- Image Retrieval and Classification Techniques
- Digital Imaging for Blood Diseases
- Visual Attention and Saliency Detection
- Domain Adaptation and Few-Shot Learning
- Multimodal Machine Learning Applications
- Advanced Image Fusion Techniques
- Remote-Sensing Image Classification
- Image Processing Techniques and Applications
- Retinal Imaging and Analysis
- Natural Language Processing Techniques
- Cognitive Functions and Memory
- Mind wandering and attention
- Smart Agriculture and AI
- EEG and Brain-Computer Interfaces
- Interpreting and Communication in Healthcare
- Remote Sensing and Land Use
- Neural dynamics and brain function
- Adversarial Robustness in Machine Learning
- Explainable Artificial Intelligence (XAI)
- Advanced Graph Neural Networks
- Retinal Diseases and Treatments
- Visual perception and processing mechanisms
- Glaucoma and retinal disorders
- Spectroscopy and Chemometric Analyses
- Face recognition and analysis
Indian Institute of Technology Kharagpur
2020-2024
Of late, convolutional neural networks (CNNs) find great attention in hyperspectral image (HSI) classification since deep CNNs exhibit commendable performance for computer vision-related areas. have already proved to be very effective feature extractors, especially the of large data sets composed 2-D images. However, due existence noisy or correlated spectral bands domain and nonuniform pixels spatial neighborhood, HSI results are often degraded unacceptable. elementary CNN models intrinsic...
Deep learning‐based approaches have become very prominent in recent years due to its outstanding performance as compared the hand‐extracted feature‐based methods. Convolutional neural network (CNN) is a type of deep learning architecture deal with image/video data. Residual and squeeze excitation (SENet) are among developments CNN for image classification. However, SENet depends on operation done by global pooling, which sometimes may lead poor performance. In this study, authors propose...
Diabetic Retinopathy, also known as diabetic eye disease, is a complication that arises due to disease called Diabetes Mellitus. It damages blood vessels in the eyes high sugar levels causing progressive damage retina which can thereby cause blindness. The main objective develop an Automated System distinguish diseased from healthy ones based on High Resolution Fundus Image of Retina. A variety Processing techniques are applied includes Greyscale Conversion, Thresholding and Binarization....
Glaucoma is one of the major and critical eye diseases discovered till date. It actually a group that damage optic nerve subsequently result in vision loss blindness. One causes intrinsic distortion resulting high fluid pressure on front portion eye. The primary objective paper to classify High Resolution Fundus images retina into Glaucomatous Non-Glaucomatous. In order achieve that, DL-ML Hybrid Model has been developed with an initial image processing. overall methodology followed this...
The human brain possesses remarkable abilities in visual processing, including image recognition and scene summarization. Efforts have been made to understand the cognitive capacities of brain, but a comprehensive understanding underlying mechanisms still needs be discovered. Advancements decoding techniques led sophisticated approaches like fMRI-to-Image reconstruction, which has implications for neuroscience medical imaging. However, challenges persist fMRI-to-image such as incorporating...
Estimation of age any sample (non-living) or living, comes into play in multiple domains. For instance, archaeological research involves estimation the excavations. In case accidents and murder, if body victim is dismembered unrecognizable, then general investigations, rough individual (victim) very crucial from legible parts like hair hand. Also apart investigation, criminal cases, Forensic Science a important domain where evaluation done via Research which thorough study (body part)...
Abstract The visual ventral system needs to be better understood despite recent technological advancements. primary cortex processes edge-based information and contrast patterns in the human brain, which are heavily studied. However, nature of non-linearity involved processing still discovered. This study aims determine type cells involved. Initially, VOneNet computational model V1, among other models was selected for its performance interpretability. used verify shift-invariance property...