Pub Date : 2018-10-01DOI: 10.1109/CONIITI.2018.8587078
Daniel Arguelles, Rafael Reyes, D. González, Osberth De Castro, P. Soares-Filho
This work presents the integration of a low-cost virtual reality video game system, called VRIMP, specifically designed in order to improve experimental procedures for impulsivity assessment in psychology, by creating an immersive experience for participants in such evaluation. Through the implementation from scratch of a client-server system, using free software tools and inexpensive commercial hardware, a platform was constructed allowing the participants to interact with special virtual reality games that implement specific measurable aspects of impulsive behavior. At the same time, VRIMP collects information on the actions performed by the user during the game session, for later analysis, and allows the connection of real-time supervisory clients. The VRIMP system has been submitted to an initial experts validation at the Psychology Laboratory of the University of San Buenaventura, in B020bl'l Colombia.
{"title":"VRIMP: Virtual Reality System for Impulsivity Assessment","authors":"Daniel Arguelles, Rafael Reyes, D. González, Osberth De Castro, P. Soares-Filho","doi":"10.1109/CONIITI.2018.8587078","DOIUrl":"https://doi.org/10.1109/CONIITI.2018.8587078","url":null,"abstract":"This work presents the integration of a low-cost virtual reality video game system, called VRIMP, specifically designed in order to improve experimental procedures for impulsivity assessment in psychology, by creating an immersive experience for participants in such evaluation. Through the implementation from scratch of a client-server system, using free software tools and inexpensive commercial hardware, a platform was constructed allowing the participants to interact with special virtual reality games that implement specific measurable aspects of impulsive behavior. At the same time, VRIMP collects information on the actions performed by the user during the game session, for later analysis, and allows the connection of real-time supervisory clients. The VRIMP system has been submitted to an initial experts validation at the Psychology Laboratory of the University of San Buenaventura, in B020bl'l Colombia.","PeriodicalId":292178,"journal":{"name":"2018 Congreso Internacional de Innovación y Tendencias en Ingeniería (CONIITI)","volume":"360 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122851752","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/CONIITI.2018.8587091
Lina J. Alfonso, J. P. Ospina, Jorge E. Ortiz
Ad hoc networks are decentralized systems that work without a fixed infrastructure and exhibit dynamic and stochastic behaviors. These conditions allow us to analyze them through properties like self-organization, adaptation and the use of the agent paradigm for managing and controlling the network. In this article, socially inspired computing is used to implement a register module for software agents through the concepts of institution and robustness; our goal is to build a software application that keeps working under the unexpected operating conditions of the ad hoc networks. IEEE FIPA protocols were used to design the agents and the communication channels. Three scenarios were developed, and failures in the systems were simulated. The results show the robustness of the module; it can recover itself from failures and allow the services to keep operating in acceptable conditions.
{"title":"A Register Module for Agents Communities Through Robustness Property in Ad Hoc Networks","authors":"Lina J. Alfonso, J. P. Ospina, Jorge E. Ortiz","doi":"10.1109/CONIITI.2018.8587091","DOIUrl":"https://doi.org/10.1109/CONIITI.2018.8587091","url":null,"abstract":"Ad hoc networks are decentralized systems that work without a fixed infrastructure and exhibit dynamic and stochastic behaviors. These conditions allow us to analyze them through properties like self-organization, adaptation and the use of the agent paradigm for managing and controlling the network. In this article, socially inspired computing is used to implement a register module for software agents through the concepts of institution and robustness; our goal is to build a software application that keeps working under the unexpected operating conditions of the ad hoc networks. IEEE FIPA protocols were used to design the agents and the communication channels. Three scenarios were developed, and failures in the systems were simulated. The results show the robustness of the module; it can recover itself from failures and allow the services to keep operating in acceptable conditions.","PeriodicalId":292178,"journal":{"name":"2018 Congreso Internacional de Innovación y Tendencias en Ingeniería (CONIITI)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128793675","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/CONIITI.2018.8587074
Renzo Arimborgo, Edson Espinoza, G. Power, G. Quispe, J. A. Merino
Because of frost in high Peruvian andean areas die alpacas (having reported more of 160 000 deaths in 2015), which can be avoided with suitable sheds. Faced with this reality, it was constructed of a prototype with solar air heating, which has been tested and calculated for use in these sheds with favorable results. During the investigation the heat transfer required by the shed was analyzed, for keep the alpacas in a closed environment at 10°C. It is worth mentioning that there are alpaca communities that supports temperatures of up to -20°C in the Altiplano area. The research was conducted in the district of San Juan de Tarucani in Arequipa and in the district of Ocuviri in Puno. The proposed model permits to keep the alpacas more protected in a warm environment using renewable energy with a system of heat storage for the low temperatures at the nights. As a result of the research, with the development of the proposed solar collector, it is possible to increase the temperature internal of the shed in an ideal range of 5°C.
{"title":"Solar Heating for Alpaca's Sheds in Peruvian Andean Region","authors":"Renzo Arimborgo, Edson Espinoza, G. Power, G. Quispe, J. A. Merino","doi":"10.1109/CONIITI.2018.8587074","DOIUrl":"https://doi.org/10.1109/CONIITI.2018.8587074","url":null,"abstract":"Because of frost in high Peruvian andean areas die alpacas (having reported more of 160 000 deaths in 2015), which can be avoided with suitable sheds. Faced with this reality, it was constructed of a prototype with solar air heating, which has been tested and calculated for use in these sheds with favorable results. During the investigation the heat transfer required by the shed was analyzed, for keep the alpacas in a closed environment at 10°C. It is worth mentioning that there are alpaca communities that supports temperatures of up to -20°C in the Altiplano area. The research was conducted in the district of San Juan de Tarucani in Arequipa and in the district of Ocuviri in Puno. The proposed model permits to keep the alpacas more protected in a warm environment using renewable energy with a system of heat storage for the low temperatures at the nights. As a result of the research, with the development of the proposed solar collector, it is possible to increase the temperature internal of the shed in an ideal range of 5°C.","PeriodicalId":292178,"journal":{"name":"2018 Congreso Internacional de Innovación y Tendencias en Ingeniería (CONIITI)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128030082","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/CONIITI.2018.8587094
José Manjarrés, Vitto Russo, J. Peñaranda, Mauricio Pardo
This paper describes a real-time human activity recognition and heart rate tracking system using ultra-low power wearables in a mobile platform. The mobile application shows both real-time and historical data of activities performed by the user along with the average heart rate for each activity. Random Forest and k-Nearest Neighbors algorithms were used to classify, showing general accuracies of 97.3% and 98.6%, respectively, with only eight features. The presented heart rate monitor is compared with an Apple Watch series 3 and they were demonstrated to display statistically equal values, confirming the reliability of the built monitor. Both wearables have a diameter of 4.5cm and consume currents below 10mA on transmission events and are powered with coin cell batteries, setting feasible characteristics for the need of modern wearable systems.
{"title":"Human Activity and Heart Rate Monitoring System in a Mobile Platform","authors":"José Manjarrés, Vitto Russo, J. Peñaranda, Mauricio Pardo","doi":"10.1109/CONIITI.2018.8587094","DOIUrl":"https://doi.org/10.1109/CONIITI.2018.8587094","url":null,"abstract":"This paper describes a real-time human activity recognition and heart rate tracking system using ultra-low power wearables in a mobile platform. The mobile application shows both real-time and historical data of activities performed by the user along with the average heart rate for each activity. Random Forest and k-Nearest Neighbors algorithms were used to classify, showing general accuracies of 97.3% and 98.6%, respectively, with only eight features. The presented heart rate monitor is compared with an Apple Watch series 3 and they were demonstrated to display statistically equal values, confirming the reliability of the built monitor. Both wearables have a diameter of 4.5cm and consume currents below 10mA on transmission events and are powered with coin cell batteries, setting feasible characteristics for the need of modern wearable systems.","PeriodicalId":292178,"journal":{"name":"2018 Congreso Internacional de Innovación y Tendencias en Ingeniería (CONIITI)","volume":"158 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115686427","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/CONIITI.2018.8587057
Andrés Baquero-Larriva, Gustavo Alvarez-Coello
Fuel consumption in transport vehicles is one of the main sources of CO2 emissions worldwide. These kind of vehicles in the most of developing countries, do not have On Board Diagnostics (OBD) systems or similar interfaces, so is essential to develop low cost, easy to build and reliable fuel consumption measurement systems. The present research focused on the development of a data acquisition system, which integrated different devices and technologies. This system uses an electronic printed circuit board (PCB) with a microcontroller that registers instant and real time measurements of diesel consumption and monitors the geographical position of the vehicle using GPS technology.
{"title":"Low Cost Data Adquisition System to Register Fuel Consumption in Diesel Engine Vehicles","authors":"Andrés Baquero-Larriva, Gustavo Alvarez-Coello","doi":"10.1109/CONIITI.2018.8587057","DOIUrl":"https://doi.org/10.1109/CONIITI.2018.8587057","url":null,"abstract":"Fuel consumption in transport vehicles is one of the main sources of CO2 emissions worldwide. These kind of vehicles in the most of developing countries, do not have On Board Diagnostics (OBD) systems or similar interfaces, so is essential to develop low cost, easy to build and reliable fuel consumption measurement systems. The present research focused on the development of a data acquisition system, which integrated different devices and technologies. This system uses an electronic printed circuit board (PCB) with a microcontroller that registers instant and real time measurements of diesel consumption and monitors the geographical position of the vehicle using GPS technology.","PeriodicalId":292178,"journal":{"name":"2018 Congreso Internacional de Innovación y Tendencias en Ingeniería (CONIITI)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131773042","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/CONIITI.2018.8587101
H. Bermudez, R. Sanchez-Iborra, W. Campo, J. Arciniegas, M. Cano
The use of adaptive transmission techniques through HTTP for the reproduction of video services is becoming more and more popular. Currently, there are problems of interoperability due to the large number of devices with different characteristics, which also used different techniques of adaptive transmission. Therefore, telecommunications service providers are looking for alternatives that allow them to support all this diversity of devices and technologies, in order to offer their users higher quality services. In this paper, we present a comparative performance evaluation of the main adaptive streaming techniques for Live Video Streaming services using QoE (Quality of Experience) as the key performance indicator. The study is carried out in an emulated LTE (Long Term Evolution) network composed by real devices (video server and client) interacting with an NS-3 simulated LTE network. We show for the constructed emulation scenario, the technology that presents the best performance in terms of delays and user mobility.
{"title":"Comparative Study of Adaptive Streaming Techniques in LTE Networks Using QoE Metrics","authors":"H. Bermudez, R. Sanchez-Iborra, W. Campo, J. Arciniegas, M. Cano","doi":"10.1109/CONIITI.2018.8587101","DOIUrl":"https://doi.org/10.1109/CONIITI.2018.8587101","url":null,"abstract":"The use of adaptive transmission techniques through HTTP for the reproduction of video services is becoming more and more popular. Currently, there are problems of interoperability due to the large number of devices with different characteristics, which also used different techniques of adaptive transmission. Therefore, telecommunications service providers are looking for alternatives that allow them to support all this diversity of devices and technologies, in order to offer their users higher quality services. In this paper, we present a comparative performance evaluation of the main adaptive streaming techniques for Live Video Streaming services using QoE (Quality of Experience) as the key performance indicator. The study is carried out in an emulated LTE (Long Term Evolution) network composed by real devices (video server and client) interacting with an NS-3 simulated LTE network. We show for the constructed emulation scenario, the technology that presents the best performance in terms of delays and user mobility.","PeriodicalId":292178,"journal":{"name":"2018 Congreso Internacional de Innovación y Tendencias en Ingeniería (CONIITI)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132336155","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/CONIITI.2018.8587100
Victor Sanchez, Ernesto Iturrizaga, J. León, C. del Carpio
This work proposes an algorithm for the detection of the intestinal parasite Giardia Lamblia from digital images obtained through a digital camera and an optical microscope. Its purpose is to reduce the time of visual inspection analysis of the sample made by the specialist in laboratory. The proposed algorithm converts the acquired RGB images to the HSV colour space. First, the saturation component is filtered using a Gaussian filter, in order to reduce the noise and standardize the areas of interest. The filtered image is thresholdized using a fixed threshold value, in order to segment the objects of interest. Then, using a labelling algorithm, a filtering is performed by object size, in order to eliminate those that do not comply with the dimensions of the parasite. Therefore, the edges are highlighted by a Canny filter, to finally apply the Hough transform and detect the morphology of the object and its physical dimensions. With this information it will be possible to validate if the object satisfies the conditions of a Giardia Lamblia parasite. The proposed method achieved a specificity of 86% and a sensitivity of 67%, processing each image in an average time of less than 2 seconds. The results were obtained from a universe of 30 images.
{"title":"An Algorithm for Giardia Lamblia Detection in Digital Images Acquired Through an Optical Microscope","authors":"Victor Sanchez, Ernesto Iturrizaga, J. León, C. del Carpio","doi":"10.1109/CONIITI.2018.8587100","DOIUrl":"https://doi.org/10.1109/CONIITI.2018.8587100","url":null,"abstract":"This work proposes an algorithm for the detection of the intestinal parasite Giardia Lamblia from digital images obtained through a digital camera and an optical microscope. Its purpose is to reduce the time of visual inspection analysis of the sample made by the specialist in laboratory. The proposed algorithm converts the acquired RGB images to the HSV colour space. First, the saturation component is filtered using a Gaussian filter, in order to reduce the noise and standardize the areas of interest. The filtered image is thresholdized using a fixed threshold value, in order to segment the objects of interest. Then, using a labelling algorithm, a filtering is performed by object size, in order to eliminate those that do not comply with the dimensions of the parasite. Therefore, the edges are highlighted by a Canny filter, to finally apply the Hough transform and detect the morphology of the object and its physical dimensions. With this information it will be possible to validate if the object satisfies the conditions of a Giardia Lamblia parasite. The proposed method achieved a specificity of 86% and a sensitivity of 67%, processing each image in an average time of less than 2 seconds. The results were obtained from a universe of 30 images.","PeriodicalId":292178,"journal":{"name":"2018 Congreso Internacional de Innovación y Tendencias en Ingeniería (CONIITI)","volume":"98 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133166488","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/CONIITI.2018.8587082
M. Chazi-Solis, C. Cajamarca-Bueno, E. Pinos-Vélez, V. Robles-Bykbaev, C. Chacón
According to estimations of the World Health Organization (WHO), approximately 4.5 million persons will become blind due to the primary Glaucoma. For these reasons, in this paper, we present a simple approach to estimate the risk of suffering from glaucoma that presents a patient. Our proposal relies on the automatic analysis of Optical Coherence Tomography (OCT) images through the calculus of nerve fiber and ganglionary vessels areas. To this aim, our system separates the two areas mentioned before through computer vision techniques and the Newton Interpolation Method. Once the two areas are separated, the system determines the size of each of them. This proposal was validated through an experiment carried out with OCT 30 images (per eye). The results are encouraging, and the suggestions provided by the system match with medical diagnosis.
{"title":"A Computer Vision Approach Based on the Retinal Nerve Fiber Thickness Analysis to Estimate the Risk of Suffering Glaucoma","authors":"M. Chazi-Solis, C. Cajamarca-Bueno, E. Pinos-Vélez, V. Robles-Bykbaev, C. Chacón","doi":"10.1109/CONIITI.2018.8587082","DOIUrl":"https://doi.org/10.1109/CONIITI.2018.8587082","url":null,"abstract":"According to estimations of the World Health Organization (WHO), approximately 4.5 million persons will become blind due to the primary Glaucoma. For these reasons, in this paper, we present a simple approach to estimate the risk of suffering from glaucoma that presents a patient. Our proposal relies on the automatic analysis of Optical Coherence Tomography (OCT) images through the calculus of nerve fiber and ganglionary vessels areas. To this aim, our system separates the two areas mentioned before through computer vision techniques and the Newton Interpolation Method. Once the two areas are separated, the system determines the size of each of them. This proposal was validated through an experiment carried out with OCT 30 images (per eye). The results are encouraging, and the suggestions provided by the system match with medical diagnosis.","PeriodicalId":292178,"journal":{"name":"2018 Congreso Internacional de Innovación y Tendencias en Ingeniería (CONIITI)","volume":"323 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116609467","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/CONIITI.2018.8587090
Rafael Enrique Rodriguez-Rodriguez, Andrés Felipe Quevedo Vega, A. F. Sánchez, Alexandra López, Jaime Fernando Pérez
Nowadays IT audit is a priority for large, medium and small companies, for this reason techniques and methodologies have been developed to manage IT audits, in order to avoid the spread of threats and risks within organizations. For the project development, methodologies and good practices were investigated in order to manage IT audits, looking for the most aligned to the collection of evidences as well as the execution of the tests that are done to the controls in the systems of information, the foregoing aims to make a diagnostic selection of the available methodologies for IT audit evidence collecting, as well as identify the characteristics of the applications used in the management of IT audits also called CAAT tools with their functional aspects to categorize their effectiveness against applied methodologies, for the analysis of the results of the methodologies and thus propose a model to automate the collection of documentary evidence in compliance tests in IT audit, based on a methodology.
{"title":"Design of an Automation Model for Taking Documentary Evidence of Compliance Tests of the IT Audit","authors":"Rafael Enrique Rodriguez-Rodriguez, Andrés Felipe Quevedo Vega, A. F. Sánchez, Alexandra López, Jaime Fernando Pérez","doi":"10.1109/CONIITI.2018.8587090","DOIUrl":"https://doi.org/10.1109/CONIITI.2018.8587090","url":null,"abstract":"Nowadays IT audit is a priority for large, medium and small companies, for this reason techniques and methodologies have been developed to manage IT audits, in order to avoid the spread of threats and risks within organizations. For the project development, methodologies and good practices were investigated in order to manage IT audits, looking for the most aligned to the collection of evidences as well as the execution of the tests that are done to the controls in the systems of information, the foregoing aims to make a diagnostic selection of the available methodologies for IT audit evidence collecting, as well as identify the characteristics of the applications used in the management of IT audits also called CAAT tools with their functional aspects to categorize their effectiveness against applied methodologies, for the analysis of the results of the methodologies and thus propose a model to automate the collection of documentary evidence in compliance tests in IT audit, based on a methodology.","PeriodicalId":292178,"journal":{"name":"2018 Congreso Internacional de Innovación y Tendencias en Ingeniería (CONIITI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130476491","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/CONIITI.2018.8587096
Álvaro Ramiro Hernández Millán, M. Mendoza-Moreno, Larry Mauricio Portocarrero López, Alexander Castro-Romero
The growing use of machine learning algorithms for image classification process using open source libraries promotes the identification of data sets to be applied on different contexts. In this way, the most outstanding algorithms for image classification with machine learning supervised techniques were implemented and their accuracy level was compared in an uncommon context, such as identification documents (combining text and images) of a university institution as a study case. The used dataset has a high complexity level in terms of design, diversity and population density. The study has provided outstanding results such as: 1) the implementation of the Transfer Learning / Image Retraining technique for which the expected performance was obtained in shorter times compared to the other techniques or algorithms studied, and 2) the need to implement a previous classifier whose objective is to refine the dataset to promote higher precision levels for new instances (new documents to classify).
{"title":"Comparative Study of Machine Learning Supervised Techniques for Image Classification Using an Institutional Identification Documents Dataset","authors":"Álvaro Ramiro Hernández Millán, M. Mendoza-Moreno, Larry Mauricio Portocarrero López, Alexander Castro-Romero","doi":"10.1109/CONIITI.2018.8587096","DOIUrl":"https://doi.org/10.1109/CONIITI.2018.8587096","url":null,"abstract":"The growing use of machine learning algorithms for image classification process using open source libraries promotes the identification of data sets to be applied on different contexts. In this way, the most outstanding algorithms for image classification with machine learning supervised techniques were implemented and their accuracy level was compared in an uncommon context, such as identification documents (combining text and images) of a university institution as a study case. The used dataset has a high complexity level in terms of design, diversity and population density. The study has provided outstanding results such as: 1) the implementation of the Transfer Learning / Image Retraining technique for which the expected performance was obtained in shorter times compared to the other techniques or algorithms studied, and 2) the need to implement a previous classifier whose objective is to refine the dataset to promote higher precision levels for new instances (new documents to classify).","PeriodicalId":292178,"journal":{"name":"2018 Congreso Internacional de Innovación y Tendencias en Ingeniería (CONIITI)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121896919","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：