Research
Welcome to our research page, where we showcase our diverse areas of focus in the fields of medical image analysis, computer vision, deep learning algorithms, machine learning, and randomized learning algorithms. Our goal is to contribute to the advancement of knowledge and innovation in these domains.
Here are the key research areas we are actively working on:
Medical Image Analysis
Brain MR Image Analysis.
Our research involves the development of advanced algorithms to analyze magnetic resonance (MR) images of the brain. We strive to improve techniques for tumor detection, segmentation, and classification, leading to more accurate diagnoses and treatment planning.
Sensorineural Hearing Loss Detection.
By leveraging medical image analysis techniques, we aim to detect and analyze patterns related to sensorineural hearing loss, contributing to early diagnosis and intervention.
Fundus Image Analysis.
We utilize image analysis methods to analyze fundus images, enabling the detection and characterization of various retinal diseases such as diabetic retinopathy and age-related macular degeneration.
Computer Vision
Large-scale Image and Video Classification.
Our research focuses on developing scalable algorithms for efficiently classifying large collections of images and videos. We aim to improve the accuracy and efficiency of classification tasks across diverse datasets.
Hyper-spectral Image Analysis.
Our research focuses on analyzing hyper-spectral images, which contain rich spectral information, for applications such as environmental monitoring, agriculture, and remote sensing. we develop deep learning algorithm which resolves some of the issues in HSI classification such as higher accuracy with lesser training data.
Action and Activity Recognition.
Our research focuses on developing scalable algorithms for efficiently classifying large collections of images and videos. We aim to improve the accuracy and efficiency of classification tasks across diverse datasets.
Image Captioning.
We explore techniques that automatically generate captions or textual descriptions for images, enabling better understanding and accessibility of visual content.
Object Detection.
Our research involves developing algorithms to detect and localize objects within images or video frames, with applications in autonomous driving, robotics, and surveillance systems.
Image Super-resolution.
We work on enhancing the resolution and quality of low-resolution images, enabling clearer visualization and analysis in various domains.
Deep Learning Algorithms
Autoencoder.
We investigate autoencoder architectures for unsupervised feature learning and dimensionality reduction, enabling efficient representation learning from large-scale datasets.
Convolutional Neural Networks (CNN) and Transformers
Randomized Learning Algorithm
Extreme Learning Machines(ELM).
Our research involves the study and utilization of ELM algorithms, which offer fast and efficient training in large-scale learning problems, facilitating real-time and online learning scenarios.
Random Vector Functional Link(RVFL).
We explore the capabilities of RVFL algorithms for supervised learning tasks, harnessing the power of randomization and non-linear mapping for improved prediction accuracy.
