This work evaluates the effect of the excitation signal used when measuring the absorption coeffi... more This work evaluates the effect of the excitation signal used when measuring the absorption coefficient on an impedance tube. This paper aims to offer some guidance on the selection of the excitation signal to perform sound absorption measurements using the impedance tube. Four possible excitation signals defined in ISO 10534-2 Standard were studied: two of them, random noise and two of them sine sweep signals. Some hypotheses tests were executed to assess the homogeneity of each measurement. The signal-to-noise ratio (SNR) was also computed to verify the measurement quality. The results show that the best performing approach was accomplished using a logarithmic sweep, giving more precise sound absorption curves with an SNR of 34.15 dB. Random signals reported similar SNR (greater than 30 dB) after executing an average with 100 repetitions.
A preliminary study is presented with the aim of modeling the thermal behavior of a passive build... more A preliminary study is presented with the aim of modeling the thermal behavior of a passive building that is ventilated merely with the promotion of natural ventilation, various models have been assessed by using the system identification process. The identification of a simplify and lite model of such thermal behavior is needed to later control the thermal comfort of the indoor environment through the building natural ventilation openings and window blinds. A physical-phenomena-based model using electrical analogies is built upon hypotheses allowed by the architectural features of the building. This helps analyze the interaction between the main elements of the physical domain, where the thermal behavior is only determined by the indoor air and concrete-slab temperatures. Three model approaches are examined with the help of the system identification toolbox: State space, Process models (linear and frequency domain), and Nonlinear representation. The nonlinear representation model i...
Spatial audio or 3D audio refers to a set of techniques whose main objective is to simulate sound... more Spatial audio or 3D audio refers to a set of techniques whose main objective is to simulate sound sources located in arbitrary positions in space. This is possible thanks to the Head-Related Transfer Functions (HRTFs), which model the anatomical characteristics of the subject and their interaction with the incident sound field. Since HRTFs are measured at discrete positions in space, interpolation techniques such as Bilinear or Triangular Spherical are necessary. However, both methods require uniform interpolation meshes. Hence, in this work, we propose a new methodology to HRTFs databases measured on non-uniform meshes for the bilinear and triangular spherical interpolation methods. Our results suggest that our proposed methodology for bilinear and triangular spherical interpolations can be used as a viable alternative for non-uniform meshes. We evaluate our approach in terms of spectral distortion, achieving similar results when compared to spline interpolation as a baseline.
El presente proyecto "DISENO E IMPLEMENTACION DE UN SISTEMA CONTROLADOR DE TEMPERATURA PID P... more El presente proyecto "DISENO E IMPLEMENTACION DE UN SISTEMA CONTROLADOR DE TEMPERATURA PID PARA LA UNIDAD AIR FLOW TEMPERATURE CONTROL SYSTEM MEDIANTE LA UTILIZACION DE LA HERRAMIENTA RTW (REAL TIME WORKSHOP) DE MATLAB"abarca el estudio de la herramienta RTW en donde se analizo la arquitectura, el algoritmo, el proceso y las etapas en el desarrollo de un modelo en tiempo real por medio del sistema SIMULINK® de MATLAB®, ademas se realizo una HMI la cual mediante subsistemas .mdl desarrollados en SIMULINK® permitieron la identificacion de la planta, el control PID experimental y la simulacion del control PID, utilizando para esto la tarjeta de adquisicion y generacion de datos Nationals Instruments PCI 6221, la misma que, luego de haberla configurado permitio interconectar la planta real con los subsistemas de la HMI desarrolla.
Digital documents are accessed by visually impaired people (VIP) through screen readers. Traditio... more Digital documents are accessed by visually impaired people (VIP) through screen readers. Traditionally, digital documents were translated to braille text, but screen readers have proved to be efficient for the acquisition of digital document knowledge by VIP. However, screen readers and other assistive technologies have significant limitations when there exist tables in digital documents such as portable document format (PDF). For instance, screen readers can not follow the correct reading sequence of the table based on its visual structure causing this content is inaccessible for VIP. In order to deal with this problem, in this work, we developed a system for the retrieval of table information from PDF documents for use in screen readers used by visually impaired people. The proposed methodology takes advantage of computer vision techniques with a deep learning approach to make documents accessible instead of the classical rule-based programming approach. We explained in detail the...
Document layout analysis plays an important role in the area of Document Understanding. It is res... more Document layout analysis plays an important role in the area of Document Understanding. It is responsible for identifying and classifying the different components of digital documents. Currently, there is no universal algorithm that fits all types of digital documents. This work presents a novel approach for identifying tables, figures, isolated equations, and text regions in scientific papers using deep learning and computer vision techniques. Our proposed approach is a three-stage system: (i) Obtaining the spectrograms of the horizontal and vertical intensity histograms of segmented regions of interest. (ii) Labeling segmented regions of interest into text, table, and figures using a deep convolutional neural network classifier. (iii) Identifying isolated equations in text regions using Bag of Visual Words (BOVW) with Zernike moments. We built a new dataset composed of 11007 papers to perform the experiments, using two common segmentation metrics to evaluate our model: (1) Adjusted Rand Index (ARI) and (2) Variation of Information (VI). The proposed document layout analysis system reached an overall accuracy of 96.2685%, outperforming prior art with a less computational cost. INDEX TERMS Computer-Vision, Deep-Learning, document layout analysis, feature engineering, PDF.
2019 IEEE Latin American Conference on Computational Intelligence (LA-CCI), 2019
Microseism classification is primordial to easily identify what type of event we are facing at in... more Microseism classification is primordial to easily identify what type of event we are facing at in a possibly dangerous situation. However, labeling events is a hard and time-consuming task since it requires expert volcanologists to do this work. To alleviate the need for abundant labeled data, we propose a semi-supervised approach using the self-training algorithm. First, we extract several relevant microseisms features from the registers on the provided database, then we apply PCA to reduce redundancy on the features and finally we classify them using an SVM classifier. As a result of this methodology we show that although the accuracy of using a supervised scheme is still better than a semi-supervised one, if we allow a 10% of false positive rate, our approach achieves similar performance to supervised techniques with only 50% of labeled data. This demonstrates the potential of semi-supervised schemes.
Precision farming is one of the most trending topics nowadays and computer vision techniques are ... more Precision farming is one of the most trending topics nowadays and computer vision techniques are increasingly gaining momentum on this subject. On the other hand, in the cocoa industry, particularly in small farms, farmers still perform the classification of fresh cocoa beans with pulp in a traditional way, i.e. through their senses. In this work, we explain a new approach for cocoa beans with pulp classification, in order to aid in the process of removing cocoa beans pulp to efficiently estimate the quality of these beans. Our approach used morphological operations, k-means clustering, a bag of visual words as a feature extractor, and finally a support vector machine classifier. We achieved an AUC of 97.757% and an accuracy of 97.57% with a low false-positive rate of 2.46%, which demonstrates the viability of using computer vision for this task. We used a real-world dataset of 247 fresh cocoa beans images, that we collected and labeled with experienced cocoa farmers.
Latin American Conference on Computational Intelligence, 2019
Assistive technologies play an important role in improving the quality of life of people with dis... more Assistive technologies play an important role in improving the quality of life of people with disabilities. In this work, we developed a system for the retrieval of table information from digital documents for use in screen readers used by visually impaired people. The proposed methodology takes advantage of computer vision techniques with a deep learning approach to make documents accessible instead of the classical rule-based programming approach. We explained in detail the methodology that we used and how to objectively evaluate the approach through entropy, information gain, and purity metrics. The results show that our proposed methodology can be used to reduce the uncertainty experienced by visually impaired people when listening to the contents of tables in digital documents through screen readers. Our table information retrieval system presents two improvements compared with traditional approaches of tagging text-based portable document format (PDF) files. First, our approach does not require supervision by sighted people. Second, our system is capable of working with image-based as well as text-based PDFs.
This work evaluates the effect of the excitation signal used when measuring the absorption coeffi... more This work evaluates the effect of the excitation signal used when measuring the absorption coefficient on an impedance tube. This paper aims to offer some guidance on the selection of the excitation signal to perform sound absorption measurements using the impedance tube. Four possible excitation signals defined in ISO 10534-2 Standard were studied: two of them, random noise and two of them sine sweep signals. Some hypotheses tests were executed to assess the homogeneity of each measurement. The signal-to-noise ratio (SNR) was also computed to verify the measurement quality. The results show that the best performing approach was accomplished using a logarithmic sweep, giving more precise sound absorption curves with an SNR of 34.15 dB. Random signals reported similar SNR (greater than 30 dB) after executing an average with 100 repetitions.
A preliminary study is presented with the aim of modeling the thermal behavior of a passive build... more A preliminary study is presented with the aim of modeling the thermal behavior of a passive building that is ventilated merely with the promotion of natural ventilation, various models have been assessed by using the system identification process. The identification of a simplify and lite model of such thermal behavior is needed to later control the thermal comfort of the indoor environment through the building natural ventilation openings and window blinds. A physical-phenomena-based model using electrical analogies is built upon hypotheses allowed by the architectural features of the building. This helps analyze the interaction between the main elements of the physical domain, where the thermal behavior is only determined by the indoor air and concrete-slab temperatures. Three model approaches are examined with the help of the system identification toolbox: State space, Process models (linear and frequency domain), and Nonlinear representation. The nonlinear representation model i...
Spatial audio or 3D audio refers to a set of techniques whose main objective is to simulate sound... more Spatial audio or 3D audio refers to a set of techniques whose main objective is to simulate sound sources located in arbitrary positions in space. This is possible thanks to the Head-Related Transfer Functions (HRTFs), which model the anatomical characteristics of the subject and their interaction with the incident sound field. Since HRTFs are measured at discrete positions in space, interpolation techniques such as Bilinear or Triangular Spherical are necessary. However, both methods require uniform interpolation meshes. Hence, in this work, we propose a new methodology to HRTFs databases measured on non-uniform meshes for the bilinear and triangular spherical interpolation methods. Our results suggest that our proposed methodology for bilinear and triangular spherical interpolations can be used as a viable alternative for non-uniform meshes. We evaluate our approach in terms of spectral distortion, achieving similar results when compared to spline interpolation as a baseline.
El presente proyecto "DISENO E IMPLEMENTACION DE UN SISTEMA CONTROLADOR DE TEMPERATURA PID P... more El presente proyecto "DISENO E IMPLEMENTACION DE UN SISTEMA CONTROLADOR DE TEMPERATURA PID PARA LA UNIDAD AIR FLOW TEMPERATURE CONTROL SYSTEM MEDIANTE LA UTILIZACION DE LA HERRAMIENTA RTW (REAL TIME WORKSHOP) DE MATLAB"abarca el estudio de la herramienta RTW en donde se analizo la arquitectura, el algoritmo, el proceso y las etapas en el desarrollo de un modelo en tiempo real por medio del sistema SIMULINK® de MATLAB®, ademas se realizo una HMI la cual mediante subsistemas .mdl desarrollados en SIMULINK® permitieron la identificacion de la planta, el control PID experimental y la simulacion del control PID, utilizando para esto la tarjeta de adquisicion y generacion de datos Nationals Instruments PCI 6221, la misma que, luego de haberla configurado permitio interconectar la planta real con los subsistemas de la HMI desarrolla.
Digital documents are accessed by visually impaired people (VIP) through screen readers. Traditio... more Digital documents are accessed by visually impaired people (VIP) through screen readers. Traditionally, digital documents were translated to braille text, but screen readers have proved to be efficient for the acquisition of digital document knowledge by VIP. However, screen readers and other assistive technologies have significant limitations when there exist tables in digital documents such as portable document format (PDF). For instance, screen readers can not follow the correct reading sequence of the table based on its visual structure causing this content is inaccessible for VIP. In order to deal with this problem, in this work, we developed a system for the retrieval of table information from PDF documents for use in screen readers used by visually impaired people. The proposed methodology takes advantage of computer vision techniques with a deep learning approach to make documents accessible instead of the classical rule-based programming approach. We explained in detail the...
Document layout analysis plays an important role in the area of Document Understanding. It is res... more Document layout analysis plays an important role in the area of Document Understanding. It is responsible for identifying and classifying the different components of digital documents. Currently, there is no universal algorithm that fits all types of digital documents. This work presents a novel approach for identifying tables, figures, isolated equations, and text regions in scientific papers using deep learning and computer vision techniques. Our proposed approach is a three-stage system: (i) Obtaining the spectrograms of the horizontal and vertical intensity histograms of segmented regions of interest. (ii) Labeling segmented regions of interest into text, table, and figures using a deep convolutional neural network classifier. (iii) Identifying isolated equations in text regions using Bag of Visual Words (BOVW) with Zernike moments. We built a new dataset composed of 11007 papers to perform the experiments, using two common segmentation metrics to evaluate our model: (1) Adjusted Rand Index (ARI) and (2) Variation of Information (VI). The proposed document layout analysis system reached an overall accuracy of 96.2685%, outperforming prior art with a less computational cost. INDEX TERMS Computer-Vision, Deep-Learning, document layout analysis, feature engineering, PDF.
2019 IEEE Latin American Conference on Computational Intelligence (LA-CCI), 2019
Microseism classification is primordial to easily identify what type of event we are facing at in... more Microseism classification is primordial to easily identify what type of event we are facing at in a possibly dangerous situation. However, labeling events is a hard and time-consuming task since it requires expert volcanologists to do this work. To alleviate the need for abundant labeled data, we propose a semi-supervised approach using the self-training algorithm. First, we extract several relevant microseisms features from the registers on the provided database, then we apply PCA to reduce redundancy on the features and finally we classify them using an SVM classifier. As a result of this methodology we show that although the accuracy of using a supervised scheme is still better than a semi-supervised one, if we allow a 10% of false positive rate, our approach achieves similar performance to supervised techniques with only 50% of labeled data. This demonstrates the potential of semi-supervised schemes.
Precision farming is one of the most trending topics nowadays and computer vision techniques are ... more Precision farming is one of the most trending topics nowadays and computer vision techniques are increasingly gaining momentum on this subject. On the other hand, in the cocoa industry, particularly in small farms, farmers still perform the classification of fresh cocoa beans with pulp in a traditional way, i.e. through their senses. In this work, we explain a new approach for cocoa beans with pulp classification, in order to aid in the process of removing cocoa beans pulp to efficiently estimate the quality of these beans. Our approach used morphological operations, k-means clustering, a bag of visual words as a feature extractor, and finally a support vector machine classifier. We achieved an AUC of 97.757% and an accuracy of 97.57% with a low false-positive rate of 2.46%, which demonstrates the viability of using computer vision for this task. We used a real-world dataset of 247 fresh cocoa beans images, that we collected and labeled with experienced cocoa farmers.
Latin American Conference on Computational Intelligence, 2019
Assistive technologies play an important role in improving the quality of life of people with dis... more Assistive technologies play an important role in improving the quality of life of people with disabilities. In this work, we developed a system for the retrieval of table information from digital documents for use in screen readers used by visually impaired people. The proposed methodology takes advantage of computer vision techniques with a deep learning approach to make documents accessible instead of the classical rule-based programming approach. We explained in detail the methodology that we used and how to objectively evaluate the approach through entropy, information gain, and purity metrics. The results show that our proposed methodology can be used to reduce the uncertainty experienced by visually impaired people when listening to the contents of tables in digital documents through screen readers. Our table information retrieval system presents two improvements compared with traditional approaches of tagging text-based portable document format (PDF) files. First, our approach does not require supervision by sighted people. Second, our system is capable of working with image-based as well as text-based PDFs.
Uploads
Papers by Byron Acuña