A5079803939- Video Surveillance and Tracking Methods
- Neural Networks and Applications
- Advanced Vision and Imaging
- Anomaly Detection Techniques and Applications
- Image Enhancement Techniques
- Advanced Image and Video Retrieval Techniques
- Remote-Sensing Image Classification
- Image and Signal Denoising Methods
- Facility Location and Emergency Management
- Advanced Image Processing Techniques
- AI in cancer detection
- Network Security and Intrusion Detection
- Vehicle Routing Optimization Methods
- Face and Expression Recognition
- Advanced Neural Network Applications
- Digital Media Forensic Detection
- Image Retrieval and Classification Techniques
- Fire Detection and Safety Systems
- Advanced Clustering Algorithms Research
- Cutaneous Melanoma Detection and Management
- Student Assessment and Feedback
- Advanced Data Compression Techniques
- Gait Recognition and Analysis
- Industrial Vision Systems and Defect Detection
- Human Pose and Action Recognition
Universidad de Málaga
2015-2024
Software (Spain)
2024
Instituto de Investigación Biomédica de Málaga
2020-2024
The University of Texas at Austin
2024
University of Michigan
2019
Universidad Autónoma de Zacatecas "Francisco García Salinas"
2005
Deep Learning (DL), a groundbreaking branch of Machine (ML), has emerged as driving force in both theoretical and applied Artificial Intelligence (AI). DL algorithms, rooted complex non-linear artificial neural systems, excel at extracting high-level features from data. demonstrated human-level performance real-world tasks, including clinical diagnostics, unlocked solutions to previously intractable problems virtual agent design, robotics, genomics, neuroimaging, computer vision, industrial...
Skin lesions are caused due to multiple factors, like allergies, infections, exposition the sun, etc. These skin diseases have become a challenge in medical diagnosis visual similarities, where image classification is an essential task achieve adequate diagnostic of different lesions. Melanoma one best-known types vast majority cancer deaths. In this work, we propose ensemble improved convolutional neural networks combined with test-time regularly spaced shifting technique for lesion...
The rise of surveillance systems has led to exponential growth in collected data, enabling several advances Deep Learning exploit them and automate tasks for autonomous systems. Vehicle detection is a crucial task the fields Intelligent Systems Transport systems, making it possible control traffic density or detect accidents potential risks. This paper presents an optimal meta-method that can be applied any instant segmentation model, such as Mask R-CNN YOLACT++. Using initial detections...
Background modeling and foreground detection are key parts of any computer vision system. These problems have been addressed in literature with several probabilistic approaches based on mixture models. Here we propose a new kind background models which is self-organising maps. This way, the pixels modeled more flexibility. On other hand, statistical correlation measure used to test similarity among nearby pixels, so as enhance performance by providing feedback process. Several well known...
Image segmentation is a common goal in many medical applications, as its use can improve diagnostic capability and outcome prediction. In order to assess the wound healing rate diabetic foot ulcers, some parameters from area are measured. However, heterogeneity of skin lesions noise present images captured by digital cameras make extraction difficult task. this work, Deep Learning based method for accurate regions proposed. proposed method, input first processed remove artifacts then fed...
The design of automated video surveillance systems often involves the detection agents which exhibit anomalous or dangerous behavior in scene under analysis. Models aimed to enhance pattern recognition abilities system are commonly integrated order increase its performance. Deep learning neural networks found among most popular models employed for this purpose. Nevertheless, large computational demands deep mean that exhaustive scans full frame make perform rather poorly terms execution...
One of the most important challenges in computer vision applications is background modeling, especially when dynamic and input distribution might not be stationary, i.e. data could change with time (e.g. changing illuminations, waving trees, water, etc.). In this work, an unsupervised learning neural network proposed which able to cope progressive changes distribution. It based on a dual mechanism manages separately from cluster detection. The proposal adequate for scenes where varies...
Human activity recognition is an application of machine learning with the aim identifying activities from gathered raw data acquired by different sensors. In medicine, human gait commonly analyzed doctors to detect abnormalities and determine possible treatments for patient. Monitoring patient’s paramount in evaluating treatment’s evolution. This type classification still not enough precise, which may lead unfavorable reactions responses. A novel methodology that reduces complexity...
Small robots have numerous interesting applications in domains like industry, education, scientific research, and services. For most vision is important, however, the limitations of computing hardware make this a challenging task. In paper, we address problem real-time object recognition propose Fast Regions Interest Search (FROIS) algorithm to quickly find ROIs objects small with low-performance hardware. Subsequently, use two methods analyze ROIs. First, develop Convolutional Neural...
Growing hierarchical self-organizing models are characterized by the flexibility of their structure, which can easily accommodate for complex input datasets. However, most proposals use Euclidean distance as only error measure. Here we propose a way to introduce Bregman divergences in these models, is based on stochastic approximation principles, so that more general distortion measures be employed. A procedure derived compare performance networks using different divergences. Moreover,...
Anomaly detection in sequences is a complex problem security and surveillance. With the exponential growth of surveillance cameras urban roads, automating them to analyze data automatically identify anomalous events efficiently essential. This paper presents methodology detect using pre-trained convolutional neural networks (CNN) super-resolution (SR) models. The proposal composed two parts. In offline stage, CNN model evaluated large dataset establish common locations elements interest....
Abstract In the area of medical imaging, one factors that can negatively influence performance prediction algorithms is limited number observations for each class within a labeled dataset. Usually, in order to increase samples, second set unlabeled images used. However, this adds two new problems (i) finding patient with different pathologies than those observed data and (ii) belonging distribution from dataset used model training process. This way, merging datasets sources have an adverse...