A5020330356- Video Surveillance and Tracking Methods
- Fire Detection and Safety Systems
- Remote Sensing and LiDAR Applications
- AI in cancer detection
- Advanced Neural Network Applications
- Gastric Cancer Management and Outcomes
- Forest ecology and management
- Advanced Image and Video Retrieval Techniques
- Colorectal Cancer Screening and Detection
- Fire effects on ecosystems
- Radiomics and Machine Learning in Medical Imaging
- Remote Sensing in Agriculture
- Image Enhancement Techniques
- Human Pose and Action Recognition
- Graph Theory and Algorithms
- Wood and Agarwood Research
- Cancer Genomics and Diagnostics
- Visual Attention and Saliency Detection
- Anomaly Detection Techniques and Applications
- Cell Image Analysis Techniques
- Gait Recognition and Analysis
- ECG Monitoring and Analysis
- Genetic factors in colorectal cancer
- Forest Biomass Utilization and Management
- Artificial Intelligence in Healthcare
Imperial College London
2018-2024
University College London
2021-2024
CRUK Lung Cancer Centre of Excellence
2021-2024
Information Technologies Institute
2013-2023
Centre for Research and Technology Hellas
2013-2023
Aristotle University of Thessaloniki
2015-2017
Information Technology Institute
2014
Every year, a large number of wildfires all over the world burn forested lands, causing adverse ecological, economic, and social impacts. Beyond taking precautionary measures, early warning immediate response are only ways to avoid great losses. To this end, in paper we propose computer vision approach for fire-flame detection be used by an early-warning fire monitoring system. Initially, candidate regions frame defined using background subtraction color analysis based on nonparametric...
In this paper, we consider the problem of multi-dimensional dynamic texture analysis, and introduce a new higher order linear dynamical system (h-LDS) descriptor. The proposed h-LDS descriptor is based on decomposition multidimensional image data enables analysis textures by using information from various elements. addition, propose methodology for its application to video-based early warning systems that focus smoke identification. More specifically, representation video subsequences as...
In this paper, we propose a novel image-based fire detection approach, which combines the power of modern deep learning networks with multidimensional texture analysis based on higher-order linear dynamical systems. The candidate regions are identified by Faster R-CNN network trained for task using set annotated images containing actual as well selected negatives. projected to Grassmannian space and each image is represented cloud points manifold. Finally, vector representation approach...
In this paper we address the problem of automated grading invasive breast carcinoma through encoding histological images as VLAD (Vector Locally Aggregated Descriptors) representations on Grassmann manifold. The proposed method considers each image a set multidimensional spatially-evolving signals that can be efficiently modeled higher-order linear dynamical systems analysis. Subsequently, H&E (Hematoxylin and Eosin) stained cancer is represented cloud points manifold, while vector...
The environmental challenges the world faces have never been greater or more complex. Global areas that are covered by forests and urban woodlands threatened large-scale forest fires increased dramatically during last decades in Europe worldwide, terms of both frequency magnitude. To this end, rapid advances remote sensing systems including ground-based, unmanned aerial vehicle-based satellite-based adopted for effective fire surveillance. In paper, recently introduced 360-degree sensor...
Tertiary lymphoid structures (TLS) are ectopic aggregates of cells in inflamed, infected, or tumoral tissues that easily recognized on an H&E histology slide as discrete entities, distinct from lymphocytes. TLS associated with improved cancer prognosis but there is no standardised method available to quantify their presence. Previous studies have used immunohistochemistry determine the presence specific a marker TLS. This has now been proven be underestimate true number Thus, we propose...
In the current context of climate change and demographic expansion, one phenomena that humanity faces are suburban wildfires. To prevent occurrence forest fires, fire risk assessment early detection approaches need to be applied. Forest mapping depends on various factors contributes identification monitoring vulnerable zones where most severe. Therefore, watchtowers, sensors, base stations autonomous unmanned aerial vehicles placed carefully in order ensure adequate visibility or battery...
Abstract Mismatch repair (MMR)-deficient cancer evolves through the stepwise erosion of coding homopolymers in target genes. Curiously, MMR genes MutS homolog 6 ( MSH6) and 3 MSH3 ) also contain homopolymers, these are frequent mutational targets MMR-deficient cancers. The impact incremental mutations on evolution is unknown. Here we show that microsatellite instability modulates DNA by toggling hypermutable mononucleotide homopolymer runs MSH6 stochastic frameshift switching. Spontaneous...
In this paper, we address the problem of classifying multidimensional time-evolving data in dynamic scenes. To take advantage correlation between different channels data, introduce a generalized form stabilized higher order linear dynamical system (sh-LDS) and represent signal as third-order tensor. addition, show that parameters proposed model lie on Grassmann manifold attempt to classification through study geometric properties sh-LDS's space. Moreover, tackle nonlinearity observation each...
In this research, the calorific value and ash content of wood bark some fast-growing hardwood species, such as tree-of-heaven, (Ailanthus altissima (Miller) Swingle), empress tree (Paulownia tomentosa (Thunberg) Steudel), trembling aspen (Populus tremuloides Michaux), oriental plane (Platanus orientalis Linnaeus) black locust (Robinia pseudoacacia were investigated in order to comprehend their behaviour during combustion estimate utilization potential solid biofuels (pellets).Beech (Fagus...
Gastric cancer is one of the most frequent causes cancer-related deaths worldwide. atrophy (GA) and gastric intestinal metaplasia (IM) mucosa stomach have been found to increase risk are considered precancerous lesions. Therefore, early detection GA IM may a valuable role in histopathological assessment. However, difficult confirm endoscopically and, following Sydney protocol, their diagnosis depends on analysis glandular morphology identification at least well-defined goblet cell set...
In this paper a new algorithm is proposed for detecting fire from video data in real time based on combination of features, including spatio-temporal consistency feature. A significant challenge flame detection systems to discriminate between actual and false alarms caused by colored objects. Towards aim, we propose an that has advantages: a) it fast since rectangular non-overlapping blocks are used as basic elements (instead arbitrary-shaped regions) b) feature addition color probability,...
Invasive insect pests and fungi, which are introduced accidentally to forests affect tree growth survival, constitute a serious threat for the trees acting on climate change its impacts. Thus, need early accurate health determination process of forest regions, has significantly increased interest in automatic monitoring methods. In this paper, order overcome great variety trees' characteristics forests' heterogeneity that affects diversity their color texture making detection diseases...
Purpose The aim of this study is to evaluate the predictive ability goodwill and other intangible assets on forecasting corporate profitability. Subsequently, compares efficiency deep learning model that machine models such as random forest (RF) support vector (SVM) well traditional statistical methods linear regression model. Design/methodology/approach Studies confirm intangibles are valuable give companies a competitive advantage increase profitability shareholders’ returns. Thus, by...
Salient object detection has achieved great improvements by using the Fully Convolutional Networks (FCNs). However, FCN-based U-shape architecture may cause dilution problems in high-level semantic information during up-sample operations top-down pathway. Thus, it can weaken ability of salient localization and produce degraded boundaries. To this end, order to overcome limitation, we propose a novel pyramid self-attention module (PSAM) adoption an independent feature-complementing strategy....
Given the urgent priority around protecting forests and limiting impacts of climate change, constant monitoring towards achievement accurate timely detection infestations catastrophic action invasive insects, pests fungi is an important challenging task. More precisely, new species insects that are introduced or already existing insect whose population multiply uncontrollably into forest area, affect tree growth, their survival, as well quality biomass constitute a serious threat to...
Purpose The purpose of this study is to investigate the value relevance goodwill and its additional aspects during a long-term period in Greece. Furthermore, by implementing two most popular models, Ohlson’s price Easton Harris’ return model, examines impact on Greek stock prices from 2007 2018, 12 years which International Financial Reporting Standards (IFRS) are applied. analyzes how goodwill’s changes as it ages debt crisis. Design/methodology/approach In order test we implemented model....
The use of agricultural biomass to produce biofuels and energy can provide many environmental socio-economic benefits. This research project examines the possibility replacing part wood material in a pellet with various proportions residues crops such as medic, maize, wheat bran, straw, sunflower, cardoon. Such substitution would contribute recycling materials sustainable other natural resources. It reduce emissions gaseous pollutants by fossil fuels solid biofuels. chosen species, well...