A5024242809- Nuclear Physics and Applications
- Radiation Detection and Scintillator Technologies
- Human Pose and Action Recognition
- Neonatal and fetal brain pathology
- Welding Techniques and Residual Stresses
- Sentiment Analysis and Opinion Mining
- Hand Gesture Recognition Systems
- Retinal Imaging and Analysis
- Blind Source Separation Techniques
- Advanced Chemical Sensor Technologies
- Fetal and Pediatric Neurological Disorders
- Prenatal Screening and Diagnostics
- AI in cancer detection
- Functional Brain Connectivity Studies
- Dental Radiography and Imaging
- Image and Object Detection Techniques
- Advanced Image Processing Techniques
- EEG and Brain-Computer Interfaces
- Mediterranean and Iberian flora and fauna
- Seismology and Earthquake Studies
- Gait Recognition and Analysis
- Neural dynamics and brain function
- African Botany and Ecology Studies
- Retinal and Optic Conditions
- Advanced Neural Network Applications
Mohammed V University
2009-2023
Hassiba Benbouali University of Chlef
2020
Fetal biometry and amniotic fluid volume assessments are two essential yet repetitive tasks in fetal ultrasound screening scans, aiding the detection of potentially life-threatening conditions. However, these assessment methods can occasionally yield unreliable results. Advances deep learning have opened up new avenues for automated measurements ultrasound, demonstrating human-level performance various tasks. Nevertheless, majority studies retrospective silico studies, with a limited number...
Detection of defects in metallic pieces is an important application the field non-destructive testing (NDT), particularly industrial setting. These are mainly due to manufacturing errors or welding processes. In this article we will focus on second category using segmentation techniques applied welded joints. Segmentation one most difficult tasks image processing, case noisy low contrast images such as radiographic welds. segmenting type image, many researchers have used neural networks and...
P300 is one of the most studied and used event related potentials (ERP) in brain computer interfaces (BCI). The classical oddball paradigm usually to evoke from Electroencephalogram (EEG) signals. However, EEG raw data are noisy which make detection very difficult. In this paper, we aim detect wave as accurate possible using appropriate feature extraction method combined with powerful classifier. We compared four methods: Kernel principal component analysis (KPCA), Principal (PCA),...
X-ray radiography is one of the most used techniques in non destructive testing(NDT). It allows detection weld defects dangerous for weld's integrity. Because images welds are noisy and low contrasted, it difficult to detect inside. The goal this paper segment images. However, segmentation remains among tasks image processing, especially case or contrasted Many researchers neural networks, fuzzy logic methods SVM-based type results impressive; however they require a complex implementation...
Human Computer Interaction received a great deal of attention this last decade. Last researches has turned to more natural interaction systems like gestural human machine interfaces. Recent works are attempting solve the problem hand gestures recognition using learning methods. Some them pretending achieve very high performance. However, few taking into account mandatory requirements apply workflow model, mainly data unbalance, model selection and generalization performance metric choice. In...
<span>In this work we implemented STA/LTA trigger algorithm, which is widely used in seismic detection, using Hadoop MapReduce. This<br />implementation allows to find out how effective it type of tasks as well accelerate the detection process by reducing processing time. We tested our implementation on a seismological dataset 14 broadband stations and compare with traditional one. The results show that MapReduce decreased time 34% compared implementation.</span>
The advent of web 2.0 platforms allowed users to generate and share textual content. This results in an explosive increase online personal opinion. Sentiment Analysis, which is a recent field Natural Language Processing, aims predict the orientation sentiment present on this massive data. plays vital role many applications, such as recommender systems, customer intelligence, information retrieval psychological study crowd. Most existing approaches analysis trait only positive, negative...
Hand gesture recognition plays an important role in human computer interaction (HCI). Despite the recent progress, accuracy of up-to-date methods is still not satisfactory. In this work, we proposed a comparative study to recognize six hand gestures real time using Kinect sensor. First, developed tracking method scene where center palm detected depth data and projected into color image. Second, geometric features were extracted from image Histogram Oriented Gradients (HOG) descriptors...
Cervical cancer will cause 460 000 deaths per year by 2040, approximately 90% are Sub-Saharan African women. A constantly increasing incidence in Africa making cervical a priority the World Health Organization (WHO) terms of screening, diagnosis, and treatment. Conventionally, diagnosis relies primarily on histopathological assessment, deeply error-prone procedure requiring intelligent computer-aided systems as low-cost patient safety mechanisms but lack labeled data digital pathology limits...
We studied a colony of the Western Cattle Egret Bubulcus ibis located in public garden on northern outskirts Chlef, north-western Algeria, site with 13 tree species; this locality Egrets nested only Ficus elastica, Jacaranda mimosifolia and Pinus halepensis. observed birds, each two weeks during wintering period weekly breeding season (February to August). To count population, we used method described by Dragonetti & Giovacchini (2009), called perimeter counts. Egg nestling were recorded...
In this paper, we tackled the neutron-gamma discrimination at output of stilbene organic scintillation detector as a blind source separation problem. The estimation signal is performed using Second order NMF (Nonnegative Matrix Factorization) and NTF-2 Nonnegative Tensor methods. recovered independent components issued from application these two methods are then used to define qualitative criterion for ray's characterization. For enhancing performance characterization proposed Otsu's image...
In order to perform a fast and accurate neutron-gamma discrimination, we present in this paper method based on supervised unsupervised machine learning that is composed of the following steps. Firstly, apply nonnegative parallel factor analysis recover original sources from mixed signals recorded at output stilbene scintillator detector (45×45 mm). Secondly, spectral continuous wavelet transform used characterize these recovered sources. Thirdly, resulting time-scale representations are...
Semantic Segmentation is a pivotal undertaking within the domain of medical imaging analysis. It involves partitioning image into distinct regions interest. Many deep learning architectures has led to automation segmentation task, yielding performance levels close those experts. The U-Net architecture and many its variants, showcased their effectiveness for this task especially biomedical images. Recently, Segment Anything Model (SAM) been released showing impressive in segmenting natural...
Abstract Fetal biometry (FB) and amniotic fluid volume (AFV) assessments are two crucial yet repetitive tasks of fetal ultrasound screening scans that help detect potential life-threatening conditions, however, they suffer from reproducibility reliability issues. Advances in deep learning have led to new applications measurement automation ultrasound, showcasing human-level performances several tasks. However, most the studies performed retrospective “in silico” few include African patients...