This study introduces a novel, drone-based approach for the detection and classification of Greater Caribbean Manatees (Trichechus manatus manatus) in Panama Canal Basin by integrating advanced deep learning techniques. Leveraging high-performance YOLOv8 model augmented with Sliced Aided Hyper Inferencing (SAHI) improved small-object detection, our system accurately identifies individual manatees, mother–calf pairs, group formations across challenging aquatic environment. Additionally, use...

This work presents a methodology to automatically detect and identify manatee vocalizations in continuous passive acoustic underwater recordings. Given that of each present slightly different frequency content, it is possible individuals using non-invasive approach. The recordings are processed four stages, including detection, denoising, classification, counting identification by vocalization clustering. main contribution this considering the spectrogram as an image (i.e., two-dimensional...

We evaluated the potential of using convolutional neural networks in classifying spectrograms Antillean manatee (Trichechus manatus manatus) vocalizations. Spectrograms binary, linear and logarithmic amplitude formats were considered. Two deep (DCNN) architectures tested: (fixed filter size) pyramidal (incremental size). Six experiments devised for testing accuracy obtained each spectrogram representation architecture combination. Results show that binary with both dropout provide a...

Malaria control in Panama is problematic due to the high diversity of morphologically similar Anopheles mosquito species, which makes identification vectors human Plasmodium challenging. Strategies by Panamanian health authorities bring malaria under targeting could be ineffective if they tackle a misidentified species.A rapid mass spectrometry procedure was developed accurately and timely sort out field-collected Neotropical mosquitoes into vector non-vector species. Matrix-assisted laser...

Passenger flow modeling and station dwelling time estimation are significant elements for railway mass transit planning, but system operators usually have limited information to model the passenger flow. In this paper, an artificial-intelligence technique known as fuzzy logic is applied of origin-destination matrix stations in a transport system. The inference engine used algorithm based principle maximum entropy. approach considers passengers’ preferences assign level congestion each car...

Introduction This work presents an unsupervised learning-based methodology to identify and count unique manatees using underwater vocalization recordings. Methods The proposed approach uses Scattering Wavelet Transform (SWT) represent individual manatee vocalizations. A Manifold Learning approach, known as PacMAP, is employed for dimensionality reduction. density-based algorithm, Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN), used clusters compared a...

Matrix-assisted laser desorption/ionization (MALDI) time-of-flight mass spectrometry is an analytical method that detects macromolecules can be used for proteomic fingerprinting and taxonomic identification in arthropods. The conventional MALDI approach uses fresh laboratory-reared arthropod specimens to build a reference spectra library with high-quality standards required achieve reliable identification. However, this may not possible accomplish some groups are difficult rear under...

Abstract Geographic variation in the vocal behavior of manatees has been reported but is largely unexplored. Vocalizations wild West Indian ( Trichechus manatus ) were recorded with hydrophones Florida from latirostris ), and Belize Panama Antillean to determine if calls varied between subspecies geographic regions. Calls visually classified into five categories: squeaks, high squeals, squeak-squeals, chirps. From these categories, only three call types (squeaks, squeaks squeals) observed...

In this work a methodology for the management of partial services on Line 1 Panama Metro is proposed.To validate proposal results are presented in series simulations that allow characterizing model proposed with at railway engineering level.

This paper proposes a new method for identifying ARMA models in the presence of additive white noise. The operates with two main steps. First, noisy model is approximated by sum an high-order AR parameters as well driving noise and variances are estimated using errors-in-variables approach. Second, coefficients extracted from those previously identified means existing techniques. In particular, three different methods considered compared solving second step. effectiveness described...

In ecology, there are several strategies used for gathering samples in orde to understand behavioral patterns of an animal, or quantify their population size. the case endangered species, techniques such as, mark and recapture, becomes impossible. Thus, researchers collect data passive ways mostly, bioacoustic, visual monitoring. this work we present elements construction a integration platform geared structure acoustic obtained from monitoring Antillean Manatee, species marine mammals that...

This paper explores the process of implementing Convolutional Neural Networks (CNNs) on resource-constrained microcontroller for real-time manatee detection in an edge computing application. The steps taken to optimize CNN model and reduce its computational burden without compromising performance are outlined. Techniques such as weight pruning quantization employed enhance efficiency deployed model. results our implementation demonstrate feasibility using CNNs microcontrollers applications....

In Panama, there are community organizations that guarantee access to water for human consumption more than 20% of the country's total population. For sustainability resource, it is essential involve communities in process monitoring quality. This can be achieved through implementation participatory biomonitoring using macroinvertebrates as indicators. fact, has been determined presence different families these organisms ecosystems associated levels their ecological work aims develop a...

The internet of things covers a large number solutions, among which we have the LoRa technology. following work present theoretical and technical aspects implementation this technology in wireless network to monitor environmental parameters. document brief look physical layer network, such as frequency bands type modulation, well details components procedures performed our results performance terms signal-to-noise ratio (SNR) function distance.

This work presents a scheme for the detection of manatee vocalizations in underwater recordings to support efforts monitoring and population estimation this species western Panama. The proposed automatic uses autoregressive model as feature extraction stage feed two-layer feedforward neural networks that classify signal or background noise. network was trained with scaled conjugate gradient backpropagation algorithm using supervised learning. provides an accuracy 92.4% on training set both classes.

Recent advances in applications of Brain-Computer Interfaces (BCIs) for control and communication more complex systems have demonstrated their potential the augmentation human cognition. This paper aims to develop a BrainComputer Interface application capable monitoring focus level user real time produce binarized output signal based on such electrophysiological activity show that real-time, low-cost BCI can be implemented industrialized settings. achieved by applying processing stage...

With the globalization and proliferation of COVID-19, new challenges have emerged in higher education regarding use integration technologies. In this context, university is called upon to organize, distribute, design best strategies overcome challenge distance times pandemic. This document proposes elements that support implementation a Digital Transformation Plan Technological University Panama. The digital resource platform developed for professors researchers share with their students,...

Comparing processes or models is of interest in various applications. Among the existing approaches, one most popular methods to use Kullback-Leibler (KL) divergence which related Shannon's entropy. Similarly, Rényi order α can be deduced from When tends 1, it leads KL divergence. In this paper, our purpose derive expression between probability density functions k consecutive samples two real first-order moving average (MA) by using eigen-decompositions their Toeplitz correlation matrices....

This paper shows the results from computation cost evaluation of three blind source separation algorithms. The algorithms tested were: FastICA, Adaptive Algorithm Based on Natural Gradient, and EASI Relative Gradient. were chosen for their relative simplicity, taking into account hardware implementation feasibility, either a FPGA or an ASIC, as part system acoustic localization mobile agents in industrial environments.

Radio intelligence is a novel approach in wireless communication systems it arises to solve the current saturation of spectrum. There are three access schemes for radius: interweave, underlay and overlay. The overlay scheme may allow higher transmission speeds bandwidth through application mitigation techniques or management interference. Through these quality efficiency guaranteed. This paper presents study analysis that be implemented future systems. characteristics some most common...