Daniel Hernández-Sosa

Universidad de Las Palmas de Gran Canaria | ULPGC · Department of Computer Sciences and Systems

We present a transfer learning analysis on a sporting environment of the expanded 3D (X3D) neural networks. Inspired by action quality assessment methods in the literature, our method uses an action recognition network to estimate athletes' cumulative race time (CRT) during an ultra-distance competition. We evaluate the performance considering the...

The uncontrolled characteristics of long-term scenarios, like ultra-running competitions, are challenging for person re-identification approaches based on computer vision methods. State-of-the-art techniques have reported hardly moderate success for whole-body runner re-identification due to the existence of distinct illumination conditions, as wel...

An article published on Medical News Today in June 2022 presented a fundamental question in its title: Can an earlobe crease predict heart attacks? The author explained that end arteries supply the heart and ears. In other words, if they lose blood supply, no other arteries can take over, resulting in tissue damage. Consequently, some earlobes have...

The scientific community and mass media have already reported the use of nonverbal behavior analysis in sports for athletes’ performance. Their conclusions stated that certain emotional expressions are linked to athlete’s performance, or even that psychological strategies serve to improve endurance performance. This paper examines the portrayal of...

Deep learning has drawn the attention of oceanographic researchers over the past few years, making the research community adopt computer vision techniques for oceanic mesoscale eddy detection on satellite altimetry gridded products. In this paper, we describe a convolutional neural network designed to detect eddies in satellite altimetry maps after...

Predicting an athlete's performance based on short footage is highly challenging. Performance prediction requires high domain knowledge and enough evidence to infer an appropriate quality assessment. Sports pundits can often infer this kind of information in real-time. In this paper, we propose regressing an ultra-distance runner cumulative race ti...

Retinopathy of prematurity (ROP) is an ocular disease caused by abnormal retinal blood vessel growth of premature infants. All premature infants who fall within a screening protocol (birth weight less than 1500 g and gestational age below 32 weeks) are diagnosed by an ophthalmological specialist for ROP. Early recognition of ROP and other diseases...

Recent advances in deep learning have made it possible to use neural networks for the detection and classification of oceanic mesoscale eddies from satellite altimetry data. Various neural network models have been proposed in recent years to address this challenge, but they have been trained using different types of input data and evaluated using d...

Underwater gliders are energy-efficient vehicles that rely on changes in buoyancy in order to convert up and down movement into forward displacement. These vehicles are conceived as multi-sensor platforms, and can be used to collect ocean data for long periods in wide range areas. This endurance is achieved at the cost of low speed, which requires...

The real-world implementation of Underwater Glider Path Planning (UGPP) over the dynamic and changing environment in deep ocean waters requires complex mission planning under very high uncertainties. Such a mission is also influenced to a large extent by remote sensing for forecasting weather models outcomes used to predict spatial currents in deep...

Planning a glider mission in an uncertain environment, with multiple parameters at the same time is a hard 4D (longitude, latitude, depth, time) Multi-Objective Optimization Problem. Multiple approaches are being used to do Glider Path Planning. In this work, we present a new system for helping multi-objective glider path planning in real missions,...

This paper presents an application of a recently well performing evolutionary algorithm for continuous numerical optimization, Success-History Based Adaptive Differential Evolution Algorithm (SHADE) including Linear population size reduction (L-SHADE), to an expert system for underwater glider path planning (UGPP). The proposed algorithm is compare...

Wind sensors are essential components of any sailboat, meanwhile they are also one of its most compromised and exposed elements. This paper introduces a novel approach that allows to estimate wind direction and speed based on the application of a particle filter technique that relies on a model dynamics of the sailboat. The proposal incorporates el...

This paper presents an approach for tackling constrained underwater glider path planning (UGPP), where the feasible path area is defined as a corridor around the border of an ocean eddy. The objective of the glider here is to sample the oceanographic variables more efficiently while keeping a bounded trajectory. Therefore, we propose a solution bas...

In this paper we present a first prototype of a teleoperated telepresence system that uses a range of devices aimed at operating a mobile platform situated in a remote location while offering the user the sensation of being at that remote location. The objective of this first stage of the project is to achieve a sensation of telepresence by simply...

While traditional glider path planning relies on constant model forecasts with only a limited range, a new approach to glider operation is based on automated topographical understanding incorporating feature detection and tracking techniques. Encapsulating the key elements of ocean structure dynamics and focusing on their behaviour an alternative p...

In this paper we present a first prototype of a teleoperated telepresence system that uses a range of devices aimed at operating a mobile platform situated in a remote location while offering the user the sensation of being at that remote location. The objective of this first stage of the project is to achieve a sensation of telepresence by simply...

The ocean, as vast as it is complex, has a plethora of phenomena that are of legitimate scientific interest, e.g., ocean fronts and Lagrangian Coherent Structures. These coherent ocean features occur from tidal mixing and ocean circulation, and are generally characterized with narrow bands of locally intensive physical gradients with enhanced circu...

This paper presents an approach to underwater glider path planning (UGPP), where the population size reduction mechanism is introduced into the differential evolution (DE) meta-heuristic and two types of DE strategies (DE/best and DE/rand) are applied interchangeably. The newly proposed DE instance algorithms using population size reduction on the...

This paper presents ongoing work aimed at the development of a multithreaded open source sailboat controller based on low cost Arduinocompatible hardware and ChibiOS/RT, a small and agile real-time operating system. The results achieved so far prove that this approach, that relies intensively on the programming resources provided by the real-time m...

The automatic extraction of biometric descriptors of anonymous people is a challenging scenario in camera networks. This task is typically accomplished making use of visual information. Calibrated RGBD sensors make possible the extraction of point cloud information. We present a novel approach for people semantic description and re-identification u...

The special characteristics of the ocean gliders propulsion scheme endows these vehicles with an extremely high endurance, at the cost of a relatively low surge speed. Hence, gliders' trajectory is highly influenced by ocean currents, which can even exceed the vehicle's nominal velocity. This calls for intelligent path planning algorithms, fact tha...

This paper presents an approach where differential evolution is applied to underwater glider path planning. The objective of a glider is to reach a target location and gather research data along its path by propelling itself underwater and returning periodically to the surface. The main hypothesis of this work is that gliders operational capabiliti...

In this paper, we prove that depth information provided by a consumer depth camera is a reliable data source to perform robust people counting. The adoption of a top view configuration reduces the space problem complexity for this task, while preserving privacy. Two different background subtraction approaches for color images are transferred to thi...

This work presents a small and affordable autonomous sailboat platform designed to be transported and operated by one or two people without any special means. The sailboat is based on a RC One Meter class vessel equipped with a low power 8-bit microcontroller board and a set of navigation sensors (compass, GPS, wind vane, ...) and a 868 MHz RF modu...

The re-identification problem has been commonly accomplished using appearance features based on salient points and color information. In this paper, we focus on the possibilities that simple geometric features obtained from depth images captured with RGB-D cameras may offer for the task, particularly working under severe illumination conditions. Th...

In the present work we propose a method for generating ocean glider trajectories that optimize the sampling of mesoscale eddy structures based on a given objective function and the predictions available from ROMS maps. The eddy structure is modeled as a 3D volume discretized into sectors rotating around its center at different velocities. The objec...

In this paper we propose a deterministic route planner for a sailboat suitable for areas where high quality wind and currents forecasts are available. An optimization based approach is used with the objective of minimizing the time required to arrive at a destination. Several simulations have been performed using high resolution regional forecasts...

Re-identification is commonly accomplished using appearance features based on salient points and color information. In this paper, we make an study on the use of different features exclusively obtained from depth images captured with RGB-D cameras. The results achieved, using simple geometric features extracted in a top-view setup, seem to provide...

In this paper, we experimentally study the combination of face and facial feature detectors to improve face detection performance. The face detection problem, as suggeted by recent face detection challenges, is still not solved. Face detectors traditionally fail in large-scale problems and/or when the face is occluded or different head rotations ar...

CoolBOT is a C++ distributed component-based programming framework for robotics. A system can be seen as a distributed network of software components interconnected by port connections where system behavior emerges from the interaction and independent execution of the components integrating the system. Recently we have endowed CoolBOT with two new...

Underwater gliders have revealed as a valuable scientific platform, with a growing number of successful environmental sampling applications. They are specially suited for long range missions due to their unmatched autonomy level, although their low surge speed makes them strongly affected by ocean currents. Path planning constitutes a real concern...

OpenCV includes different object detectors based on the Viola-Jones framework. Most of them are specialized to deal with the frontal face pattern and its inner elements: eyes, nose, and mouth. In this paper, we focus on the ear pattern detection, particularly when a head profile or almost profile view is present in the image. We aim at creating rea...

Underwater gliders are vehicles characterized by their persistence as they allow missions extremely long requiring very low power consumption. However, this also implicates that they are much more affected by adverse environmental conditions during navigation, making the use of path planning a crucial tool for this vehicles. In this work we analyse...

Path planning for Unmanned Underwater Vehicles (UUVs) is a key issue for the success and efficiency of the missions these vehicles perform. This problem is very challenging, because it must cope with dynamic and uncertain models both for the vehicle and for the environment. In the case of ocean gliders, this aspect is critical due to the strong inf...

Unmanned Underwater Vehicles (UUVs) are commonly used in Oceanography due to their relative low cost and wide range of capabilities. Gliders are a type of UUV particularly suitable for long-range missions because of their large autonomy. They change their buoyancy to dive and climb describing a vertical saw tooth pattern, which produces an effectiv...

An accurate estimation of the number of people entering/ leaving a controlled area is an interesting capability for automatic surveillance systems. Potential applications where this technology can be applied include those related to security, safety, energy saving or fraud control. In this paper we present a novel configuration of a multi-sensor sy...

This paper presents a study on the facial feature detection performance achieved using the Viola-Jones framework. A set of classifiers using two different focuses to gather the training samples is created and tested on four different datasets covering a wide range of possibilities. The results achieved should serve researchers to choose the classif...

In this paper an architecture for an estimator of short-term wind farm power is proposed. The estimator is made up of a Linear Machine classifier and a set of k Multilayer Perceptrons, training each one for a specific subspace of the input space. The splitting of the input dataset into the k clusters is done using a k-means technique, obtaining the...

Underwater gliders constitute a technology in active development, which has proven very promising in Ocean Research because of its relative low cost and long mission range. Due to their low surge speed, however, gliders are strongly affected by ocean currents, making path planning a crucial tool for this type of vehicles. In this work, we present a...

The human face provides useful information during interaction; therefore, any system integrating Vision-Based Human Computer Interaction requires fast and reliable face and facial feature detection. Different approaches have focused on this ability but only open source implementations have been extensively used by researchers. A good example is the...

Underwater gliders are a technology that have demonstrated to be a valid tool for diverse applications in the oceans including validation of currents models, environmental control or security. Due to their low speed, gliders might drift significantly from the planned trajectory by effect of ocean currents, making path planning a crucial tool for th...

In this paper some experiments have been realized to test how the introduction of variable selection has an effect on the predictor performance in short-term wind farm power prediction. Variable selection based on Kraskov estimation of the mutual information will be used due to its capability to deal with sets of continuous random variables. A Mult...

Programming robotic systems is not an easy task, even developing software for simple systems may be difficult, or at least cumbersome and error prone. In this document it is presented a C++ distributed component based programming framework for robotics socalled CoolBOT, around which we have started an open source initiative which is freely availabl...

Low cost real-time depth cameras offer new sensors for a wide field of applications apart from the gaming world. Other active research scenarios as for example surveillance, can take advantage of the capabilities offered by this kind of sensors that integrate depth and visual information.

An automatic system aimed at producing a compacttridimensional description of indoor environments using a mobile3D laser scanner is described in this paper. The resultingdescription is made up of a Multi-Level Map (ML map) and aseries of plane patches extracted from the MLSM. We propose anovel plane detection algorithm, based on the efficient RANSA...

Unmanned Underwater Vehicles (UUVs) are commonly used in Oceanography due to their relative low cost and wide range of capabilities. Gliders, being UUVs, are particularly suitable for long-range missions because of their large autonomy. They change their buoyancy to dive and climb describing a vertical saw tooth route, which produces an effective b...

The use of renewable energy sources (RES) in electricity generation has many economical and environmental advantages, but has a downside in the instability and unpredictability introduced into the public electric systems. High variable energies such as wind power have a lack of stability and, to avoid short-term variations in power supplied to the...

In this chapter we describe the main problems associated with the integration of active vision and multitasking. This configuration, though attractive, must be properly handled by means of simplification strategies that cope with its inherent complexity. Besides, the programming of this kind of complex system is prone to conclude often in monolithi...

This paper describes the utilization of software design patterns and plan-based mission specification in the definition of AUVs missions. Within this approach, a mission is described in terms of a set of task-oriented plans in order to simplify mission definition and favor reutilization of some aspects of a mission. Each plan organizes how and when...

This work is related to the analysis of reactive obstacle avoidance in general, and specifically to ND algorithms family. Contrary to many previous methods, the ND approach is not aimed at devising a general motion law; instead, it operates over a reduced set of possible situations that are treated by a particular motion law. The big earning of thi...

Endowing a physical agent with the capability of moving in a safe and agile manner is an issue of capital importance during the development of these systems. In this context, our work is related to obstacle avoidance in general, and specifically to ND algorithms. Contrary to many previous methods , the ND approach is not aimed at devising a general...

Automatic face recognition has been mainly tackled by matching a new image to a set of previously computed identity models. The literature describes approximations where those identity models are based on a single sample or a set of them. However, face representation keeps being a topic of great debate in the psychology literature, with some result...

This paper describes the architecture of an active-vision system that has been conceived to ease the concurrent utilization of a visual sensor by several tasks on a mobile robot. We describe in detail the functional architecture of the system and provide several solutions to the problem of sharing the visual atten-tion when several visual tasks nee...

Los mecanismos de atención visual han sido frecuente objeto de interés por parte de investigadores en las áreas de robótica y visión por computador. Sin embargo, aspectos como el control de la mirada y la composición modular de capacidades visuales han sido estudiados en mucha menor medida. En este artículo se realiza una propuesta de arquitectura...

The use of the vision in humans is a source of inspiration for many research works in robotics. Attention mechanisms has received much of this effort, however, aspects such as gaze control and modular composition of vision capabilities have been much less analyzed. This paper describes the architecture of an active vision system that has been conce...

In general, we face recurrently some common problems when programming robotic systems: multithreading and multiprocessing, distributed computing, hardware abstraction, hardware and software integration, multiple levels of abstraction and control, the development of a programming tool for a group of users which may become wide and diverse, etc. In t...

Very little attention has been devoted to the problem of modular composition of vision capabilities in perception-action systems. While new algorithms and techniques have paved the way for important developments, the majority of vision systems are still designed and integrated in a very primitive way according to modern software engineering princip...

This paper describes the architecture of an active-vision system that has been conceived to ease the concurrent utilization of a visual sensor by several tasks on a mobile robot. We describe in detail the functional architecture of the system and provide several solutions to the problem of sharing the visual attention among multiple visual tasks. T...

We investigate mechanisms which can endow the computer with the ability of describing a human face by means of computer vision techniques. This is a necessary requirement in order to develop HCI approaches which make the user feel himself/herself perceived. This paper describes our experiences considering gender, race and the presence of moustache...

In this work, we present a hybrid control structure for a mobile robotic system which combines concepts from homeostatic control and adaptive behavior. The homeostatic control consists of a set of control lo ops operating on state variables. These variables are computed from pre-categorical sensory data and also from high-level application results....