|
|
Jorge L. Charco, Angel D. Sappa, Boris X. Vintimilla, & Henry O. Velesaca. (2020). Transfer Learning from Synthetic Data in the Camera Pose Estimation Problem. In The 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2020); Valletta, Malta; 27-29 Febrero 2020 (Vol. 4, pp. 498–505).
Abstract: This paper presents a novel Siamese network architecture, as a variant of Resnet-50, to estimate the relative camera pose on multi-view environments. In order to improve the performance of the proposed model
a transfer learning strategy, based on synthetic images obtained from a virtual-world, is considered. The
transfer learning consist of first training the network using pairs of images from the virtual-world scenario
considering different conditions (i.e., weather, illumination, objects, buildings, etc.); then, the learned weight
of the network are transferred to the real case, where images from real-world scenarios are considered. Experimental results and comparisons with the state of the art show both, improvements on the relative pose
estimation accuracy using the proposed model, as well as further improvements when the transfer learning
strategy (synthetic-world data – transfer learning – real-world data) is considered to tackle the limitation on
the training due to the reduced number of pairs of real-images on most of the public data sets.
|
|
|
|
Charco, J. L., Sappa, A.D., Vintimilla, B.X., Velesaca, H.O. (2021). Camera pose estimation in multi-view environments:from virtual scenarios to the real world. In Image and Vision Computing Journal. (Article number 104182), Vol. 110.
Abstract: This paper presents a domain adaptation strategy to efficiently train network architectures for estimating the relative camera pose in multi-view scenarios. The network architectures are fed by a pair of simultaneously acquired
images, hence in order to improve the accuracy of the solutions, and due to the lack of large datasets with pairs of
overlapped images, a domain adaptation strategy is proposed. The domain adaptation strategy consists on transferring the knowledge learned from synthetic images to real-world scenarios. For this, the networks are firstly
trained using pairs of synthetic images, which are captured at the same time by a pair of cameras in a virtual environment; and then, the learned weights of the networks are transferred to the real-world case, where the networks are retrained with a few real images. Different virtual 3D scenarios are generated to evaluate the
relationship between the accuracy on the result and the similarity between virtual and real scenarios—similarity
on both geometry of the objects contained in the scene as well as relative pose between camera and objects in the
scene. Experimental results and comparisons are provided showing that the accuracy of all the evaluated networks for estimating the camera pose improves when the proposed domain adaptation strategy is used,
highlighting the importance on the similarity between virtual-real scenarios.
|
|
|
|
Ma. Paz Velarde, Erika Perugachi, Dennis G. Romero, Ángel D. Sappa, & Boris X. Vintimilla. (2015). Análisis del movimiento de las extremidades superiores aplicado a la rehabilitación física de una persona usando técnicas de visión artificial. Revista Tecnológica ESPOL-RTE, Vol. 28, pp. 1–7.
Abstract: Comúnmente durante la rehabilitación física, el diagnóstico dado por el especialista se basa en observaciones cualitativas que sugieren, en algunos casos, conclusiones subjetivas. El presente trabajo propone un enfoque cuantitativo, orientado a servir de ayuda a fisioterapeutas, a través de una herramienta interactiva y de bajo costo que permite medir los movimientos de miembros superiores. Estos movimientos son capturados por un sensor RGB-D y procesados mediante la metodología propuesta, dando como resultado una eficiente representación de movimientos, permitiendo la evaluación cuantitativa de movimientos de los miembros superiores.
|
|
|
|
Arias Alexandra. Ing., Peña Roxanna. Ing., Chávez Patricia. MSEE., & Basurto Juan. Ing. (2011). Análisis Comparativo de la Implementación de una PBX de Código Abierto instalada en un Servidor Tradicional y en un Enrutador Inalámbrico en términos de Calidad de Servicio en Redes Inalámbricas Amalladas. Revista Tecnologica ESPOL RTE, Vol. 24, pp. 1–6.
Abstract: El presente trabajo compara dos Implementaciones de Centrales Telefónicas VoIP de Código Abierto implementados sobre una Red Inalámbrica Amallada. El primero comprende la instalación de la PBX en un servidor tradicional y el segundo la instalación de una PBX en un enrutador inalámbrico. Nuestro objetivo es
determinar cuál de estos dos sistemas es superior en cuanto a calidad de servicio se refiere. Para determinar la mejor solución, realizamos un estudio técnico de los paquetes capturados durante diferentes pruebas, considerando parámetros como el ancho de banda, retardo y jitter. Nuestros métodos de análisis pueden ser utilizados para futuros trabajos con una mayor complejidad y número de enrutadores inalámbrico, así como establecer el grado de afectación y el comportamiento de las dos PBX cuando haya congestión en la red.
|
|
|
|
Luis Jacome-Galarza, M. V. - C., Miguel Realpe-Robalino, Jose Benavides-Maldonado. (2021). Software Engineering and Distributed Computing in image processing intelligent systems: a systematic literature review. In 19th LACCEI International Multi-Conference for Engineering, Education, and Technology.
Abstract: Deep learning is experiencing an upward technology trend that is revolutionizing intelligent systems in several domains, such as image and speech recognition, machine translation, social network filtering, and the like. By reviewing a total of 80 studies reported from 2016 to 2020, the present article evaluates the application of software engineering to the field
of intelligent image processing systems, it also offers insights about aspects related to distributed computing for this type of systems. Results indicate that several topics of software engineering are mostly applied when academics are involved in developing projects associated to this kind of intelligent systems. The findings provide evidences that Apache Spark is the most
utilized distributed computing framework for image processing. In addition, Tensorflow is a popular framework used to build convolutional neural networks, which are the prevailing deep learning algorithms used in intelligent image processing systems.
Also, among big cloud providers, Amazon Web Services is the preferred computing platform across the industry sectors, followed by Google cloud.
|
|
|
|
Rubio, G. A., Agila, W.E. (2021). A fuzzy model to manage water in polymer electrolyte membrane fuel cells. In Processes Journal. (Article number 904), Vol. 9(Issue 6).
Abstract: In this paper, a fuzzy model is presented to determine in real-time the degree of dehydration or flooding of a proton exchange membrane of a fuel cell, to optimize its electrical response and consequently, its autonomous operation. By applying load, current and flux variations in the dry, normal, and flooded states of the membrane, it was determined that the temporal evolution of the fuel cell voltage is characterized by changes in slope and by its voltage oscillations. The results were validated using electrochemical impedance spectroscopy and show slope changes from 0.435 to 0.52 and oscillations from 3.6 mV to 5.2 mV in the dry state, and slope changes from 0.2 to 0.3 and oscillations from 1 mV to 2 mV in the flooded state. The use of fuzzy logic is a novelty and constitutes a step towards the progressive automation of the supervision, perception, and intelligent control of fuel cells, allowing them to reduce their risks and increase their economic benefits.
|
|
|
|
W. Agila, Gomer Rubio, L. Miranda, & D. Sanaguano. (2019). Open Control Architecture for the Characterization and Control of the PEM Fuel Cell. In IEEE ETCM 2019 Fourth Ecuador Technical Chapters Meeting; Guayaquil, Ecuador (pp. 1–5).
Abstract: Proton exchange membrane (PEM) fuel cells, are an efficient and clean source of electrical energy. The analysis of its operation requires experimental work, which allows measuring, modeling and optimizing PEM fuel cells electrical behavior under different operating conditions. Therefore, having an experimentation platform that allows to easily carry out its study and control is essential. This research presents the design and development of an open instrumental system that allows measuring, controlling and determining the operating parameters of a PEM fuel cell. As results, the polarization curves, voltage-current, obtained by the system itself in different experimental conditions are shown. These curves are a very useful tool to evaluate the electrical behavior of the PEM battery.
|
|
|
|
Dennis G. Romero, A. F. Neto, T. F. Bastos, & Boris X. Vintimilla. (2012). An approach to automatic assistance in physiotherapy based on on-line movement identification. In VI Andean Region International Conference – ANDESCON 2012. Andean Region International Conference (ANDESCON), 2012 VI: IEEE.
Abstract: This paper describes a method for on-line movement identification, oriented to patient’s movement evaluation during physiotherapy. An analysis based on Mahalanobis distance between temporal windows is performed to identify the “idle/motion” state, which defines the beginning and end of the patient’s movement, for posterior patterns extraction based on Relative Wavelet Energy from sequences of invariant moments.
|
|
|
|
Ángel Morera, Á. S., A. Belén Moreno, Angel D. Sappa, & José F. Vélez. (2020). SSD vs. YOLO for Detection of Outdoor Urban Advertising Panels under Multiple Variabilities. In Sensors, Vol. 2020-August(16), pp. 1–23.
Abstract: This work compares Single Shot MultiBox Detector (SSD) and You Only Look Once (YOLO)
deep neural networks for the outdoor advertisement panel detection problem by handling multiple
and combined variabilities in the scenes. Publicity panel detection in images o ers important
advantages both in the real world as well as in the virtual one. For example, applications like Google
Street View can be used for Internet publicity and when detecting these ads panels in images, it could
be possible to replace the publicity appearing inside the panels by another from a funding company.
In our experiments, both SSD and YOLO detectors have produced acceptable results under variable
sizes of panels, illumination conditions, viewing perspectives, partial occlusion of panels, complex
background and multiple panels in scenes. Due to the diculty of finding annotated images for the
considered problem, we created our own dataset for conducting the experiments. The major strength
of the SSD model was the almost elimination of False Positive (FP) cases, situation that is preferable
when the publicity contained inside the panel is analyzed after detecting them. On the other side,
YOLO produced better panel localization results detecting a higher number of True Positive (TP)
panels with a higher accuracy. Finally, a comparison of the two analyzed object detection models
with di erent types of semantic segmentation networks and using the same evaluation metrics is
also included.
|
|
|
|
Mildred Cruz, Cristhian A. Aguilera, Boris X. Vintimilla, Ricardo Toledo, & Ángel D. Sappa. (2015). Cross-spectral image registration and fusion: an evaluation study. In 2nd International Conference on Machine Vision and Machine Learning (Vol. 331). Barcelona, Spain: Computer Vision Center.
Abstract: This paper presents a preliminary study on the registration and fusion of cross-spectral imaging. The objective is to evaluate the validity of widely used computer vision approaches when they are applied at different spectral bands. In particular, we are interested in merging images from the infrared (both long wave infrared: LWIR and near infrared: NIR) and visible spectrum (VS). Experimental results with different data sets are presented.
|
|
