The envelope of Discrete Cosine Transform (DCT) coefficients exhibits a faster decay rate in a PVC beat than that of a regular ECG beat. This algorithm clearly distinguishes PVC patterns under noisy conditions

VMD-based spectral and entropy features are used to analyse and classify Amyotrophic lateral sclerosis (ALS), myopathy, and normal electromyogram (EMG) signals. VMD-based statistical feature for classifying physical actions using surface EMG signals These features are fed into the multiclass least squares support vector machine (MC-LS-SVM) classifier with the radial basis function to classify normal physical actions of surface EMG signals

A novel dataset consisting of road images for Indian road scenarios has been developed with various environmental conditions. Developed a robust algorithm for ego-lane detection with straight and curved lanes for adverse environmental conditions Generalized ego-lane detection and road marking detection for the Indian highway road. Developing a lane departure warning system based on the detected lane information. Hardware implementation of lane detection and departure warning system is developed. Two-dimensional Sliding discrete Fourier transform (2D-SDFT) kernels-based cellular neural network (CNN) for edge and lane features extraction. The feature-based lane detection algorithm is proposed and tested on the Caltech lane dataset. Design of driver state monitoring and assistance system using physiological, physical, and vision sensors Time-Frequency based continuous wave transform-synchrosqueezing transform (CWT-SST) proposed for detection of driver drowsiness state and stress level. Modified a road lane detection model that is capable of segmenting an image into the different road lanes present Used FCN architecture and pre-trained VGG-16 model without top layers for semantic segmentation