We evaluated the potential of using convolutional neural networks in classifying spectrograms of Antillean manatee (Trichechus manatus manatus) vocalizations. Spectrograms using binary, linear and logarithmic amplitude formats were considered. Two deep convolutional neural networks (DCNN) architectures were tested: linear (fixed filter size) and pyramidal (incremental filter size). Six experiments were devised for testing the accuracy obtained for each spectrogram representation and architecture combination. Results show that binary spectrograms with both linear and pyramidal architectures with dropout provide a classification rate of 94–99% on the training and 92–98% on the testing set, respectively. The pyramidal network presents a shorter training and inference time. Results from the convolutional neural networks (CNN) are substantially better when compared with a signal processing fast Fourier transform (FFT)-based harmonic search approach in terms of accuracy and F1 Score. Taken together, these results prove the validity of using spectrograms and using DCNNs for manatee vocalization classification. These results can be used to improve future software and hardware implementations for the estimation of the manatee population in Panama.

This work presents a methodology to automatically detect and identify manatee vocalizations in continuous passive acoustic underwater recordings. Given that vocalizations of each manatee present a slightly different frequency content, it is possible to identify individuals using a non-invasive acoustic approach. The recordings are processed in four stages, including detection, denoising, classification, and manatee counting and identification by vocalization clustering. The main contribution of this work is considering the vocalization spectrogram as an image (i.e., two-dimensional pattern) and representing it in terms of principal component analysis coefficients that feed a clustering approach. A performance study is carried out for each stage of the scheme. The methodology is tested to analyze three years of recordings from two wetlands in Panama to support ongoing efforts to estimate the manatee population.

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 for the estimation of the elements of the origin-destination matrix and the dwelling time of stations in a railway transport system. The fuzzy inference engine used in the algorithm is based in the principle of maximum entropy. The approach considers passengers’ preferences to assign a level of congestion in each car of the train in function of the properties of the station platforms. This approach is implemented to estimate the passenger flow and dwelling times of the recently opened Line 1 of the Panama Metro. The dwelling times obtained from the simulation are compared to real measurements to validate the approach.

This chapter focuses on the contributions made in the development of assistive technologies for the navigation of blind and visually impaired (BVI) individuals. A special interest is placed on vision-based systems that make use of image (RGB) and depth (D) information to assist their indoor navigation. Many commercial RGB-D cameras exist on the market, but for many years the Microsoft Kinect has been used as a tool for research in this field. Therefore, first-hand experience and advances on the use of Kinect for the development of an indoor navigation aid system for BVI individuals is presented. Limitations that can be encountered in building such a system are addressed at length. Finally, an overview of novel avenues of research in indoor navigation for BVI individuals such as integration of computer vision algorithms, deep learning for the classification of objects, and recent developments with stereo depth vision are discussed.