A5102000256- Advanced Image Processing Techniques
- Advanced Neural Network Applications
- Surgical Simulation and Training
- Image and Signal Denoising Methods
- Artificial Intelligence in Healthcare and Education
- Multimodal Machine Learning Applications
- Coal Properties and Utilization
- Visual Attention and Saliency Detection
- Advanced Image Fusion Techniques
- Image Enhancement Techniques
- Emotion and Mood Recognition
- Image Processing Techniques and Applications
- Machine Learning and Data Classification
- Image and Video Quality Assessment
- Biomedical and Engineering Education
- Geomechanics and Mining Engineering
- Neural Networks and Applications
- Color perception and design
- Face and Expression Recognition
- Human Pose and Action Recognition
- Hand Gesture Recognition Systems
- Advanced Image and Video Retrieval Techniques
- Machine Learning and ELM
- Advanced X-ray and CT Imaging
- Coal and Its By-products
University of Plymouth
2024
Oxford Brookes University
2019-2023
Academy of Scientific and Innovative Research
2022
Central Institute of Mining and Fuel Research
2022
Thapar Institute of Engineering & Technology
2016-2019
Southern Illinois University Carbondale
2015
For an autonomous robotic system, monitoring surgeon actions and assisting the main during a procedure can be very challenging. The challenges come from peculiar structure of surgical scene, greater similarity in appearance performed via tools cavity compared to, say, human unconstrained environments, as well motion endoscopic camera. This paper presents ESAD, first large-scale dataset designed to tackle problem action detection minimally invasive surgery. ESAD aims at contributing increase...
The number of international benchmarking competitions is steadily increasing in various fields machine learning (ML) research and practice. So far, however, little known about the common practice as well bottlenecks faced by community tackling questions posed. To shed light on status quo algorithm development specific field biomedical imaging analysis, we designed an survey that was issued to all participants challenges conducted conjunction with IEEE ISBI 2021 MICCAI conferences (80 total)....
This paper presents ultra wide band low noise amplifier (LNA) with three stages for 1.9 to 8.2 GHz frequency based on CMOS technology. In this LNA, first stage is designed current reuse topology achieve higher gain, while second complementary push pull configuration in self biased state, which provides high gain and contribution the third common source stage. proposes two different designs. These designs are optimized set of parameters, makes them suitable categories applications. The design...
In this work, we take aim towards increasing the effectiveness of surgical assistant robots. We intended to make robots safer by making them aware about actions surgeon, so it can appropriate assisting actions. other words, solve problem surgeon action detection in endoscopic videos. To this, introduce a challenging dataset for real-world Action classes are picked based on feedback surgeons and annotated medical professional. Given video frame, draw bounding box around tool which is...
Every individual's perception of multimedia content varies based on their interpretation. Therefore, it is quite challenging to predict likability any just its content. This paper presents a novel system for analysis facial expressions subject against the be evaluated. First, we developed dataset by recording subjects under uncontrolled environment. These are volunteers recruited watch videos different genre, and provide feedback in terms likability. Subject responses divided into three...
Deep learning methods for the super-resolution problem are showing great performance compared to other traditional techniques. However, these unable learn complex spatial structures and high frequency details; which leads over-smooth results. In present paper, a novel Generative Adversarial Network based architecture named as Residue Semantic feature Dual Subpixel has been proposed generator discriminator networks solve problem. The network is residue semantic dual subpixel generative...
In this work, we take aim towards increasing the effectiveness of surgical assistant robots. We intended to make robots safer by making them aware about actions surgeon, so it can appropriate assisting actions. other words, solve problem surgeon action detection in endoscopic videos. To this, introduce a challenging dataset for real-world Action classes are picked based on feedback surgeons and annotated medical professional. Given video frame, draw bounding box around tool which is...
Advanced driver assistance systems (ADAS) have been developed to automate and modify vehicles for safety better driving experience. Among all computer vision modules in ADAS, 360-degree surround view generation of immediate surroundings the vehicle is very important, due application on-road traffic assistance, parking etc. This paper presents a novel algorithm fast computationally efficient transformation input fisheye images into required top down view. also generalized framework generating...
Image super resolution has gained a lot of attention due to its applications in different fields image processing. It is used produce high-resolution images from low-resolution input. Because the excellent learning capability convolution neural networks, these networks are able learn complex spatial structures for super-resolution. In this paper, two architectures have been proposed resolution. The first architecture Dual Subpixel Layer Convolution Neural Network (DSL-CNN), which stacks...
With the inclusion of camera in daily life, an automatic no reference image quality evaluation index is required for classification images. The present manuscripts proposes a new No Reference Regional Mutual Information based technique evaluating image. We use regional mutual information on subsets complete Proposed tested four benchmark natural databases, and one synthetic database. A comparative analysis with classical state-of-art methods indicate superiority high images comparable other...
Technological advancements in smart assistive technology enable navigating and manipulating various types of computer-aided devices the operating room through a contactless gesture interface. Understanding surgeon actions is crucial to natural human-robot interaction since it means sort prediction human behavior so that robot can foresee surgeon's intention, early choose appropriate action reduce waiting time. In this paper, we present new deep network based on Convolution Long Short-Term...
This paper summarizes the findings of a study carried out to explain sudden increases in permeability coal, commonly encountered northern San Juan basin, with continued production methane. Change coal was first measured laboratory using from basin and replicating ideal field conditions, that is, uniaxial strain. The results showed increased slowly continuously for depletion 1500 ~600 psi, at which time there large increase its value. After that, until pressure dropped ~50 psi. total excess...
International benchmarking competitions have become fundamental for the comparative performance assessment of image analysis methods. However, little attention has been given to investigating what can be learnt from these competitions. Do they really generate scientific progress? What are common and successful participation strategies? makes a solution superior competing method? To address this gap in literature, we performed multi-center study with all 80 that were conducted scope IEEE ISBI...
International benchmarking competitions have become fundamental for the comparative performance assessment of image analysis methods. However, little attention has been given to investigating what can be learnt from these competitions. Do they really generate scientific progress? What are common and successful participation strategies? makes a solution superior competing method? To address this gap in literature, we performed multicenter study with all 80 that were conducted scope IEEE ISBI...
In this manuscript neural networks architecture is used for image compression. We analyzed the PCA technique with help of in which synaptic weights act as principal components are trained through Generalized Hebbian Algorithm (GHA). A comparison traditional performed to demonstrate and illustrate training capabilities GHA
Deep learning methods for the super-resolution problem are showing great performance compared to other traditional techniques. However, these unable learn complex spatial structures and high frequency details; which leads over-smooth results. In present paper, a novel Generative Adversarial Network based architecture named as Residue Semantic feature Dual Subpixel has been proposed generator discriminator networks solve problem. The network is residue semantic dual subpixel generative...
Recently, there has been numerous breakthroughs in face hallucination tasks. However, the task remains rather challenging videos comparison to images due inherent consistency issues. The presence of extra temporal dimension video makes it non-trivial learn facial motion through out sequence. In order these fine spatio-temporal details, we propose a novel cross-modal audio-visual Video Face Hallucination Generative Adversarial Network (VFH-GAN). architecture exploits semantic correlation...