Cookies Policy
The website need some cookies and similar means to function. If you permit us, we will use those means to collect data on your visits for aggregated statistics to improve our service. Find out More
Accept Reject
  • Menu
Publications

2024

Fusion of Time-of-Flight Based Sensors with Monocular Cameras for a Robotic Person Follower

Authors
Sarmento, J; dos Santos, FN; Aguiar, AS; Filipe, V; Valente, A;

Publication
JOURNAL OF INTELLIGENT & ROBOTIC SYSTEMS

Abstract
Human-robot collaboration (HRC) is becoming increasingly important in advanced production systems, such as those used in industries and agriculture. This type of collaboration can contribute to productivity increase by reducing physical strain on humans, which can lead to reduced injuries and improved morale. One crucial aspect of HRC is the ability of the robot to follow a specific human operator safely. To address this challenge, a novel methodology is proposed that employs monocular vision and ultra-wideband (UWB) transceivers to determine the relative position of a human target with respect to the robot. UWB transceivers are capable of tracking humans with UWB transceivers but exhibit a significant angular error. To reduce this error, monocular cameras with Deep Learning object detection are used to detect humans. The reduction in angular error is achieved through sensor fusion, combining the outputs of both sensors using a histogram-based filter. This filter projects and intersects the measurements from both sources onto a 2D grid. By combining UWB and monocular vision, a remarkable 66.67% reduction in angular error compared to UWB localization alone is achieved. This approach demonstrates an average processing time of 0.0183s and an average localization error of 0.14 meters when tracking a person walking at an average speed of 0.21 m/s. This novel algorithm holds promise for enabling efficient and safe human-robot collaboration, providing a valuable contribution to the field of robotics.

2024

Active Supervision: Human in the Loop

Authors
Cruz, RPM; Shihavuddin, ASM; Maruf, MH; Cardoso, JS;

Publication
PROGRESS IN PATTERN RECOGNITION, IMAGE ANALYSIS, COMPUTER VISION, AND APPLICATIONS, CIARP 2023, PT I

Abstract
After the learning process, certain types of images may not be modeled correctly because they were not well represented in the training set. These failures can then be compensated for by collecting more images from the real-world and incorporating them into the learning process - an expensive process known as active learning. The proposed twist, called active supervision, uses the model itself to change the existing images in the direction where the boundary is less defined and requests feedback from the user on how the new image should be labeled. Experiments in the context of class imbalance show the technique is able to increase model performance in rare classes. Active human supervision helps provide crucial information to the model during training that the training set lacks.

2024

Reagentless Vis-NIR Spectroscopy Point-of-Care for Feline Total White Blood Cell Counts

Authors
Barroso, TG; Queirós, C; Monteiro-Silva, F; Santos, F; Gregório, AH; Martins, RC;

Publication
BIOSENSORS-BASEL

Abstract
Spectral point-of-care technology is reagentless with minimal sampling (<10 mu L) and can be performed in real-time. White blood cells are non-dominant in blood and in spectral information, suffering significant interferences from dominant constituents such as red blood cells, hemoglobin and billirubin. White blood cells of a bigger size can account for 0.5% to 22.5% of blood spectra information. Knowledge expansion was performed using data augmentation through the hybridization of 94 real-world blood samples into 300 synthetic data samples. Synthetic data samples are representative of real-world data, expanding the detailed spectral information through sample hybridization, allowing us to unscramble the spectral white blood cell information from spectra, with correlations of 0.7975 to 0.8397 and a mean absolute error of 32.25% to 34.13%; furthermore, we achieved a diagnostic efficiency between 83% and 100% inside the reference interval (5.5 to 19.5 x 10(9) cell/L), and 85.11% for cases with extreme high white blood cell counts. At the covariance mode level, white blood cells are quantified using orthogonal information on red blood cells, maximizing sensitivity and specificity towards white blood cells, and avoiding the use of non-specific natural correlations present in the dataset; thus, the specifity of white blood cells spectral information is increased. The presented research is a step towards high-specificity, reagentless, miniaturized spectral point-of-care hematology technology for Veterinary Medicine.

2024

A large-scale empirical study on mobile performance: energy, run-time and memory

Authors
Rua, R; Saraiva, J;

Publication
EMPIRICAL SOFTWARE ENGINEERING

Abstract
Software performance concerns have been attracting research interest at an increasing rate, especially regarding energy performance in non-wired computing devices. In the context of mobile devices, several research works have been devoted to assessing the performance of software and its underlying code. One important contribution of such research efforts is sets of programming guidelines aiming at identifying efficient and inefficient programming practices, and consequently to steer software developers to write performance-friendly code.Despite recent efforts in this direction, it is still almost unfeasible to obtain universal and up-to-date knowledge regarding software and respective source code performance. Namely regarding energy performance, where there has been growing interest in optimizing software energy consumption due to the power restrictions of such devices. There are still many difficulties reported by the community in measuring performance, namely in large-scale validation and replication. The Android ecosystem is a particular example, where the great fragmentation of the platform, the constant evolution of the hardware, the software platform, the development libraries themselves, and the fact that most of the platform tools are integrated into the IDE's GUI, makes it extremely difficult to perform performance studies based on large sets of data/applications. In this paper, we analyze the execution of a diversified corpus of applications of significant magnitude. We analyze the source-code performance of 1322 versions of 215 different Android applications, dynamically executed with over than 27900 tested scenarios, using state-of-the-art black-box testing frameworks with different combinations of GUI inputs. Our empirical analysis allowed to observe that semantic program changes such as adding functionality and repairing bugfixes are the changes more associated with relevant impact on energy performance. Furthermore, we also demonstrate that several coding practices previously identified as energy-greedy do not replicate such behavior in our execution context and can have distinct impacts across several performance indicators: runtime, memory and energy consumption. Some of these practices include some performance issues reported by the Android Lint and Android SDK APIs. We also provide evidence that the evaluated performance indicators have little to no correlation with the performance issues' priority detected by Android Lint. Finally, our results allowed us to demonstrate that there are significant differences in terms of performance between the most used libraries suited for implementing common programming tasks, such as HTTP communication, JSON manipulation, image loading/rendering, among others, providing a set of recommendations to select the most efficient library for each performance indicator. Based on the conclusions drawn and in the extension of the developed work, we also synthesized a set of guidelines that can be used by practitioners to replicate energy studies and build more efficient mobile software.

2024

Map-matching methods in agriculture

Authors
Silva, A; Mendes-Moreira, J; Ferreira, C; Costa, N; Dias, D;

Publication
COMPUTERS AND ELECTRONICS IN AGRICULTURE

Abstract
In this paper, a solution to monitor the location of humans during their activity in the agriculture sector with the aim to boost productivity and efficiency is provided. Our solution is based on map-matching methods, that are used to track the path spanned by a worker along a specific activity in an agriculture culture. Two different cultures are taken into consideration in this study olives and vines. We leverage the symmetry of the geometry of these cultures into our solution and divide the problem three-fold initially, we estimate a path of a worker along the fields, then we apply the map-matching to such path and finally, a post-processing method is applied to ensure local continuity of the sequence obtained from map-matching. The proposed methods are experimentally evaluated using synthetic and real data in the region of Mirandela, Portugal. Evaluation metrics show that results for synthetic data are robust under several sampling periods, while for real-world data, results for the vine culture are on par with synthetic, and for the olive culture performance is reduced.

2024

Assessment of Multiple Fiducial Marker Trackers on Hololens 2

Authors
Costa, GM; Petry, MR; Martins, JG; Moreira, APGM;

Publication
IEEE ACCESS

Abstract
Fiducial markers play a fundamental role in various fields in which precise localization and tracking are paramount. In Augmented Reality, they provide a known reference point in the physical world so that AR systems can accurately identify, track, and overlay virtual objects. This accuracy is essential for creating a seamless and immersive AR experience, particularly when prompted to cope with the sub-millimeter requirements of medical and industrial applications. This research article presents a comparative analysis of four fiducial marker tracking algorithms, aiming to assess and benchmark their accuracy and precision. The proposed methodology compares the pose estimated by four algorithms running on Hololens 2 with those provided by a highly accurate ground truth system. Each fiducial marker was positioned in 25 sampling points with different distances and orientations. The proposed evaluation method is not influenced by human error, relying only on a high-frequency and accurate motion tracking system as ground truth. This research shows that it is possible to track the fiducial markers with translation and rotation errors as low as 1.36 mm and 0.015 degrees using ArUco and Vuforia, respectively.

  • 4
  • 3741