Emission Tomography) scans. Early detection of BT is important and MRI is one of the primary imaging techniques used to diagnose and assess BT. Deep learning (DL) techniques, particularly convolutional neural networks (CNNs) have shown promising results in assisting with BT detection on MRI scans. This study designs an Ensemble Learning Driven Computer-Aided Diagnosis Model for Brain Tumor Classification (ELCAD-BTC) technique on MRIs. The presented system purposes to detect and classify various steps of BTs. The presented system contains a Gabor filtering (GF) approach to remove the noise and increase the quality of MRI images. Moreover, ensemble learning of three DL models namely EfficientNet, DenseNet, and MobileNet is utilized as feature extractors. Furthermore, the denoising autoencoder (DAE) approach can be exploited to detect the presence of BTs. Finally, a social spider optimization algorithm (SSOA) was carried out for the hyperparameter tuning of the DL models. For simulating the improved BT classification outcome, a brief set of simulations occur on BRATS 2015 database.

The augmentation of hyperspectral images requires the design of high-density feature analysis & band-fusion models that can generate multimodal imagery from limited information sets. The feature analysis models use deep learning operations to maximize inter-class variance while minimizing inter-class variance levels for efficient classification operations. When combined with intelligent band-fusion methods, such models allow the augmentation model to enhance its classification efficiency under different use cases. Existing band-fusion-based augmentation models for hyperspectral images do not incorporate continuous efficiency enhancements and showcase higher complexity levels. Furthermore, these models can’t be scaled for more varied use cases because their use is restricted to specific image types. To overcome these issues, we designed a novel multimodal hybrid bioinspired model for the augmentation of hyperspectral imagery via iterative learning for continuous efficiency enhancements. The proposed model initially represents input images into Fourier, Laplacian, Cosine, multimodal Wavelet, Mellin, and Z-Transform domains, which will assist in describing the images in multimodal domains. These transformed image sets are passed through a convolutional filter to extract windowed feature sets. A Grey Wolf Optimizer (GWO) is used to identify high inter-class variance features from the extracted image sets, which assists in selecting transformed images that can help improve hyperspectral classification performance. The selected hyperspectral images are fused via a Bacterial Foraging Optimization (BFO) model, which assists in reducing intra-class variance levels. The final set of selected images is intelligently augmented via Particle Swarm Optimization (PSO), which performs rotation, zooming, shifting, and brightness variation operations selectively. These augmented images are classified via a customized VGGNet-19-based Convolutional Neural Network (CNN) classifier that assists in continuously estimating accuracy levels for different application scenarios. Based on these accuracy levels, the model is reconfigured via hyperparameter tuning of GWO and PSO optimizers. Due to combining these models and incremental accuracy optimizations, the proposed model has improved its hyperspectral classification accuracy by 10.6% and precision by 10.4%, as compared to standard deep learning-based augmentation techniques.

The growth in sustainable generation technology such as fuel cell, wind energy conversion system, photovoltaic system, increase in fuel cost, energy necessity and the reduction in the fossil fuel reserve, for better power quality and reliability, is obliging the power sector to use the renewable based energy sources. In India, wind energy is gradually becoming an important and significant energy resource. Keeping in opinion the aforementioned wind energy prediction is becoming an essential study for harnessing the wind energy prospective. This paper proposes an effective technique based on intelligent approach for predicting wind power in different areas. This technique is based on using an intelligent model concerning the predicted gap to its similar one and two year old data. There are many intelligent and conventional models existed in literature for the wind power prediction like support vector machines (SVM), back propagation (BP) prediction etc. In this paper an effective fuzzy logic and model predictive control based models have been developed and offered for the wind power prediction for microgrid application by using air density and wind speed as the input parameters for fuzzy system. The outcomes are compared with the computed data and existing models and it can be observed that the different errors are found within the permissible limits. The outcomes obtained from fuzzy based technique are very close to calculated values if compared with model predictive based technique. Hence, the proposed models can be employed for the prediction of wind speed and wind power generation in the selected stations. The existing models results are compared with Kolkata city outcomes. The Error RMSE with Support vector machine, Back propagation, Model of forecast error correction +SVM and Model of forecast errors correction +BP, Neural Network method, model predictive based system, and proposed fuzzy logic based system are 30.48%, 32.83%, 26.81%, 28.58%, 1.1431%, 1.38% and 1.12% respectively. Therefore, the proposed techniques provide the best results and even these are observed within the suitable limits. Additionally, the achieved outcomes can be used for Microgrid/SmartGrid applications.

Breast cancer is one the most critical disease and suffered many people around the world. The efficient and correct detection of breast cancer is still needed to ensure this medical issue although the researchers around the world are proposed different diagnostic methods for detection of this disease, however these existing methods still needed further improvement to correct and efficient detection of this disease. In this study, we proposed a new breast cancer identification method by using machine learning algorithms and clinical data. In the proposed method supervised (Relief algorithm) and unsupervised (Autoencoder, PCA algorithms) techniques have been used for related features selection from data set and then these selected features have been used for training and testing of classifier support vector machine for accurate and on time detection of breast cancer. Additionally, in the proposed approach k fold cross validation method has been used for model validation and best hyperparameters selection. The model performance evaluation metrics have been used for model performance evaluation. The BC data sets have been used for testing of the proposed method. The analysis of experimental results has been demonstrated that the features selected by Relief algorithm are more related for accurate detection of Breast cancer instead of features selected by Auotencoder and PCA algorithms. The proposed method has been attained high results in terms of accuracy on selected feature selected by Relief algorithm and achieved 99.91% accuracy. We have been employed McNemar's statistical test for performance comparison of our different models. Further, the proposed method performance has been compared with baseline methods in the literature and the proposed method performance is high as compared to base line methods. Due to the high performance of the proposed method (Relief-Support vector machine) we highly recommended it for the diagnosis of breast cancer. In addition, the proposed method can be easily incorporated into the healthcare system for reliable diagnosis of Breast cancer.