All-in-one aerial image enhancement network for forest scenes.

Drone monitoring plays an irreplaceable and significant role in forest firefighting due to its characteristics of wide-range observation and real-time messaging. However, aerial images are often susceptible to different degradation problems before performing high-level visual tasks including but not limited to smoke detection, fire classification, and regional localization. Recently, the majority of image enhancement methods are centered around particular types of degradation, necessitating the memory unit to accommodate different models for distinct scenarios in practical applications. Furthermore, such a paradigm requires wasted computational and storage resources to determine the type of degradation, making it difficult to meet the real-time and lightweight requirements of real-world scenarios. In this paper, we propose an All-in-one Image Enhancement Network (AIENet) that can restore various degraded images in one network. Specifically, we design a new multi-scale receptive field image enhancement block, which can better reconstruct high-resolution details of target regions of different sizes. In particular, this plug-and-play module enables it to be embedded in any learning-based model. And it has better flexibility and generalization in practical applications. This paper takes three challenging image enhancement tasks encountered in drone monitoring as examples, whereby we conduct task-specific and all-in-one image enhancement experiments on a synthetic forest dataset. The results show that the proposed AIENet outperforms the state-of-the-art image enhancement algorithms quantitatively and qualitatively. Furthermore, extra experiments on high-level vision detection also show the promising performance of our method compared with some recent baselines.

Determinants of the Attitude to COVID-19 Vaccine in Lima-Peru: Path Analysis and Structural Regression.

Introduction: Research on the effects of COVID-19 has shown that a favorable attitude toward the COVID-19 vaccine would help reduce the pandemic's sequelae and avoid lethal variants. Objective: A theoretical model was tested through the strategy of path analysis and structural equation modeling, seeking to evaluate the direct effect of neuroticism and the indirect effects of risk-avoidance and rule-following behaviors, mediated by attitudes toward science. Methods: A total of 459 adults, mostly women (61%), mean age 28.51 (SD = 10.36), living in Lima (Peru), participated. The scales of neuroticism, risk avoidance behavior (RAB), norm following (NF), attitudes toward science, and attitudes toward vaccination were administered. Results: The path analysis explained 36% of the variance in vaccine attitude, whereas the latent structural regression model achieved a 54% explanation; according to this model attitude toward science (ß=.70, p < .01) and neuroticism (ß=-.16, p < .01) are significant predictors of vaccine attitude. Likewise, risk avoidance behavior and rule-following have indirect effects on attitudes toward vaccination. Conclusion: Low neuroticism and a positive attitude toward the science that mediates the effects of RAB and NF directly condition the possibility of vaccination against COVID-19 in the adult population.

Semi-supervised wildfire smoke detection based on smoke-aware consistency.

The semi-transparency property of smoke integrates it highly with the background contextual information in the image, which results in great visual differences in different areas. In addition, the limited annotation of smoke images from real forest scenarios brings more challenges for model training. In this paper, we design a semi-supervised learning strategy, named smoke-aware consistency (SAC), to maintain pixel and context perceptual consistency in different backgrounds. Furthermore, we propose a smoke detection strategy with triple classification assistance for smoke and smoke-like object discrimination. Finally, we simplified the LFNet fire-smoke detection network to LFNet-v2, due to the proposed SAC and triple classification assistance that can perform the functions of some specific module. The extensive experiments validate that the proposed method significantly outperforms state-of-the-art object detection algorithms on wildfire smoke datasets and achieves satisfactory performance under challenging weather conditions.

A Hybrid Visual-Based SLAM Architecture: Local Filter-Based SLAM with KeyFrame-Based Global Mapping.

This work presents a hybrid visual-based SLAM architecture that aims to take advantage of the strengths of each of the two main methodologies currently available for implementing visual-based SLAM systems, while at the same time minimizing some of their drawbacks. The main idea is to implement a local SLAM process using a filter-based technique, and enable the tasks of building and maintaining a consistent global map of the environment, including the loop closure problem, to use the processes implemented using optimization-based techniques. Different variants of visual-based SLAM systems can be implemented using the proposed architecture. This work also presents the implementation case of a full monocular-based SLAM system for unmanned aerial vehicles that integrates additional sensory inputs. Experiments using real data obtained from the sensors of a quadrotor are presented to validate the feasibility of the proposed approach.

Monocular Visual SLAM Based on a Cooperative UAV-Target System.

To obtain autonomy in applications that involve Unmanned Aerial Vehicles (UAVs), the capacity of self-location and perception of the operational environment is a fundamental requirement. To this effect, GPS represents the typical solution for determining the position of a UAV operating in outdoor and open environments. On the other hand, GPS cannot be a reliable solution for a different kind of environments like cluttered and indoor ones. In this scenario, a good alternative is represented by the monocular SLAM (Simultaneous Localization and Mapping) methods. A monocular SLAM system allows a UAV to operate in a priori unknown environment using an onboard camera to simultaneously build a map of its surroundings while at the same time locates itself respect to this map. So, given the problem of an aerial robot that must follow a free-moving cooperative target in a GPS denied environment, this work presents a monocular-based SLAM approach for cooperative UAV-Target systems that addresses the state estimation problem of (i) the UAV position and velocity, (ii) the target position and velocity, (iii) the landmarks positions (map). The proposed monocular SLAM system incorporates altitude measurements obtained from an altimeter. In this case, an observability analysis is carried out to show that the observability properties of the system are improved by incorporating altitude measurements. Furthermore, a novel technique to estimate the approximate depth of the new visual landmarks is proposed, which takes advantage of the cooperative target. Additionally, a control system is proposed for maintaining a stable flight formation of the UAV with respect to the target. In this case, the stability of control laws is proved using the Lyapunov theory. The experimental results obtained from real data as well as the results obtained from computer simulations show that the proposed scheme can provide good performance.

Design and Implementation of a Virtual Sensor Network for Smart Waste Water Monitoring.

Monitoring and analysis of open air basins is a critical task in waste water plant management. These tasks generally require sampling waters at several hard to access points, be it real time with multiparametric sensor probes, or retrieving water samples. Full automation of these processes would require deploying hundreds (if not thousands) of fixed sensors, unless the sensors can be translated. This work proposes the utilization of robotized unmanned aerial vehicle (UAV) platforms to work as a virtual high density sensor network, which could analyze in real time or capture samples depending on the robotic UAV equipment. To check the validity of the concept, an instance of the robotized UAV platform has been fully designed and implemented. A multi-agent system approach has been used (implemented over a Robot Operating System, ROS, middleware layer) to define a software architecture able to deal with the different problems, optimizing modularity of the software; in terms of hardware, the UAV platform has been designed and built, as a sample capturing probe. A description on the main features of the multi-agent system proposed, its architecture, and the behavior of several components is discussed. The experimental validation and performance evaluation of the system components has been performed independently for the sake of safety: autonomous flight performance has been tested on-site; the accuracy of the localization technologies deemed as deployable options has been evaluated in controlled flights; and the viability of the sample capture device designed and built has been experimentally tested.

Visual-Based SLAM Configurations for Cooperative Multi-UAV Systems with a Lead Agent: An Observability-Based Approach.

In this work, the problem of the cooperative visual-based SLAM for the class of multi-UA systems that integrates a lead agent has been addressed. In these kinds of systems, a team of aerial robots flying in formation must follow a dynamic lead agent, which can be another aerial robot, vehicle or even a human. A fundamental problem that must be addressed for these kinds of systems has to do with the estimation of the states of the aerial robots as well as the state of the lead agent. In this work, the use of a cooperative visual-based SLAM approach is studied in order to solve the above problem. In this case, three different system configurations are proposed and investigated by means of an intensive nonlinear observability analysis. In addition, a high-level control scheme is proposed that allows to control the formation of the UAVs with respect to the lead agent. In this work, several theoretical results are obtained, together with an extensive set of computer simulations which are presented in order to numerically validate the proposal and to show that it can perform well under different circumstances (e.g., GPS-challenging environments). That is, the proposed method is able to operate robustly under many conditions providing a good position estimation of the aerial vehicles and the lead agent as well.

UAV Visual and Laser Sensors Fusion for Detection and Positioning in Industrial Applications.

This work presents a solution to localize Unmanned Autonomous Vehicles with respect to pipes and other cylindrical elements found in inspection and maintenance tasks both in industrial and civilian infrastructures. The proposed system exploits the different features of vision and laser based sensors, combining them to obtain accurate positioning of the robot with respect to the cylindrical structures. A probabilistic (RANSAC-based) procedure is used to segment possible cylinders found in the laser scans, and this is used as a seed to accurately determine the robot position through a computer vision system. The priors obtained from the laser scan registration help to solve the problem of determining the apparent contour of the cylinders. In turn this apparent contour is used in a degenerate quadratic conic estimation, enabling to visually estimate the pose of the cylinder.

Cooperative Monocular-Based SLAM for Multi-UAV Systems in GPS-Denied Environments.

This work presents a cooperative monocular-based SLAM approach for multi-UAV systems that can operate in GPS-denied environments. The main contribution of the work is to show that, using visual information obtained from monocular cameras mounted onboard aerial vehicles flying in formation, the observability properties of the whole system are improved. This fact is especially notorious when compared with other related visual SLAM configurations. In order to improve the observability properties, some measurements of the relative distance between the UAVs are included in the system. These relative distances are also obtained from visual information. The proposed approach is theoretically validated by means of a nonlinear observability analysis. Furthermore, an extensive set of computer simulations is presented in order to validate the proposed approach. The numerical simulation results show that the proposed system is able to provide a good position and orientation estimation of the aerial vehicles flying in formation.

Human Collaborative Localization and Mapping in Indoor Environments with Non-Continuous Stereo.

A new approach to the monocular simultaneous localization and mapping (SLAM) problem is presented in this work. Data obtained from additional bearing-only sensors deployed as wearable devices is fully fused into an Extended Kalman Filter (EKF). The wearable device is introduced in the context of a collaborative task within a human-robot interaction (HRI) paradigm, including the SLAM problem. Thus, based on the delayed inverse-depth feature initialization (DI-D) SLAM, data from the camera deployed on the human, capturing his/her field of view, is used to enhance the depth estimation of the robotic monocular sensor which maps and locates the device. The occurrence of overlapping between the views of both cameras is predicted through geometrical modelling, activating a pseudo-stereo methodology which allows to instantly measure the depth by stochastic triangulation of matched points found through SIFT/SURF. Experimental validation is provided through results from experiments, where real data is captured as synchronized sequences of video and other data (relative pose of secondary camera) and processed off-line. The sequences capture indoor trajectories representing the main challenges for a monocular SLAM approach, namely, singular trajectories and close turns with high angular velocities with respect to linear velocities.

Monocular SLAM for autonomous robots with enhanced features initialization.

This work presents a variant approach to the monocular SLAM problem focused in exploiting the advantages of a human-robot interaction (HRI) framework. Based upon the delayed inverse-depth feature initialization SLAM (DI-D SLAM), a known monocular technique, several but crucial modifications are introduced taking advantage of data from a secondary monocular sensor, assuming that this second camera is worn by a human. The human explores an unknown environment with the robot, and when their fields of view coincide, the cameras are considered a pseudo-calibrated stereo rig to produce estimations for depth through parallax. These depth estimations are used to solve a related problem with DI-D monocular SLAM, namely, the requirement of a metric scale initialization through known artificial landmarks. The same process is used to improve the performance of the technique when introducing new landmarks into the map. The convenience of the approach taken to the stereo estimation, based on SURF features matching, is discussed. Experimental validation is provided through results from real data with results showing the improvements in terms of more features correctly initialized, with reduced uncertainty, thus reducing scale and orientation drift. Additional discussion in terms of how a real-time implementation could take advantage of this approach is provided.

New validation algorithm for data association in SLAM.

In this work, a novel data validation algorithm for a single-camera SLAM system is introduced. A 6-degree-of-freedom monocular SLAM method based on the delayed inverse-depth (DI-D) feature initialization is used as a benchmark. This SLAM methodology has been improved with the introduction of the proposed data association batch validation technique, the highest order hypothesis compatibility test, HOHCT. This new algorithm is based on the evaluation of statistically compatible hypotheses, and a search algorithm designed to exploit the characteristics of delayed inverse-depth technique. In order to show the capabilities of the proposed technique, experimental tests have been compared with classical methods. The results of the proposed technique outperformed the results of the classical approaches.