Pub Date : 2024-02-28DOI: 10.22146/jnteti.v13i1.8730
I. Santoso, Shoffin Nahwa Utama, Supriyono
Salah satu faktor yang berpengaruh terhadap kinerja Gaussian naive Bayes classifier (GNBC) dalam klasifikasi citra tekstur adalah ukuran (dimensi) citra. Ukuran citra adalah salah satu kriteria citra tekstur terbaik di samping nilai pikselnya. Pada penelitian ini, diusulkan metode untuk mendapatkan ukuran citra tekstur terbaik terhadap GNBC dengan optimalisasi interpolasi nearest neighbor (NN). Ukuran citra tekstur terbaik dengan nilai piksel hasil interpolasi tersebut membuat GNBC mampu membedakan citra tekstur pada setiap kelasnya dengan kinerja paling tinggi. Langkah pertama metode usulan tersebut adalah menentukan ukuran citra tekstur untuk pelatihan melalui kombinasi ukuran baris dan kolom pada proses optimalisasi. Pengubahan ukuran (resizing) semua citra tekstur asal dengan interpolasi NN adalah langkah penting berikutnya untuk mendapatkan citra tekstur baru. Langkah selanjutnya adalah membangun GNBC berdasarkan citra baru hasil interpolasi dan menentukan akurasi klasifikasinya. Langkah terakhir yaitu memilih ukuran citra tekstur terbaik berdasarkan nilai akurasi klasifikasi terbesar sebagai kriteria pertama dan ukuran citra sebagai kriteria kedua. Evaluasi terhadap metode yang diusulkan tersebut dilakukan menggunakan data citra tekstur dari dataset publik CVonline dengan melibatkan beberapa skenario uji coba dan metode interpolasi. Hasil uji coba menunjukkan bahwa pada skenario yang melibatkan lima kelas citra tekstur, GNBC dengan interpolasi NN memberikan nilai akurasi klasifikasi terkecil 89% dan terbesar 100% pada ukuran citra terbaik, masing-masing 14 × 32 dan 47 × 42. Pada skenario yang melibatkan jumlah kelas kecil hingga besar, GNBC dengan interpolasi NN memberikan akurasi klasifikasi 81,6% - 95%. Dari hasil tersebut, GNBC dengan optimalisasi NN memberikan hasil lebih baik daripada metode interpolasi nonadaptif lainnya (bilinear, bicubic, dan lanczos) dan principal component analysis (PCA).
在纹理图像分类中，影响高斯天真贝叶斯分类器（GNBC）性能的一个因素是图像尺寸（维度）。除了像素值之外，图像尺寸也是最佳纹理图像的标准之一。本研究提出了一种通过优化近邻（NN）插值来获得 GNBC 最佳纹理图像尺寸的方法。最佳纹理图像大小加上插值像素值使得 GNBC 能够以最高性能区分每一类纹理图像。建议方法的第一步是在优化过程中通过行和列大小的组合来确定用于训练的纹理图像大小。下一个重要步骤是使用 NN 插值重新调整所有原始纹理图像的大小，以获得新的纹理图像。下一步是根据新的插值图像建立 GNBC，并确定分类精度。最后一步是以分类准确率最大值为第一标准，图像大小为第二标准，选择最佳纹理图像大小。我们使用 CVonline 公共数据集中的纹理图像数据对所提出的方法进行了评估，涉及多个测试场景和插值方法。测试结果表明，在涉及五类纹理图像的场景中，采用 NN 插值的 GNBC 分类准确率最小，为 89%；在最佳图像尺寸 14 × 32 和 47 × 42 时，分类准确率最大，分别为 100%。 在涉及少量到大量类别的场景中，采用 NN 插值的 GNBC 分类准确率为 81.6% - 95%。从这些结果来看，采用 NN 优化的 GNBC 比其他非自适应插值方法（双线性、双三次和 Lanczos）和主成分分析（PCA）效果更好。
{"title":"Citra Tekstur Terbaik Untuk Gaussian Naïve Bayes Dengan Interpolasi Nearest Neighbor","authors":"I. Santoso, Shoffin Nahwa Utama, Supriyono","doi":"10.22146/jnteti.v13i1.8730","DOIUrl":"https://doi.org/10.22146/jnteti.v13i1.8730","url":null,"abstract":"Salah satu faktor yang berpengaruh terhadap kinerja Gaussian naive Bayes classifier (GNBC) dalam klasifikasi citra tekstur adalah ukuran (dimensi) citra. Ukuran citra adalah salah satu kriteria citra tekstur terbaik di samping nilai pikselnya. Pada penelitian ini, diusulkan metode untuk mendapatkan ukuran citra tekstur terbaik terhadap GNBC dengan optimalisasi interpolasi nearest neighbor (NN). Ukuran citra tekstur terbaik dengan nilai piksel hasil interpolasi tersebut membuat GNBC mampu membedakan citra tekstur pada setiap kelasnya dengan kinerja paling tinggi. Langkah pertama metode usulan tersebut adalah menentukan ukuran citra tekstur untuk pelatihan melalui kombinasi ukuran baris dan kolom pada proses optimalisasi. Pengubahan ukuran (resizing) semua citra tekstur asal dengan interpolasi NN adalah langkah penting berikutnya untuk mendapatkan citra tekstur baru. Langkah selanjutnya adalah membangun GNBC berdasarkan citra baru hasil interpolasi dan menentukan akurasi klasifikasinya. Langkah terakhir yaitu memilih ukuran citra tekstur terbaik berdasarkan nilai akurasi klasifikasi terbesar sebagai kriteria pertama dan ukuran citra sebagai kriteria kedua. Evaluasi terhadap metode yang diusulkan tersebut dilakukan menggunakan data citra tekstur dari dataset publik CVonline dengan melibatkan beberapa skenario uji coba dan metode interpolasi. Hasil uji coba menunjukkan bahwa pada skenario yang melibatkan lima kelas citra tekstur, GNBC dengan interpolasi NN memberikan nilai akurasi klasifikasi terkecil 89% dan terbesar 100% pada ukuran citra terbaik, masing-masing 14 × 32 dan 47 × 42. Pada skenario yang melibatkan jumlah kelas kecil hingga besar, GNBC dengan interpolasi NN memberikan akurasi klasifikasi 81,6% - 95%. Dari hasil tersebut, GNBC dengan optimalisasi NN memberikan hasil lebih baik daripada metode interpolasi nonadaptif lainnya (bilinear, bicubic, dan lanczos) dan principal component analysis (PCA). ","PeriodicalId":31477,"journal":{"name":"Jurnal Nasional Teknik Elektro dan Teknologi Informasi","volume":"131 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140417703","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 : 2024-02-28DOI: 10.22146/jnteti.v13i1.8532
Anugerah Simanjuntak, Rosni Lumbantoruan, Kartika Sianipar, Rut Gultom, Mario Simaremare, Samuel Situmeang, Erwin Panggabean
The rapid advancement of communication technology has transformed how information is shared, but it has also brought concerns about the proliferation of false information. A recent report by the Ministry of Communication and Informatics in Indonesia revealed that around 800,000 websites were involved in spreading false information, underscoring the seriousness of the problem. To combat this issue, researchers have focused on developing techniques to detect and combat fake news. This research centers on using IndoBERT-base-p1 for fake news detection and aims to enhance its performance through three methods to tune the hyperparameter value of the model namely: Bayesian optimization, grid search, and random search. After comparing the outcomes of the three hyperparameter tuning methods, Bayesian Optimization emerged as the most effective approach. Achieving a precision of 88.79%, recall of 94.5%, and F1-score of 91.56% for the “fake” label, Bayesian Optimization outperformed the other hyperparameter tuning methods as well as the model using the fine-tuning hyperparameter value. These findings emphasize the importance of hyperparameter tuning in improving the accuracy of fake news detection models. Utilizing Bayesian Optimization and optimizing the specified hyperparameters, the model demonstrated superior performance in accurately identifying instances of fake news, providing a valuable tool in the ongoing battle against disinformation in the digital realm.
{"title":"Research and Analysis of IndoBERT Hyperparameter Tuning in Fake News Detection","authors":"Anugerah Simanjuntak, Rosni Lumbantoruan, Kartika Sianipar, Rut Gultom, Mario Simaremare, Samuel Situmeang, Erwin Panggabean","doi":"10.22146/jnteti.v13i1.8532","DOIUrl":"https://doi.org/10.22146/jnteti.v13i1.8532","url":null,"abstract":"The rapid advancement of communication technology has transformed how information is shared, but it has also brought concerns about the proliferation of false information. A recent report by the Ministry of Communication and Informatics in Indonesia revealed that around 800,000 websites were involved in spreading false information, underscoring the seriousness of the problem. To combat this issue, researchers have focused on developing techniques to detect and combat fake news. This research centers on using IndoBERT-base-p1 for fake news detection and aims to enhance its performance through three methods to tune the hyperparameter value of the model namely: Bayesian optimization, grid search, and random search. After comparing the outcomes of the three hyperparameter tuning methods, Bayesian Optimization emerged as the most effective approach. Achieving a precision of 88.79%, recall of 94.5%, and F1-score of 91.56% for the “fake” label, Bayesian Optimization outperformed the other hyperparameter tuning methods as well as the model using the fine-tuning hyperparameter value. These findings emphasize the importance of hyperparameter tuning in improving the accuracy of fake news detection models. Utilizing Bayesian Optimization and optimizing the specified hyperparameters, the model demonstrated superior performance in accurately identifying instances of fake news, providing a valuable tool in the ongoing battle against disinformation in the digital realm.","PeriodicalId":31477,"journal":{"name":"Jurnal Nasional Teknik Elektro dan Teknologi Informasi","volume":"22 7","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140420059","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 : 2024-02-23DOI: 10.22146/jnteti.v13i1.4395
Arif Indra Irawan, Iman Hedi Santoso, Maya Rahayu, Arif Indra, Irawan
The quick response (QR) code-based attendance application contributes to reducing paper usage and attendance input errors. However, in its implementation process, the QR-code-based attendance at a Bandung school demonstrates weaknesses. Absent students can fake their attendance for themselves or friends. This type of attack is known as fake QR code generation. This research proposes a security authentication system using the Rivest–Shamir–Adleman (RSA) encryption algorithm and the secure hash algorithm 1 (SHA-1) to secure QR code-based attendance applications from fake QR code generation attacks. The RSA algorithm encrypts QR code data to maintain privacy, while the SHA-1 algorithm ensures data integrity. Based on this method, the mutual authentication process between the QR code data generated by the student and the attendance reading application by the teacher can be established. The results obtained from a series of tests showed that the security system in the student attendance recording application that had been implemented at Madrasah Aliyah (MA) Al-Mukhlishin could detect and prevent fake QR code generation attacks. The test was conducted by changing the impact of the key length on RSA-1024 bits and RSA-2048 bits. The results showed that in RSA-1024 bits, energy consumption of 0.14 J and time of 1.66 s is more efficient than that in RSA-2048 bits, with energy consumption of 0.19 J and time of 2.09 s. Interestingly, if a higher level of security is required, the key length should be increased at the expense of some energy and time efficiency.
{"title":"Implementation of QR Code Attendance Security System Using RSA and Hash Algorithms","authors":"Arif Indra Irawan, Iman Hedi Santoso, Maya Rahayu, Arif Indra, Irawan","doi":"10.22146/jnteti.v13i1.4395","DOIUrl":"https://doi.org/10.22146/jnteti.v13i1.4395","url":null,"abstract":"The quick response (QR) code-based attendance application contributes to reducing paper usage and attendance input errors. However, in its implementation process, the QR-code-based attendance at a Bandung school demonstrates weaknesses. Absent students can fake their attendance for themselves or friends. This type of attack is known as fake QR code generation. This research proposes a security authentication system using the Rivest–Shamir–Adleman (RSA) encryption algorithm and the secure hash algorithm 1 (SHA-1) to secure QR code-based attendance applications from fake QR code generation attacks. The RSA algorithm encrypts QR code data to maintain privacy, while the SHA-1 algorithm ensures data integrity. Based on this method, the mutual authentication process between the QR code data generated by the student and the attendance reading application by the teacher can be established. The results obtained from a series of tests showed that the security system in the student attendance recording application that had been implemented at Madrasah Aliyah (MA) Al-Mukhlishin could detect and prevent fake QR code generation attacks. The test was conducted by changing the impact of the key length on RSA-1024 bits and RSA-2048 bits. The results showed that in RSA-1024 bits, energy consumption of 0.14 J and time of 1.66 s is more efficient than that in RSA-2048 bits, with energy consumption of 0.19 J and time of 2.09 s. Interestingly, if a higher level of security is required, the key length should be increased at the expense of some energy and time efficiency.","PeriodicalId":31477,"journal":{"name":"Jurnal Nasional Teknik Elektro dan Teknologi Informasi","volume":"15 10","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140435361","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 : 2024-02-22DOI: 10.22146/jnteti.v13i1.10095
Haposan Yoga, Pradika Napitupulu, I. Gde, Dharma Nugraha
The growth of vehicle number is unavoidable whilst the availability of parking is not directly proportional with this condition. Nowadays, many shopping centers do not have sufficient parking spot, causing customers to have difficulty in finding available parking spots. Research has been conducted to tackle the issue of finding available parking spots. Much of this research proposed the narrowband-Internet of things (NB-IoT) as a fog node. For communication purposes, this NB-IoT-based fog node has some shortcomings, such as security and privacy, lower data rate, higher cost in development, dependency with wireless system, and only covers one area. In this research, the fog computing was proposed to decentralize smart parking system by using Firebase to cover several areas or malls in one system and interface. Instead of using NB-IoT, this research employed decentralized local server as a fog node to deliver a fast data exchange. Firestore database (Firebase) was also used to secure, manage, and analyze the data in the cloud. Conjunctively, the Android application was created as a user interface to book and find the availability of parking spots. The Android application was built using Android Studio and implemented authentication to keep the data access secure and private. The testing scenario was done following the design unified modeling language (UML). The research results confirmed that the fog computing system successfully supported the decentralized smart parking system and was able to be implemented for covering several areas or malls in one system.
{"title":"Fog Computing-Based System for Decentralized Smart Parking System by Using Firebase","authors":"Haposan Yoga, Pradika Napitupulu, I. Gde, Dharma Nugraha","doi":"10.22146/jnteti.v13i1.10095","DOIUrl":"https://doi.org/10.22146/jnteti.v13i1.10095","url":null,"abstract":"The growth of vehicle number is unavoidable whilst the availability of parking is not directly proportional with this condition. Nowadays, many shopping centers do not have sufficient parking spot, causing customers to have difficulty in finding available parking spots. Research has been conducted to tackle the issue of finding available parking spots. Much of this research proposed the narrowband-Internet of things (NB-IoT) as a fog node. For communication purposes, this NB-IoT-based fog node has some shortcomings, such as security and privacy, lower data rate, higher cost in development, dependency with wireless system, and only covers one area. In this research, the fog computing was proposed to decentralize smart parking system by using Firebase to cover several areas or malls in one system and interface. Instead of using NB-IoT, this research employed decentralized local server as a fog node to deliver a fast data exchange. Firestore database (Firebase) was also used to secure, manage, and analyze the data in the cloud. Conjunctively, the Android application was created as a user interface to book and find the availability of parking spots. The Android application was built using Android Studio and implemented authentication to keep the data access secure and private. The testing scenario was done following the design unified modeling language (UML). The research results confirmed that the fog computing system successfully supported the decentralized smart parking system and was able to be implemented for covering several areas or malls in one system.","PeriodicalId":31477,"journal":{"name":"Jurnal Nasional Teknik Elektro dan Teknologi Informasi","volume":"9 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140439208","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 : 2024-02-19DOI: 10.22146/jnteti.v13i1.8564
Eni Dwi Wardihani, Eka Ulia Sari, Helmy, Ari Sriyanto Nugroho, Yusnan Badruzzaman, Arif Nursyahid, Thomas Agung Setyawan, M. Fitri, I. Nugraha
Modernisasi pada sektor pertanian diharapkan berpengaruh pada peningkatan kualitas, jumlah produksi, dan kontinuitas suplai hasil tani. Saat ini banyak dikembangkan pertanian cerdas (smart agriculture) dalam greenhouse. Namun, beberapa parameter greenhouse harus diperhatikan supaya pertumbuhan tanaman berlangsung secara optimal. Pada penelitian ini dibuat sebuah sistem pemantauan dan pengendalian beberapa parameter greenhouse berbasis internet of things (IoT) yang memungkinkan petani melakukan pemantauan dan pengendalian kapan pun dan di mana pun. Hal ini juga dapat meningkatkan efisiensi kerja petani dalam melakukan pemantauan dan pengendalian, terlebih jika greenhouse yang dipantau dan dikendalikan lebih dari satu. Data-data hasil pemantauan greenhouse dapat dilihat secara real-time dan tersimpan di server, sehingga memudahkan petani melakukan evaluasi terhadap greenhouse dan tanaman. Beberapa parameter yang dipantau adalah suhu greenhouse, kelembapan greenhouse, dan intensitas cahaya dalam greenhouse, sedangkan parameter yang dikendalikan adalah suhu greenhouse dan kelembapan greenhouse, dengan menggunakan exhaust fan. Proses pengiriman data pemantauan dan pengendalian parameter greenhouse dilakukan menggunakan protokol message queue telemetry transport (MQTT). Pengujian data loss dan delay pada sistem diperlukan untuk mengetahui keandalan alat dalam proses pengiriman dan penerimaan data. Hasil pengujian quality of service (QoS) adalah sebagai berikut: rata-rata data loss pemantauan Gateway-Server sebesar 10,6 %, rata-rata delay pemantauan gateway-server sebesar 1,9 s, dan rata-rata delay pengendalian server-gateway sebesar 7,1 s. Saat nilai parameter suhu greenhouse kurang dari batas ambang maksimal yang ditentukan, sistem akan menyalakan drum fan agar suhu mencapai nilai minimal pada batas ambang.
{"title":"Pemantauan dan Pengendalian Parameter Greenhouse Berbasis IoT Dengan Protokol MQTT","authors":"Eni Dwi Wardihani, Eka Ulia Sari, Helmy, Ari Sriyanto Nugroho, Yusnan Badruzzaman, Arif Nursyahid, Thomas Agung Setyawan, M. Fitri, I. Nugraha","doi":"10.22146/jnteti.v13i1.8564","DOIUrl":"https://doi.org/10.22146/jnteti.v13i1.8564","url":null,"abstract":"Modernisasi pada sektor pertanian diharapkan berpengaruh pada peningkatan kualitas, jumlah produksi, dan kontinuitas suplai hasil tani. Saat ini banyak dikembangkan pertanian cerdas (smart agriculture) dalam greenhouse. Namun, beberapa parameter greenhouse harus diperhatikan supaya pertumbuhan tanaman berlangsung secara optimal. Pada penelitian ini dibuat sebuah sistem pemantauan dan pengendalian beberapa parameter greenhouse berbasis internet of things (IoT) yang memungkinkan petani melakukan pemantauan dan pengendalian kapan pun dan di mana pun. Hal ini juga dapat meningkatkan efisiensi kerja petani dalam melakukan pemantauan dan pengendalian, terlebih jika greenhouse yang dipantau dan dikendalikan lebih dari satu. Data-data hasil pemantauan greenhouse dapat dilihat secara real-time dan tersimpan di server, sehingga memudahkan petani melakukan evaluasi terhadap greenhouse dan tanaman. Beberapa parameter yang dipantau adalah suhu greenhouse, kelembapan greenhouse, dan intensitas cahaya dalam greenhouse, sedangkan parameter yang dikendalikan adalah suhu greenhouse dan kelembapan greenhouse, dengan menggunakan exhaust fan. Proses pengiriman data pemantauan dan pengendalian parameter greenhouse dilakukan menggunakan protokol message queue telemetry transport (MQTT). Pengujian data loss dan delay pada sistem diperlukan untuk mengetahui keandalan alat dalam proses pengiriman dan penerimaan data. Hasil pengujian quality of service (QoS) adalah sebagai berikut: rata-rata data loss pemantauan Gateway-Server sebesar 10,6 %, rata-rata delay pemantauan gateway-server sebesar 1,9 s, dan rata-rata delay pengendalian server-gateway sebesar 7,1 s. Saat nilai parameter suhu greenhouse kurang dari batas ambang maksimal yang ditentukan, sistem akan menyalakan drum fan agar suhu mencapai nilai minimal pada batas ambang.","PeriodicalId":31477,"journal":{"name":"Jurnal Nasional Teknik Elektro dan Teknologi Informasi","volume":"26 8","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140450253","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 : 2024-02-12DOI: 10.22146/jnteti.v13i1.9292
Hustinawaty, Muhammad Farell
Despite a decline in global COVID-19 cases, the persisting threat of SARS-CoV-2 coupled with waning public awareness of the virus threat has raised concerns. A notable number of individuals disregard mask usage or do so incorrectly. It is particularly concerning given that COVID-19 has high transmissibility, especially in crowded areas like shopping centers. Enforcement officers often face challenges in identifying those wearing masks improperly. Herein lies the significance of automated mask detection to aid enforcement officers in containing the spread of the virus. Hence, this paper aims to highlight the importance of automated mask detection in combatting COVID-19 transmission. Previous mask detection algorithms were intricate because they relied heavily on resource-intensive machine learning algorithms and libraries. These algorithms, however, failed to address the problem of incorrect mask usage adequately. Therefore, despite the apparent usage of masks, the virus managed to find transmission pathways. In contrast, this research focuses on creating algorithms that pinpoint improper mask usage and optimize resource utilization without compromising detection quality. The Haar cascade algorithm was utilized to detect faces and the support vector machine (SVM) was used to train the dataset. The model attained an average accuracy of 95.8%, precision of 99.7%, recall of 92.3%, and F1-score of 93.7%. The metrics aligned with prior studies, affirming their reliability. Nevertheless, limitations exist as the model faces challenges in detecting obscured facial features, requiring further research to enhance its detection capabilities. This research contributes to ongoing efforts to improve mask detection technology for more effective virus containment.
{"title":"Implementation of Mask Use Detection With SVM and Haar Cascade in OpenCV","authors":"Hustinawaty, Muhammad Farell","doi":"10.22146/jnteti.v13i1.9292","DOIUrl":"https://doi.org/10.22146/jnteti.v13i1.9292","url":null,"abstract":"Despite a decline in global COVID-19 cases, the persisting threat of SARS-CoV-2 coupled with waning public awareness of the virus threat has raised concerns. A notable number of individuals disregard mask usage or do so incorrectly. It is particularly concerning given that COVID-19 has high transmissibility, especially in crowded areas like shopping centers. Enforcement officers often face challenges in identifying those wearing masks improperly. Herein lies the significance of automated mask detection to aid enforcement officers in containing the spread of the virus. Hence, this paper aims to highlight the importance of automated mask detection in combatting COVID-19 transmission. Previous mask detection algorithms were intricate because they relied heavily on resource-intensive machine learning algorithms and libraries. These algorithms, however, failed to address the problem of incorrect mask usage adequately. Therefore, despite the apparent usage of masks, the virus managed to find transmission pathways. In contrast, this research focuses on creating algorithms that pinpoint improper mask usage and optimize resource utilization without compromising detection quality. The Haar cascade algorithm was utilized to detect faces and the support vector machine (SVM) was used to train the dataset. The model attained an average accuracy of 95.8%, precision of 99.7%, recall of 92.3%, and F1-score of 93.7%. The metrics aligned with prior studies, affirming their reliability. Nevertheless, limitations exist as the model faces challenges in detecting obscured facial features, requiring further research to enhance its detection capabilities. This research contributes to ongoing efforts to improve mask detection technology for more effective virus containment.","PeriodicalId":31477,"journal":{"name":"Jurnal Nasional Teknik Elektro dan Teknologi Informasi","volume":"14 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140457884","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 : 2024-02-06DOI: 10.22146/jnteti.v13i1.6332
Muhammad Bhayu Bramantyo, Sasongko Pramonohadi, Sarjiya
Upaya dalam mengurangi emisi karbon pada sektor ketenagalistrikan adalah dengan menerapkan pembangkit energi terbarukan yang lebih ramah lingkungan. Pembangkit fotovoltaik (photovoltaic, PV) sebagai pembangkit terdistribusi (distributed generator, DG) merupakan salah satu pembangkit dari energi terbarukan yang menjadi tren saat ini. DG merupakan pembangkit yang letaknya dekat dengan beban pada jaringan distribusi. Dalam penerapannya, PV DG akan memengaruhi besar rugi daya pada jaringan listrik yang sudah ada (existing), sehingga juga memengaruhi besar biaya rugi energi. Selain itu, ketersediaan lahan yang memadai juga diperlukan dalam pemasangan PV DG. Hal ini menyebabkan kerja sama antara pelaku usaha penyedia tenaga listrik PV DG dengan partner beban dilakukan secara berjauhan, sehingga menimbulkan masalah penyaluran. Pembangunan saluran distribusi oleh pelaku usaha untuk evakuasi produksi listriknya hampir tidak mungkin dilakukan. Sewa jaringan distribusi menjadi solusi yang menarik, yaitu melalui skema pemanfaatan bersama jaringan distribusi atau power wheeling. Penelitian ini bertujuan untuk melakukan studi penerapan power wheeling pembangkit PV pada sistem jaringan distribusi IEEE 33 bus, dengan mencari lokasi bus penempatan pembangkit wheeling PV yang menghasilkan total biaya rugi energi dan biaya sewa jaringan distribusi terkecil. Metode MW-km digunakan untuk perhitungan biaya sewa jaringan. Ketersediaan lahan tiap bus juga dipertimbangkan dalam penelitian ini. Hasil penelitian menunjukkan bahwa penempatan pembangkit wheeling PV di bus 8 menghasilkan biaya total rugi energi dan biaya sewa jaringan distribusi terkecil selama setahun. Hal tersebut menunjukkan bahwa penempatan pembangkit wheeling PV di sembarang tempat belum tentu menghasilkan total biaya rugi energi dan biaya sewa jaringan distribusi terkecil.
{"title":"Studi Penerapan Power Wheeling Pembangkit Fotovoltaik Dengan Metode MW-km","authors":"Muhammad Bhayu Bramantyo, Sasongko Pramonohadi, Sarjiya","doi":"10.22146/jnteti.v13i1.6332","DOIUrl":"https://doi.org/10.22146/jnteti.v13i1.6332","url":null,"abstract":"Upaya dalam mengurangi emisi karbon pada sektor ketenagalistrikan adalah dengan menerapkan pembangkit energi terbarukan yang lebih ramah lingkungan. Pembangkit fotovoltaik (photovoltaic, PV) sebagai pembangkit terdistribusi (distributed generator, DG) merupakan salah satu pembangkit dari energi terbarukan yang menjadi tren saat ini. DG merupakan pembangkit yang letaknya dekat dengan beban pada jaringan distribusi. Dalam penerapannya, PV DG akan memengaruhi besar rugi daya pada jaringan listrik yang sudah ada (existing), sehingga juga memengaruhi besar biaya rugi energi. Selain itu, ketersediaan lahan yang memadai juga diperlukan dalam pemasangan PV DG. Hal ini menyebabkan kerja sama antara pelaku usaha penyedia tenaga listrik PV DG dengan partner beban dilakukan secara berjauhan, sehingga menimbulkan masalah penyaluran. Pembangunan saluran distribusi oleh pelaku usaha untuk evakuasi produksi listriknya hampir tidak mungkin dilakukan. Sewa jaringan distribusi menjadi solusi yang menarik, yaitu melalui skema pemanfaatan bersama jaringan distribusi atau power wheeling. Penelitian ini bertujuan untuk melakukan studi penerapan power wheeling pembangkit PV pada sistem jaringan distribusi IEEE 33 bus, dengan mencari lokasi bus penempatan pembangkit wheeling PV yang menghasilkan total biaya rugi energi dan biaya sewa jaringan distribusi terkecil. Metode MW-km digunakan untuk perhitungan biaya sewa jaringan. Ketersediaan lahan tiap bus juga dipertimbangkan dalam penelitian ini. Hasil penelitian menunjukkan bahwa penempatan pembangkit wheeling PV di bus 8 menghasilkan biaya total rugi energi dan biaya sewa jaringan distribusi terkecil selama setahun. Hal tersebut menunjukkan bahwa penempatan pembangkit wheeling PV di sembarang tempat belum tentu menghasilkan total biaya rugi energi dan biaya sewa jaringan distribusi terkecil.","PeriodicalId":31477,"journal":{"name":"Jurnal Nasional Teknik Elektro dan Teknologi Informasi","volume":"92 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140461185","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 : 2024-02-05DOI: 10.22146/jnteti.v13i1.8795
Rian Putra Pratama, S. Supangkat
Perlintasan sebidang masih menjadi permasalahan di beberapa kota akibat tingkat pelanggaran yang tinggi. Saat ini, pengawasan di perlintasan sebidang masih dilakukan secara konvensional. Permasalahan di perlintasan sebidang makin kompleks dan solusi konvensional tidak lagi efektif, sehingga diperlukan sistem pengawasan video cerdas sebagai solusi. Penerapan sistem pengawasan video cerdas merupakan tugas yang kompleks dan memerlukan perangkat dengan sumber daya komputasi besar. Penelitian ini bertujuan mengoptimalkan sistem dalam melakukan pemrosesan data secara real-time dengan melakukan komputasi dekat dengan sumber data dan membagi tugas komputasi ke beberapa perangkat tepi (edge device). Sistem pengawasan video cerdas berbasis komputasi tepi (edge computing) dengan metode computation offloading pada perangkat terbatas diusulkan sebagai solusi pada penelitian ini. Penelitian ini terdiri atas dua tahap pengembangan. Tahap pertama mengembangkan model deteksi objek menggunakan dataset perlintasan sebidang di Kota Bandung dan tahap kedua mengembangkan sistem berbasis komputasi tepi dengan menerapkan metode computation offloading pada perangkat komputasi terbatas. Metode komputasi tepi memperluas komputasi awan (cloud computing) ke tepi jaringan untuk melakukan perhitungan dekat dengan sumber data, sedangkan metode computation offloading meningkatkan kinerja komputasi tepi dengan membagi tugas komputasi. Hasil pengujian menunjukkan terjadinya peningkatan kecepatan komputasi sekitar 1,5 kali lebih cepat dengan tingkat akurasi deteksi pelanggaran mencapai 89,4%. Selain itu, terjadi penurunan suhu GPU sebesar 5,50 °C, penggunaan GPU turun 44,05%, penggunaan memori berkurang 301 Mb, dan konsumsi daya menurun 2,28 W. Sistem yang dikembangkan efektif dan efisien dalam mengoptimalkan kinerja sistem deteksi pelanggaran di perlintasan sebidang pada perangkat komputasi terbatas.
{"title":"Penerapan Computation Offloading Pada Sistem Deteksi Pelanggaran Perlintasan Sebidang Berbasis Komputasi Tepi","authors":"Rian Putra Pratama, S. Supangkat","doi":"10.22146/jnteti.v13i1.8795","DOIUrl":"https://doi.org/10.22146/jnteti.v13i1.8795","url":null,"abstract":"Perlintasan sebidang masih menjadi permasalahan di beberapa kota akibat tingkat pelanggaran yang tinggi. Saat ini, pengawasan di perlintasan sebidang masih dilakukan secara konvensional. Permasalahan di perlintasan sebidang makin kompleks dan solusi konvensional tidak lagi efektif, sehingga diperlukan sistem pengawasan video cerdas sebagai solusi. Penerapan sistem pengawasan video cerdas merupakan tugas yang kompleks dan memerlukan perangkat dengan sumber daya komputasi besar. Penelitian ini bertujuan mengoptimalkan sistem dalam melakukan pemrosesan data secara real-time dengan melakukan komputasi dekat dengan sumber data dan membagi tugas komputasi ke beberapa perangkat tepi (edge device). Sistem pengawasan video cerdas berbasis komputasi tepi (edge computing) dengan metode computation offloading pada perangkat terbatas diusulkan sebagai solusi pada penelitian ini. Penelitian ini terdiri atas dua tahap pengembangan. Tahap pertama mengembangkan model deteksi objek menggunakan dataset perlintasan sebidang di Kota Bandung dan tahap kedua mengembangkan sistem berbasis komputasi tepi dengan menerapkan metode computation offloading pada perangkat komputasi terbatas. Metode komputasi tepi memperluas komputasi awan (cloud computing) ke tepi jaringan untuk melakukan perhitungan dekat dengan sumber data, sedangkan metode computation offloading meningkatkan kinerja komputasi tepi dengan membagi tugas komputasi. Hasil pengujian menunjukkan terjadinya peningkatan kecepatan komputasi sekitar 1,5 kali lebih cepat dengan tingkat akurasi deteksi pelanggaran mencapai 89,4%. Selain itu, terjadi penurunan suhu GPU sebesar 5,50 °C, penggunaan GPU turun 44,05%, penggunaan memori berkurang 301 Mb, dan konsumsi daya menurun 2,28 W. Sistem yang dikembangkan efektif dan efisien dalam mengoptimalkan kinerja sistem deteksi pelanggaran di perlintasan sebidang pada perangkat komputasi terbatas.","PeriodicalId":31477,"journal":{"name":"Jurnal Nasional Teknik Elektro dan Teknologi Informasi","volume":"110 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140461825","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 : 2024-02-02DOI: 10.22146/jnteti.v13i1.8145
Bagas Alif Fimaskoro, S. Aulia, Dery Rimasa
Over time, numerous developments in digital technology have benefited people, including anthropometric measurements that provide information on an athlete’s ability in sports. The use of digital technology in sports must continue, particularly in the National Sports Committee of Indonesia (Komite Olahraga Nasional Indonesia, KONI) of Bandung City. This study proposed a technique for classifying and identifying fencing athletes’ talents. This work developed a methodology for evaluating sports talent based on anthropometric measurements of athletes’ bodies using the posture detection approach. Fencing and nonfencing athletes in KONI Bandung City were categorized using this talent identification. This study used 36 datasets of body posture images from various skills of the sport. These images were in JPEG or JPG format with a resolution of 3,024 × 4,032 and were acquired using a Canon EOS 1300D camera. This study utilized four points landmarks, which are usually used as measurement components in KONI, to categorize fencing athletes. The four points are shoulder (S), elbow (E), index (I), and hip (H) landmarks. The testing was done using three different dataset settings. According to the test results of all scenarios, scenario 2 had the highest accuracy. This scenario was able to categorize fencing and nonfencing athletes with an accuracy rate of 89% and an average processing time of less than 3 s per image.
随着时间的推移，数字技术的众多发展造福了人们，其中包括提供运动员运动能力信息的人体测量技术。必须继续在体育运动中使用数字技术，特别是在万隆市的印度尼西亚国家体育委员会（Komite Olahraga Nasional Indonesia, KONI）。本研究提出了一种对击剑运动员的天赋进行分类和鉴定的技术。这项工作利用姿势检测方法，根据对运动员身体的人体测量结果，开发了一种评估运动天赋的方法。通过这种天赋识别方法，对万隆科尼市的击剑和非击剑运动员进行了分类。本研究使用了 36 个来自不同运动技能的身体姿势图像数据集。这些图像均为 JPEG 或 JPG 格式，分辨率为 3,024 × 4,032 ，使用佳能 EOS 1300D 相机获取。本研究利用 KONI 中通常用作测量组件的四个点地标对击剑运动员进行分类。这四个点分别是肩（S）、肘（E）、食指（I）和髋（H）地标。测试使用了三种不同的数据集设置。根据所有方案的测试结果，方案 2 的准确率最高。该方案能够对击剑和非击剑运动员进行分类，准确率为 89%，每张图像的平均处理时间少于 3 秒。
{"title":"Identification of Fencing Athletes Based on Anthropometric Measurements Using MediaPipe Pose","authors":"Bagas Alif Fimaskoro, S. Aulia, Dery Rimasa","doi":"10.22146/jnteti.v13i1.8145","DOIUrl":"https://doi.org/10.22146/jnteti.v13i1.8145","url":null,"abstract":"Over time, numerous developments in digital technology have benefited people, including anthropometric measurements that provide information on an athlete’s ability in sports. The use of digital technology in sports must continue, particularly in the National Sports Committee of Indonesia (Komite Olahraga Nasional Indonesia, KONI) of Bandung City. This study proposed a technique for classifying and identifying fencing athletes’ talents. This work developed a methodology for evaluating sports talent based on anthropometric measurements of athletes’ bodies using the posture detection approach. Fencing and nonfencing athletes in KONI Bandung City were categorized using this talent identification. This study used 36 datasets of body posture images from various skills of the sport. These images were in JPEG or JPG format with a resolution of 3,024 × 4,032 and were acquired using a Canon EOS 1300D camera. This study utilized four points landmarks, which are usually used as measurement components in KONI, to categorize fencing athletes. The four points are shoulder (S), elbow (E), index (I), and hip (H) landmarks. The testing was done using three different dataset settings. According to the test results of all scenarios, scenario 2 had the highest accuracy. This scenario was able to categorize fencing and nonfencing athletes with an accuracy rate of 89% and an average processing time of less than 3 s per image.","PeriodicalId":31477,"journal":{"name":"Jurnal Nasional Teknik Elektro dan Teknologi Informasi","volume":"87 15","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140462676","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 : 2024-01-30DOI: 10.22146/jnteti.v13i1.6876
Yohan Fajar Sidik, F. Danang Wijaya, Roni Irnawan, Muhammad Ridwan, Kevin Gausultan, Sriyono
Solid-state transformer (SST) could be a solution for a future distribution system, in which many renewable energy sources (RES) are integrated. The SST consists of a single-phase dual-active bridge (DAB) converter, which is scale-down the dc voltage level. The control objective of the DAB converter used in the SST is to control its output voltage. This control strategy consists of a proportional-integral (PI) controller and a single-phase shift (SPS) modulation. Numerous literatures have mentioned about the SPS modulation for the DAB converter. However, they do not provide procedures in implementing the SPS modulation in the real controller. This paper aims to develop the SPS modulation in the real controller of the STM32F446RE microcontroller. The proposed SPS modulation is based on a master-slave timer feature, which is available in the STM32 microcontroller. The development process and testing of the complete control strategy of the DAB converter were carried out in the hardware-in-the-loop (HIL) simulation using Typhoon HIL. This scheme speeds up the development of process and reduces the costs. The experiment in the HIL environment shows that proposed control strategy of the DAB converter consisting of the PI controller and the SPS modulation is successfully implemented in the real microcontroller of the STM32F446RE. The proposed control strategy of the DAB converter is capable of bidirectional power flow, which is useful for integrating distributed generators in the load side. Moreover, this control strategy can reject the disturbance caused by loads.
固态变压器（SST）是未来配电系统的一种解决方案，该系统集成了多种可再生能源（RES）。固态变压器由一个单相双有源桥（DAB）转换器组成，可降低直流电压水平。SST 中使用的 DAB 转换器的控制目标是控制其输出电压。该控制策略由比例积分（PI）控制器和单相移位（SPS）调制组成。许多文献都提到了 DAB 转换器的 SPS 调制。但是，它们并没有提供在实际控制器中实施 SPS 调制的程序。本文的目的是在 STM32F446RE 微控制器的实际控制器中开发 SPS 调制。建议的 SPS 调制基于 STM32 微控制器中可用的主从定时器功能。DAB 转换器完整控制策略的开发过程和测试是在使用 Typhoon HIL 进行的硬件在环 (HIL) 仿真中完成的。该方案加快了开发过程并降低了成本。HIL 环境下的实验表明，由 PI 控制器和 SPS 调制组成的 DAB 转换器控制策略已在 STM32F446RE 的实际微控制器中成功实现。所提出的 DAB 转换器控制策略能够实现双向功率流，这对于在负载侧集成分布式发电机非常有用。此外，这种控制策略还能抑制负载造成的干扰。
{"title":"Single-Phase Shift Modulation of DAB Converter in Typhoon HIL Simulation","authors":"Yohan Fajar Sidik, F. Danang Wijaya, Roni Irnawan, Muhammad Ridwan, Kevin Gausultan, Sriyono","doi":"10.22146/jnteti.v13i1.6876","DOIUrl":"https://doi.org/10.22146/jnteti.v13i1.6876","url":null,"abstract":"Solid-state transformer (SST) could be a solution for a future distribution system, in which many renewable energy sources (RES) are integrated. The SST consists of a single-phase dual-active bridge (DAB) converter, which is scale-down the dc voltage level. The control objective of the DAB converter used in the SST is to control its output voltage. This control strategy consists of a proportional-integral (PI) controller and a single-phase shift (SPS) modulation. Numerous literatures have mentioned about the SPS modulation for the DAB converter. However, they do not provide procedures in implementing the SPS modulation in the real controller. This paper aims to develop the SPS modulation in the real controller of the STM32F446RE microcontroller. The proposed SPS modulation is based on a master-slave timer feature, which is available in the STM32 microcontroller. The development process and testing of the complete control strategy of the DAB converter were carried out in the hardware-in-the-loop (HIL) simulation using Typhoon HIL. This scheme speeds up the development of process and reduces the costs. The experiment in the HIL environment shows that proposed control strategy of the DAB converter consisting of the PI controller and the SPS modulation is successfully implemented in the real microcontroller of the STM32F446RE. The proposed control strategy of the DAB converter is capable of bidirectional power flow, which is useful for integrating distributed generators in the load side. Moreover, this control strategy can reject the disturbance caused by loads.","PeriodicalId":31477,"journal":{"name":"Jurnal Nasional Teknik Elektro dan Teknologi Informasi","volume":"368 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140480321","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：