Pub Date : 2021-04-28DOI: 10.1109/AIMS52415.2021.9466010
M. F. Hirzi, S. Efendi, R. Sembiring
The face is the front part of a human expression comprising the eyes, nose, lips, cheeks, forehead, and chin. These characters are uniquely placed according to a human's pattern of face. The Viola-Jones algorithm is used to recognize and detect objects, including this human face. It consists of several stages, such as Haar-like Filter, Integral Image, Adaboost Algorithm, and Cascade. Haar-Like filter is used to determine feature values from images containing certain objects. Furthermore, integral image helps to find the feature value to quicken the calculation process. Adaboost algorithm processes feature selection by determining the threshold value in order to determine the existing object. Meanwhile, cascade performs an image selection process that contains or excludes objects with large amounts of test data. It directly discards the figure when no objects are detected to produce images containing objects. This study is a literature review on facial recognition using the Viola-Jones algorithm. It contributes to the search for the suitability of using the Viola-Jones Algorithm in certain cases. The research contribution also lies in the researcher's idea for future research, namely testing the Viola-Jones algorithm in recognizing objects other than facial images. Furthermore, five studies are analyzed and described the application of the Viola-Jones algorithm for facial recognition with their respective advantages. The first study had a very good accuracy level of 85%–95% in detecting faces. The second study had accuracy, precision, recall, and achievement times of 0.74, 0.73, 0.76, and 15 seconds in recognizing a person's emotions through facial expressions. Meanwhile, the third study had a very good accuracy level of 94.5% in recognizing faces that are 1 meter away.
{"title":"Literature Study of Face Recognition using The Viola-Jones Algorithm","authors":"M. F. Hirzi, S. Efendi, R. Sembiring","doi":"10.1109/AIMS52415.2021.9466010","DOIUrl":"https://doi.org/10.1109/AIMS52415.2021.9466010","url":null,"abstract":"The face is the front part of a human expression comprising the eyes, nose, lips, cheeks, forehead, and chin. These characters are uniquely placed according to a human's pattern of face. The Viola-Jones algorithm is used to recognize and detect objects, including this human face. It consists of several stages, such as Haar-like Filter, Integral Image, Adaboost Algorithm, and Cascade. Haar-Like filter is used to determine feature values from images containing certain objects. Furthermore, integral image helps to find the feature value to quicken the calculation process. Adaboost algorithm processes feature selection by determining the threshold value in order to determine the existing object. Meanwhile, cascade performs an image selection process that contains or excludes objects with large amounts of test data. It directly discards the figure when no objects are detected to produce images containing objects. This study is a literature review on facial recognition using the Viola-Jones algorithm. It contributes to the search for the suitability of using the Viola-Jones Algorithm in certain cases. The research contribution also lies in the researcher's idea for future research, namely testing the Viola-Jones algorithm in recognizing objects other than facial images. Furthermore, five studies are analyzed and described the application of the Viola-Jones algorithm for facial recognition with their respective advantages. The first study had a very good accuracy level of 85%–95% in detecting faces. The second study had accuracy, precision, recall, and achievement times of 0.74, 0.73, 0.76, and 15 seconds in recognizing a person's emotions through facial expressions. Meanwhile, the third study had a very good accuracy level of 94.5% in recognizing faces that are 1 meter away.","PeriodicalId":299121,"journal":{"name":"2021 International Conference on Artificial Intelligence and Mechatronics Systems (AIMS)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124413149","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 : 2021-04-28DOI: 10.1109/AIMS52415.2021.9466022
Md. Iqbal Hossain, Maqsudur Rahman, M. T. Ahmed, Md. Saifur Rahman, A. Z. M. T. Islam
The only way to provide feedback about a product is through reviews. When a new shopper proceeds to an online shop to purchase a product but does not have adequate time to study the reviews provided by other shoppers to get an opinion about the product, the shopper determines whether to buy the product or not on the number rating. Through reviews, shoppers can acquaint everyone about the product's quality and the manufacturer can enhance their products and business by interpreting that review. However, manufacturers demand a number review more than a text review for business analysis. This paper represents a machine learning based model for predicting the number rating from written text for Bangla product review. This study performs on a dataset collected manually from Daraz.com.bd, a Bangladeshi leading e-commerce shop. We have implemented Support Vector Machine (SVM), Random Forest, XGBoost, and Logistic Regression with Term Frequency-Inverse Document Frequency (TF-IDF) Vectorizer on our collected dataset and record all the performance metrics like accuracy, precision, recall and f1-score. From these above four algorithms, SVM showed more outstanding results than others in terms of performance metrics. SVM achieved 90% accuracy on the applied dataset. The other SVM performance metrics are 0.90, 0.92, and 0.91 for precision, recall and f1-score respectively.
提供产品反馈的唯一途径就是通过评论。当一个新购物者进入网上商店购买产品，但没有足够的时间研究其他购物者提供的评论以获得对产品的意见时，购物者根据数字评级决定是否购买该产品。通过评论，购物者可以让每个人都了解产品的质量，制造商可以通过解释评论来提高他们的产品和业务。然而，对于业务分析，制造商更需要数字审查而不是文本审查。本文提出了一种基于机器学习的模型，用于从孟加拉产品评论的书面文本中预测数字评级。这项研究的数据集是从Daraz.com.bd手工收集的，这是一家孟加拉国领先的电子商务商店。我们在收集的数据集上实现了支持向量机(SVM)、随机森林(Random Forest)、XGBoost和具有Term Frequency- inverse Document Frequency (TF-IDF)矢量器的逻辑回归，并记录了所有的性能指标，如准确性、精密度、召回率和f1-score。在以上四种算法中，SVM在性能指标方面表现出比其他算法更突出的结果。SVM在应用数据集上的准确率达到90%。其他支持向量机性能指标分别为0.90,0.92和0.91的精度，召回率和f1-score。
{"title":"Rating Prediction of Product Reviews of Bangla Language using Machine Learning Algorithms","authors":"Md. Iqbal Hossain, Maqsudur Rahman, M. T. Ahmed, Md. Saifur Rahman, A. Z. M. T. Islam","doi":"10.1109/AIMS52415.2021.9466022","DOIUrl":"https://doi.org/10.1109/AIMS52415.2021.9466022","url":null,"abstract":"The only way to provide feedback about a product is through reviews. When a new shopper proceeds to an online shop to purchase a product but does not have adequate time to study the reviews provided by other shoppers to get an opinion about the product, the shopper determines whether to buy the product or not on the number rating. Through reviews, shoppers can acquaint everyone about the product's quality and the manufacturer can enhance their products and business by interpreting that review. However, manufacturers demand a number review more than a text review for business analysis. This paper represents a machine learning based model for predicting the number rating from written text for Bangla product review. This study performs on a dataset collected manually from Daraz.com.bd, a Bangladeshi leading e-commerce shop. We have implemented Support Vector Machine (SVM), Random Forest, XGBoost, and Logistic Regression with Term Frequency-Inverse Document Frequency (TF-IDF) Vectorizer on our collected dataset and record all the performance metrics like accuracy, precision, recall and f1-score. From these above four algorithms, SVM showed more outstanding results than others in terms of performance metrics. SVM achieved 90% accuracy on the applied dataset. The other SVM performance metrics are 0.90, 0.92, and 0.91 for precision, recall and f1-score respectively.","PeriodicalId":299121,"journal":{"name":"2021 International Conference on Artificial Intelligence and Mechatronics Systems (AIMS)","volume":"473 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125837153","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 : 2021-04-28DOI: 10.1109/AIMS52415.2021.9466063
Emilliano, C. Chakrabarty, A. Basri, A. Ramasamy, N. Suhendi
The FPGA (Field Programmable Gate Array) technology is widely used today for signal processing and control owing to its fast digital processing capabilities. Using this, a test-bed high-speed data acquisition system that combines a commercial FPGA board (ML405) with the ADC in PIC micro-controller 16F877A that has 8 bit in resolution, sampling rate of 20 MS/s and bandwidth of 10 MHz for counting very high speed transient signals has been developed and successfully tested in the lab. This system enables direct measurement and counting of the transient signals at 50–100 ns pulse width at a sampling frequency 20MHz from Pico pulse generator as simulator of Partial Discharge signals on the site. All the results are shown to proof the concept for detecting partial discharge with pulse width of 5 ns in high power underground cables. The advantage of this system is that it can easily be deployed and count simulated Partial Discharge signal without the use of an oscilloscope and PC. The work in this paper comprises of using VHDL programming in FPGA to capture and discriminate the high-speed transient signals that has been digitized by the ADC in PIC micro-controller. In implementation of the test in the laboratory for the detection circuit, it is shown that it can detect the amount of impulses quite accurately. This is shown in PD detector system whereby LCD reads 84,746 impulses per second when it was set at 84.746 KHz repetitive using Pico pulse generator. The result show that the output peak detector can detect peak signal from input signal of the ADC when the pulse width of PD signal more than 30 ns using the ADC of the PIC Microcontroller 16F877A. Several features such as counting and discrimination between pulses are integrated in the system are also shown. This concept will be used in the future to detect real partial discharge generated in high power underground cables in the field.
{"title":"Embedded System Real Time Data Acquisition System Using FPGA Technology for Detection and Counting of PD Signal from PICO Pulse Generator","authors":"Emilliano, C. Chakrabarty, A. Basri, A. Ramasamy, N. Suhendi","doi":"10.1109/AIMS52415.2021.9466063","DOIUrl":"https://doi.org/10.1109/AIMS52415.2021.9466063","url":null,"abstract":"The FPGA (Field Programmable Gate Array) technology is widely used today for signal processing and control owing to its fast digital processing capabilities. Using this, a test-bed high-speed data acquisition system that combines a commercial FPGA board (ML405) with the ADC in PIC micro-controller 16F877A that has 8 bit in resolution, sampling rate of 20 MS/s and bandwidth of 10 MHz for counting very high speed transient signals has been developed and successfully tested in the lab. This system enables direct measurement and counting of the transient signals at 50–100 ns pulse width at a sampling frequency 20MHz from Pico pulse generator as simulator of Partial Discharge signals on the site. All the results are shown to proof the concept for detecting partial discharge with pulse width of 5 ns in high power underground cables. The advantage of this system is that it can easily be deployed and count simulated Partial Discharge signal without the use of an oscilloscope and PC. The work in this paper comprises of using VHDL programming in FPGA to capture and discriminate the high-speed transient signals that has been digitized by the ADC in PIC micro-controller. In implementation of the test in the laboratory for the detection circuit, it is shown that it can detect the amount of impulses quite accurately. This is shown in PD detector system whereby LCD reads 84,746 impulses per second when it was set at 84.746 KHz repetitive using Pico pulse generator. The result show that the output peak detector can detect peak signal from input signal of the ADC when the pulse width of PD signal more than 30 ns using the ADC of the PIC Microcontroller 16F877A. Several features such as counting and discrimination between pulses are integrated in the system are also shown. This concept will be used in the future to detect real partial discharge generated in high power underground cables in the field.","PeriodicalId":299121,"journal":{"name":"2021 International Conference on Artificial Intelligence and Mechatronics Systems (AIMS)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123798520","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 : 2021-04-28DOI: 10.1109/AIMS52415.2021.9466049
Muhammad Moiz, Hazique Malik, Muhammad Bilal, Noman Naseer
Over the past many years the popularity of robotic workers has seen a tremendous surge. Several tasks which were previously considered insurmountable are able to be performed by robots efficiently, with much ease. This is mainly due to the advances made in the field of control systems and artificial intelligence in recent years. Lately, we have seen Reinforcement Learning (RL) capture the spotlight, in the field of robotics. Instead of explicitly specifying the solution of a particular task, RL enables the robot (agent) to explore its environment and through trial and error choose the appropriate response. In this paper, a comparative analysis of biasing techniques for the Q-biased softmax regression (QBIASSR) algorithm has been presented. In QBIASSR, decision-making for un-explored states depends upon the set of previously explored states. This algorithm improves the learning process when the robot reaches unexplored states. A vector bias(s) is calculated on the basis of variable values of experienced states and added to the Q-value function for action selection. To obtain the optimized reward, different techniques to calculate bias(s) are adopted. The performance of all the techniques has been evaluated and compared for obstacle avoidance in the case of a mobile robot. In the end, we have demonstrated that the cumulative reward generated by the technique proposed in our paper is at least 2 times greater than the baseline.
{"title":"A Comparative Analysis of Multiple Biasing Techniques for $Q_{biased}$ Softmax Regression Algorithm","authors":"Muhammad Moiz, Hazique Malik, Muhammad Bilal, Noman Naseer","doi":"10.1109/AIMS52415.2021.9466049","DOIUrl":"https://doi.org/10.1109/AIMS52415.2021.9466049","url":null,"abstract":"Over the past many years the popularity of robotic workers has seen a tremendous surge. Several tasks which were previously considered insurmountable are able to be performed by robots efficiently, with much ease. This is mainly due to the advances made in the field of control systems and artificial intelligence in recent years. Lately, we have seen Reinforcement Learning (RL) capture the spotlight, in the field of robotics. Instead of explicitly specifying the solution of a particular task, RL enables the robot (agent) to explore its environment and through trial and error choose the appropriate response. In this paper, a comparative analysis of biasing techniques for the Q-biased softmax regression (QBIASSR) algorithm has been presented. In QBIASSR, decision-making for un-explored states depends upon the set of previously explored states. This algorithm improves the learning process when the robot reaches unexplored states. A vector bias(s) is calculated on the basis of variable values of experienced states and added to the Q-value function for action selection. To obtain the optimized reward, different techniques to calculate bias(s) are adopted. The performance of all the techniques has been evaluated and compared for obstacle avoidance in the case of a mobile robot. In the end, we have demonstrated that the cumulative reward generated by the technique proposed in our paper is at least 2 times greater than the baseline.","PeriodicalId":299121,"journal":{"name":"2021 International Conference on Artificial Intelligence and Mechatronics Systems (AIMS)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126599883","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 : 2021-04-28DOI: 10.1109/AIMS52415.2021.9466059
Iswahyuli, A. Bustamam, Arry Yanuar, W. Mangunwardoyo
Prediction task of drug-target interactions (DTI) is an important step of drug development and repositioning. Experimental identification of drugs and target interactions is expensive and time-consuming. Therefore, predictive drug-target interactions with computational approaches are being developed to alleviate work in drug development. In recent years, many computational approaches aimed at predicting drug-target interactions have been developed. One of the most popular models for predicting drug interactions and targets in recent times is the machine learning-based approach and homogeneous network information. However, the accuracy and efficiency of the methods used still need to be improved. Therefore, this research aims to propose a deep learning-based prediction model for DTI implemented in heterogeneous networks. We use 12,015 nodes and 1,895,445 edges that extract from several databases to build the heterogeneous network. The model of DTI prediction that we proposed implements the random walk with restart (RWR) algorithm to build a heterogeneous network of drug and protein targets, and utilizes diffusion component analysis (DCA) algorithm to obtain low-dimensional vectors. Furthermore, a one-dimensional convolutional neural network (1D-CNN) was used as a predictive model between drug and target. The results show that our proposed model provides good performance with a mean score of AUROC was 0.9332, and a mean score of AUPR was 0.9402.
{"title":"One-Dimensional Convolutional Neural Network Method as The Predicting Model for Interactions Between Drug and Protein on Heterogeneous Network","authors":"Iswahyuli, A. Bustamam, Arry Yanuar, W. Mangunwardoyo","doi":"10.1109/AIMS52415.2021.9466059","DOIUrl":"https://doi.org/10.1109/AIMS52415.2021.9466059","url":null,"abstract":"Prediction task of drug-target interactions (DTI) is an important step of drug development and repositioning. Experimental identification of drugs and target interactions is expensive and time-consuming. Therefore, predictive drug-target interactions with computational approaches are being developed to alleviate work in drug development. In recent years, many computational approaches aimed at predicting drug-target interactions have been developed. One of the most popular models for predicting drug interactions and targets in recent times is the machine learning-based approach and homogeneous network information. However, the accuracy and efficiency of the methods used still need to be improved. Therefore, this research aims to propose a deep learning-based prediction model for DTI implemented in heterogeneous networks. We use 12,015 nodes and 1,895,445 edges that extract from several databases to build the heterogeneous network. The model of DTI prediction that we proposed implements the random walk with restart (RWR) algorithm to build a heterogeneous network of drug and protein targets, and utilizes diffusion component analysis (DCA) algorithm to obtain low-dimensional vectors. Furthermore, a one-dimensional convolutional neural network (1D-CNN) was used as a predictive model between drug and target. The results show that our proposed model provides good performance with a mean score of AUROC was 0.9332, and a mean score of AUPR was 0.9402.","PeriodicalId":299121,"journal":{"name":"2021 International Conference on Artificial Intelligence and Mechatronics Systems (AIMS)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115199201","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 : 2021-04-28DOI: 10.1109/AIMS52415.2021.9466075
D. P. Wicaksa, B. Trilaksono, E. Hidayat
Ganesh Blue Underwater Glider is an autonomous underwater vehicle that moves by changing its buoyancy and attitude to perform gliding movement. To perform a maritime exploration mission, integration between vehicle, control station, and website is required. Ganesh Blue Underwater Glider consists of several subsystems which handle algorithm, actuator control, communications, and sensor reading while the control station manages and monitors mission execution and also acts as a connector between Ganesh Blue Underwater Glider and website. This paper focuses on designing architecture and integration between Ganesh Blue Underwater Glider, Ground Control Station, and website. By separating high-level and low-level processes, the designed system able to complete sensor data acquisition module with average time 333 ms and standard deviation of 8.188 ms. Navigation, guidance, and guidance module is completed from 7.428 ms to 16.532 ms with an average of 10.687 ms.
Ganesh Blue Underwater Glider是一种自主水下航行器，通过改变其浮力和姿态来进行滑翔运动。为了执行海上探测任务，需要将车辆、控制站和网站集成在一起。Ganesh Blue水下滑翔机由几个子系统组成，这些子系统处理算法、执行器控制、通信和传感器读取，同时控制站管理和监视任务执行，并充当Ganesh Blue水下滑翔机和网站之间的连接器。本文重点研究了Ganesh Blue水下滑翔机、地面控制站和网站的体系结构设计和集成。通过对高级和低级流程的分离，设计的系统能够完成传感器数据采集模块，平均时间为333 ms，标准差为8.188 ms。导航、制导和制导模块完成时间为7.428 ms ~ 16.532 ms，平均为10.687 ms。
{"title":"System Design and Integration for Exploration Mission on Autonomous Underwater Glider","authors":"D. P. Wicaksa, B. Trilaksono, E. Hidayat","doi":"10.1109/AIMS52415.2021.9466075","DOIUrl":"https://doi.org/10.1109/AIMS52415.2021.9466075","url":null,"abstract":"Ganesh Blue Underwater Glider is an autonomous underwater vehicle that moves by changing its buoyancy and attitude to perform gliding movement. To perform a maritime exploration mission, integration between vehicle, control station, and website is required. Ganesh Blue Underwater Glider consists of several subsystems which handle algorithm, actuator control, communications, and sensor reading while the control station manages and monitors mission execution and also acts as a connector between Ganesh Blue Underwater Glider and website. This paper focuses on designing architecture and integration between Ganesh Blue Underwater Glider, Ground Control Station, and website. By separating high-level and low-level processes, the designed system able to complete sensor data acquisition module with average time 333 ms and standard deviation of 8.188 ms. Navigation, guidance, and guidance module is completed from 7.428 ms to 16.532 ms with an average of 10.687 ms.","PeriodicalId":299121,"journal":{"name":"2021 International Conference on Artificial Intelligence and Mechatronics Systems (AIMS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128330502","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 : 2021-04-28DOI: 10.1109/AIMS52415.2021.9466084
Nabeeha Ehsan Mughal, Khurram Khalil, Muhammad Jawad Khan
The ever-increasing human-machine interaction and advancement in socio-technical systems have made it essential to analyze the vital human factors such as mental workload, vigilance, fatigue, stress, etc., via monitoring brain states. Similarly, brain signals are becoming paramount for rehabilitation and assistive purposes in fields such as brain- computer interface (BCI), closed-loop neuromodulation for neurological disorders, etc. The complex, non-stationary, and very low signal-to-noise ratio of brain signals poses a significant challenge for researchers to design robust and reliable BCI systems outside the laboratory environment. In this study, we present novel recurrence plots (RPs) based on convolutional neural network and long short term memory (CNN-LSTM) algorithm for four class functional near-infrared spectroscopy (fNIRS) BCI. The acquired brain signals are projected into a non-linear dimension with RPs and fed into the CNN, which extracts the important features. Then LSTM learns the chronological and time-dependent relations. The average accuracy achieved with the proposed model is 77.7%, while the maximum accuracy is 85.9%. The results confirm the viability of RPs based deep learning algorithms for successful BCI systems.
{"title":"fNIRS Based Multi-Class Mental Workload Classification Using Recurrence Plots and CNN-LSTM","authors":"Nabeeha Ehsan Mughal, Khurram Khalil, Muhammad Jawad Khan","doi":"10.1109/AIMS52415.2021.9466084","DOIUrl":"https://doi.org/10.1109/AIMS52415.2021.9466084","url":null,"abstract":"The ever-increasing human-machine interaction and advancement in socio-technical systems have made it essential to analyze the vital human factors such as mental workload, vigilance, fatigue, stress, etc., via monitoring brain states. Similarly, brain signals are becoming paramount for rehabilitation and assistive purposes in fields such as brain- computer interface (BCI), closed-loop neuromodulation for neurological disorders, etc. The complex, non-stationary, and very low signal-to-noise ratio of brain signals poses a significant challenge for researchers to design robust and reliable BCI systems outside the laboratory environment. In this study, we present novel recurrence plots (RPs) based on convolutional neural network and long short term memory (CNN-LSTM) algorithm for four class functional near-infrared spectroscopy (fNIRS) BCI. The acquired brain signals are projected into a non-linear dimension with RPs and fed into the CNN, which extracts the important features. Then LSTM learns the chronological and time-dependent relations. The average accuracy achieved with the proposed model is 77.7%, while the maximum accuracy is 85.9%. The results confirm the viability of RPs based deep learning algorithms for successful BCI systems.","PeriodicalId":299121,"journal":{"name":"2021 International Conference on Artificial Intelligence and Mechatronics Systems (AIMS)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134265064","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 : 2021-04-28DOI: 10.1109/AIMS52415.2021.9466033
Mudiarta Tauda, Z. Zainuddin, Z. Tahir
The integration of client applications and server is an inseparable part of the process of creating a system. For these two types of applications to communicate well, developers often use the architectural model of Representational State Transfer (REST) or commonly known as the RESTful API. In practice, client application developers have difficulty in integrating their applications with applications running on the server-side, because it requires parameter consistency in every request and response. To minimize errors that can occur, we designed a programming language translator system using the architecture of Long Short-Term Memory (LSTM). The proposed system can translate applications running on the server-side (backend) written using the typescript programming language with the Nest JS framework into an Android-based client application using the Kotlin language with Retrofit modules. The results of this study indicate 93.33% accuracy.
客户端应用程序和服务器的集成是创建系统过程中不可分割的一部分。为了使这两种类型的应用程序能够很好地通信，开发人员经常使用具象状态传输(Representational State Transfer, REST)的体系结构模型，或者通常称为RESTful API。在实践中，客户端应用程序开发人员很难将他们的应用程序与运行在服务器端上的应用程序集成在一起，因为这需要每个请求和响应的参数一致性。为了尽量减少可能发生的错误，我们设计了一个使用长短期记忆(LSTM)架构的编程语言翻译系统。该系统可以将使用typescript编程语言和Nest JS框架编写的运行在服务器端(后端)上的应用程序转换为使用Kotlin语言和Retrofit模块的基于android的客户端应用程序。本研究结果表明准确率为93.33%。
{"title":"Programming Language Translator For Integration Client Application With Web APIs","authors":"Mudiarta Tauda, Z. Zainuddin, Z. Tahir","doi":"10.1109/AIMS52415.2021.9466033","DOIUrl":"https://doi.org/10.1109/AIMS52415.2021.9466033","url":null,"abstract":"The integration of client applications and server is an inseparable part of the process of creating a system. For these two types of applications to communicate well, developers often use the architectural model of Representational State Transfer (REST) or commonly known as the RESTful API. In practice, client application developers have difficulty in integrating their applications with applications running on the server-side, because it requires parameter consistency in every request and response. To minimize errors that can occur, we designed a programming language translator system using the architecture of Long Short-Term Memory (LSTM). The proposed system can translate applications running on the server-side (backend) written using the typescript programming language with the Nest JS framework into an Android-based client application using the Kotlin language with Retrofit modules. The results of this study indicate 93.33% accuracy.","PeriodicalId":299121,"journal":{"name":"2021 International Conference on Artificial Intelligence and Mechatronics Systems (AIMS)","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134270305","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 : 2021-04-28DOI: 10.1109/AIMS52415.2021.9466021
Syed Usama Bin Sabir, Kashif Ahmed, Usama Sabir, Noman Naseer
In this paper, design and testing phase of a powered exoskeleton built for enhancement of upper limb, is discussed. An exoskeleton for right upper limb is made having optimal weight actuated by a stepper motor. The input signal for this device to operate is an EMG signal taken from muscular movements of muscles inside forearm. Clenching fist and palm extension are the hand gestures to operate exoskeleton. A binary classifier, Linear Discriminant Analysis, is trained for the corresponding EMG signals of these hand movements to generate control commands for operating device with accuracy. The project also includes acquiring and processing EMG signals. The augmenting factor of this device is set to 1.5 to assist user for lifting loads ranging from 5 to 10 KG for an extended period without causing muscular fatigue.
{"title":"Design and Fabrication of Exoskeleton for Power Augmentation of Arm using Intuitive Control","authors":"Syed Usama Bin Sabir, Kashif Ahmed, Usama Sabir, Noman Naseer","doi":"10.1109/AIMS52415.2021.9466021","DOIUrl":"https://doi.org/10.1109/AIMS52415.2021.9466021","url":null,"abstract":"In this paper, design and testing phase of a powered exoskeleton built for enhancement of upper limb, is discussed. An exoskeleton for right upper limb is made having optimal weight actuated by a stepper motor. The input signal for this device to operate is an EMG signal taken from muscular movements of muscles inside forearm. Clenching fist and palm extension are the hand gestures to operate exoskeleton. A binary classifier, Linear Discriminant Analysis, is trained for the corresponding EMG signals of these hand movements to generate control commands for operating device with accuracy. The project also includes acquiring and processing EMG signals. The augmenting factor of this device is set to 1.5 to assist user for lifting loads ranging from 5 to 10 KG for an extended period without causing muscular fatigue.","PeriodicalId":299121,"journal":{"name":"2021 International Conference on Artificial Intelligence and Mechatronics Systems (AIMS)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133911164","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 : 2021-04-28DOI: 10.1109/AIMS52415.2021.9466058
Siska Armalivia, Z. Zainuddin, A. Achmad, Muh. Arief Wicaksono
Shrimp farming activities can be divided into hatchery and rearing. During the hatchery process, there is an activity of counting shrimp larvae. Currently, the method of counting larvae was still done manually using the sampling method by taking a full cup of shrimp larvae and counting them manually. This process was timewasting and can make the result in miscalculations due to human error. Therefore, this study proposed the You Only Look Once (YOLOv3) algorithm to perform automatic counting of the quantity of shrimp larvae. The results of the YOLOv3 final model show a good performance with a mean Average Precision (mAP) value of 96.83% and an average accuracy value of 76.48%. This YOLO algorithm can calculate shrimp larvae with hight precision.
对虾养殖活动可分为孵卵和饲养。在孵化过程中，有虾仔计数活动。目前，对幼虫的计数方法仍采用人工取样法，取满杯虾幼虫，人工计数。这个过程很浪费时间，并且可能由于人为错误而导致计算错误。为此，本研究提出You Only Look Once (YOLOv3)算法，实现对虾幼虫数量的自动计数。结果表明，YOLOv3最终模型的平均精度(mAP)为96.83%，平均精度为76.48%，具有良好的性能。该YOLO算法能以较高的精度计算虾仔。
{"title":"Automatic Counting Shrimp Larvae Based You Only Look Once (YOLO)","authors":"Siska Armalivia, Z. Zainuddin, A. Achmad, Muh. Arief Wicaksono","doi":"10.1109/AIMS52415.2021.9466058","DOIUrl":"https://doi.org/10.1109/AIMS52415.2021.9466058","url":null,"abstract":"Shrimp farming activities can be divided into hatchery and rearing. During the hatchery process, there is an activity of counting shrimp larvae. Currently, the method of counting larvae was still done manually using the sampling method by taking a full cup of shrimp larvae and counting them manually. This process was timewasting and can make the result in miscalculations due to human error. Therefore, this study proposed the You Only Look Once (YOLOv3) algorithm to perform automatic counting of the quantity of shrimp larvae. The results of the YOLOv3 final model show a good performance with a mean Average Precision (mAP) value of 96.83% and an average accuracy value of 76.48%. This YOLO algorithm can calculate shrimp larvae with hight precision.","PeriodicalId":299121,"journal":{"name":"2021 International Conference on Artificial Intelligence and Mechatronics Systems (AIMS)","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133073576","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：