cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Nizirwan Anwar
Contact Email
nizirwan.anwar@esaunggul.ac.id
Phone
-
Journal Mail Official
telkomnika@ee.uad.ac.id
Editorial Address
Ahmad Yani st. (Southern Ring Road), Tamanan, Banguntapan, Bantul, Yogyakarta 55191, Indonesia
Location
Kota yogyakarta,
Daerah istimewa yogyakarta
INDONESIA
TELKOMNIKA (Telecommunication Computing Electronics and Control)
Published by Universitas Ahmad Dahlan
ISSN : 16936930     EISSN : 23029293     DOI : 10.12928
Core Subject : Science,
Computer Science & IT
Submitted papers are evaluated by anonymous referees by single blind peer review for contribution, originality, relevance, and presentation. The Editor shall inform you of the results of the review as soon as possible, hopefully in 10 weeks. Please notice that because of the great number of submissions that TELKOMNIKA has received during the last few months the duration of the review process can be up to 14 weeks. Communication Engineering, Computer Network and System Engineering, Computer Science and Information System, Machine Learning, AI and Soft Computing, Signal, Image and Video Processing, Electronics Engineering, Electrical Power Engineering, Power Electronics and Drives, Instrumentation and Control Engineering, Internet of Things (IoT)
Arjuna Subject : Ilmu Komputer (semua kategori) - Aplikasi Ilmu Komputer
Articles 2,614 Documents
 247   248   249   250   251 
Solid waste classification using pyramid scene parsing network segmentation and combined features Khadijah Khadijah; Sukmawati Nur Endah; Retno Kusumaningrum; Rismiyati Rismiyati; Priyo Sidik Sasongko; Iffa Zainan Nisa
TELKOMNIKA (Telecommunication Computing Electronics and Control) Vol 19, No 6: December 2021
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/telkomnika.v19i6.18402

Abstract

Solid waste problem become a serious issue for the countries around the world since the amount of generated solid waste increase annually. As an effort to reduce and reuse of solid waste, a classification of solid waste image is needed  to support automatic waste sorting. In the image classification task, image segmentation and feature extraction play important roles. This research applies recent deep leaning-based segmentation, namely pyramid scene parsing network (PSPNet). We also use various combination of image feature extraction (color, texture, and shape) to search for the best combination of features. As a comparison, we also perform experiment without using segmentation to see the effect of PSPNet. Then, support vector machine (SVM) is applied in the end as classification algorithm. Based on the result of experiment, it can be concluded that generally applying segmentation provide better source for feature extraction, especially in color and shape feature, hence increase the accuracy of classifier. It is also observed that the most important feature in this problem is color feature. However, the accuracy of classifier increase if additional features are introduced. The highest accuracy of 76.49% is achieved when PSPNet segmentation is applied and all combination of features are used.
Shared-hidden-layer Deep Neural Network for Under-resourced Language the Content Devin Hoesen; Dessi Puji Lestari; Dwi Hendratmo Widyantoro
TELKOMNIKA (Telecommunication Computing Electronics and Control) Vol 16, No 3: June 2018
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/telkomnika.v16i3.7984

Abstract

Training speech recognizer with under-resourced language data still proves difficult. Indonesian language is considered under-resourced because the lack of a standard speech corpus, text corpus, and dictionary. In this research, the efficacy of augmenting limited Indonesian speech training data with highly-resourced-language training data, such as English, to train Indonesian speech recognizer was analyzed. The training was performed in form of shared-hidden-layer deep-neural-network (SHL-DNN) training. An SHL-DNN has language-independent hidden layers and can be pre-trained and trained using multilingual training data without any difference with a monolingual deep neural network. The SHL-DNN using Indonesian and English speech training data proved effective for decreasing word error rate (WER) in decoding Indonesian dictated-speech by achieving 3.82% absolute decrease compared to a monolingual Indonesian hidden Markov model using Gaussian mixture model emission (GMM-HMM). The case was confirmed when the SHL-DNN was also employed to decode Indonesian spontaneous-speech by achieving 4.19% absolute WER decrease.
Brain Tumor Segmentation using Multi-level Otsu Thresholding and Chan-Vese Active Contour Model Hadi, Heru Pramono; Faisal, Edi; Rachmawanto, Eko Hari
TELKOMNIKA (Telecommunication Computing Electronics and Control) Vol 19, No 6: December 2021
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/telkomnika.v19i6.21679

Abstract

Research on brain tumor segmentation has been developed, ranging from threshold-based methods to the use of the deep learning algorithm. In this study, we proposed a region-based brain tumor segmentation method, namely the active contour model (ACM). Tumor segmentation was carried out using (Fluid Attenuated Inversion Recovery) FLAIR modality MRI image data obtained from the BRATS 2015 dataset of 86 images. The initial stage of our segmentation method is to find the initial initialization point/area for the ACM algorithm using Multi-Level Otsu Thresholding, with the level used in this study is 3 levels. After the initial initialization area has been obtained, the segmentation process is continued with ACM which explores the tumor area to obtain a full and accurate tumor area result. The results of this study obtained Dice Similarity for our study of 0.7856 with a total time required of 28.080722 seconds, which better than other method that we also compared with ours, 0.75 compared to 0.78 in term of Dice Similarity.
An assisting model for the visually challenged to detect bus door accurately Sreenu Ponnada; Praveen Kumar Sekharamantry; Abhinav Dayal; Srinivas Yarramalle; Nagesh Vadaparthi; Jude Hemanth
TELKOMNIKA (Telecommunication Computing Electronics and Control) Vol 19, No 6: December 2021
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/telkomnika.v19i6.19811

Abstract

Visually impaired individuals are increasing and as per global statistics, around 39 million are blind, and 246 million are affected by low vision. Even in India, as per the recent reviews, over 5 million visually challenged people are present. Authors performed a survey of some critical problems the visually challenged people faced in India from the centre for visually challenged (CVC) School established by UVSM Hospitals. Among the major problems identified through survey, most of these persons prefer carrying out their tasks independently, and depend on public transport buses for migration. However, critical sub-problems being faced include; bus door identification and identifying the bus route number accurately. This article aims to provide solutions in helping visually challenged individuals to identify exact bus that drives them to their destination, its door, bus number, and the path for boarding bus. A video sequence of current scenario would be sent to mobile, in which the actual processing of image is carried out. After the video sequence processing, generated output is a voice message that specifies the bus's location, door, and exact information of the bus number along the road path directly to the user using a wireless device aiming foa a low-cost solution.
Advertisement billboard detection and geotagging system with inductive transfer learning in deep convolutional neural network Romi Fadillah Rahmat; Dennis Dennis; Opim Salim Sitompul; Sarah Purnamawati; Rahmat Budiarto
TELKOMNIKA (Telecommunication Computing Electronics and Control) Vol 17, No 5: October 2019
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/telkomnika.v17i5.11276

Abstract

In this paper, we propose an approach to detect and geotag advertisement billboard in real-time condition. Our approach is using AlexNet’s Deep Convolutional Neural Network (DCNN) as a pre-trained neural network with 1000 categories for image classification. To improve the performance of the pre-trained neural network, we retrain the network by adding more advertisement billboard images using inductive transfer learning approach. Then, we fine-tuned the output layer into advertisement billboard related categories. Furthermore, the detected advertisement billboard images will be geotagged by inserting Exif metadata into the image file. Experimental results show that the approach achieves 92.7% training accuracy for advertisement billboard detection, while for overall testing results it will give 71,86% testing accuracy.
Improved fuzzy miner algorithm for business process discovery Yutika Amelia Effendi; Riyanarto Sarno; Danica Virlianda Marsha
TELKOMNIKA (Telecommunication Computing Electronics and Control) Vol 19, No 6: December 2021
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/telkomnika.v19i6.19015

Abstract

Return material authorization (RMA) is a process in which a company decides to repair or replace customer’s defect product during the warranty period. To execute RMA, both company and customer obliged to follow standard operating procedure (SOP) which usually consists of many business processes of a company well. As the business process could cause inefficiencies, a company should improve their business process regularly. The best way is using process discovery. This research proposes a new improved fuzzy miner algorithm to represent binary correlation between activities. This new algorithm utilizes binary significance and binary correlation equally to acquire fuzzy model. While the original fuzzy miner algorithm uses various binary correlation metrics, the improved fuzzy miner algorithm uses only one metric and could capture the fuzzy model, accurately based on the event logs to capture more accurate business process model. In this research, ProM fuzzy miner is used as a comparison to the proposed improved time-based fuzzy miner. The results showed that the improved algorithm has higher value on conformance checking and able to capture business process model based on time interval, by using only time-interval significance as a binary correlation metrics.
An implentation of IoT for environmental monitoring and its analysis using k-NN algorithm Eko Prayitno; Nurul Fahmi; M. Udin Harun Al Rasyid; Amang Sudarsono
TELKOMNIKA (Telecommunication Computing Electronics and Control) Vol 19, No 6: December 2021
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/telkomnika.v19i6.15724

Abstract

Environmental monitoring is a process for observing around with various conditions. Recently, internet of things (IoT) and wireless sensor network (WSN) technologies support to solve these problems. In this paper, we implemented a system to monitor environmental conditions using IoT and WSN technology. The data measure is temperature, humidity, carbon monoxide (CO) and carbon dioxide (CO2) sensors. All sensor data will be sent and stored to the cloud through the internet in real-time. We provide applications for monitoring website and mobile phone-based environmental conditions, so users can access wherever and whenever. Furthermore, we also confirm the evaluation of analyst data that usedk-NN method is better than other methods with an accuracy rate of 99.0657%.
Human activity recognition for static and dynamic activity using convolutional neural network Agus Eko Minarno; Wahyu Andhyka Kusuma; Yoga Anggi Kurniawan
TELKOMNIKA (Telecommunication Computing Electronics and Control) Vol 19, No 6: December 2021
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/telkomnika.v19i6.20994

Abstract

Evaluated activity as a detail of the human physical movement has become a leading subject for researchers. Activity recognition application is utilized in several areas, such as living, health, game, medical, rehabilitation, and other smart home system applications. An accelerometer was popular sensors to recognize the activity, as well as a gyroscope, which can be embedded in a smartphone. Signal was generated from the accelerometer as a time-series data is an actual approach like a human actifvity pattern. Motion data have acquired in 30 volunteers. Dynamic actives (walking, walking upstairs, walking downstairs) as DA and static actives (laying, standing, sitting) as SA were collected from volunteers. SA and DA it's a challenging problem with the different signal patterns, SA signals coincide between activities but with a clear threshold, otherwise the DA signal is clearly distributed but with an adjacent upper threshold. The proposed network structure achieves a significant performance with the best overall accuracy of 97%. The result indicated the ability of the model for human activity recognition purposes.
Extraction of object image features with gradation contour Fachruddin Fachruddin; Saparudin Saparudin; Errissya Rasywir; Yovi Pratama; Beni Irawan
TELKOMNIKA (Telecommunication Computing Electronics and Control) Vol 19, No 6: December 2021
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/telkomnika.v19i6.19491

Abstract

Image retrieval using features has been used in previous studies including shape, color, texture, but these features are lagging. With the selection of high-level features with contours, this research is done with the hypothesis that images on objects can also be subjected to representations that are commonly used in natural images. Considering the above matters, we need to research the feature extraction of object images using gradation contour. From the results of the gradation contour test results, there is linearity between the results of accuracy with the large number of images tested. Therefore, it can be said that the influence of the number of images will affect the accuracy of classification. The use of contour gradation can be accepted and treated equally in all image types, so there is no more differentiation between image features. The complexity of the image does not affect the method of extracting features that are only used uniquely by an image. From the results of testing the polynomial coefficient savings data as a result of the gradation contour, the highest result is 81.40% with the highest number of categories and the number of images tested in the category is also higher.
5G NOMA user grouping using discrete particle swarm optimization approach Hadhrami Ab. Ghani; Farah Najwa Roslim; Muhammad Akmal Remli; Eissa Mohammed Mohsen Al-Shari; Nurul Izrin Md Saleh; Azizul Azizan
TELKOMNIKA (Telecommunication Computing Electronics and Control) Vol 19, No 6: December 2021
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/telkomnika.v19i6.18580

Abstract

Non-orthogonal multiple access (NOMA) technology meets the increasing demand for high-seed cellular networks such as 5G by offering more users to be accommodated at once in accessing the cellular and wireless network. Moreover, the current demand of cellular networks for enhanced user fairness, greater spectrum efficiency and improved sum capacity further increase the need for NOMA improvement. However, the incurred interference in implementing NOMA user grouping constitutes one of the major barriers in achieving high throughput in NOMA systems. Therefore, this paper presents a computationally lower user grouping approach based on discrete particle swarm intelligence in finding the best user-pairing for 5G NOMA networks and beyond. A discrete particle swarm optimization (DPSO) algorithm is designed and proposed as a promising scheme in performing the user-grouping mechanism. The performance of this proposed approach is measured and demonstrated to have comparable result against the existing state-of-the art approach.

Page 249 of 262 | Total Record : 2614

 247   248   249   250   251 

Filter by Year

2004 2022


Reset
Filter By Issues
All Issue Vol 20, No 3: June 2022 Vol 20, No 2: April 2022 Vol 20, No 1: February 2022 Vol 19, No 6: December 2021 Vol 19, No 5: October 2021 Vol 19, No 4: August 2021 Vol 19, No 3: June 2021 Vol 19, No 2: April 2021 Vol 19, No 1: February 2021 Vol 18, No 6: December 2020 Vol 18, No 5: October 2020 Vol 18, No 4: August 2020 Vol 18, No 3: June 2020 Vol 18, No 2: April 2020 Vol 18, No 1: February 2020 Vol 17, No 6: December 2019 Vol 17, No 5: October 2019 Vol 17, No 4: August 2019 Vol 17, No 3: June 2019 Vol 17, No 2: April 2019 Vol 17, No 1: February 2019 Vol 16, No 6: December 2018 Vol 16, No 5: October 2018 Vol 16, No 4: August 2018 Vol 16, No 3: June 2018 Vol 16, No 2: April 2018 Vol 16, No 1: February 2018 Vol 15, No 4: December 2017 Vol 15, No 3: September 2017 Vol 15, No 2: June 2017 Vol 15, No 1: March 2017 Vol 14, No 4: December 2016 Vol 14, No 3: September 2016 Vol 14, No 2: June 2016 Vol 14, No 1: March 2016 Vol 13, No 4: December 2015 Vol 13, No 3: September 2015 Vol 13, No 2: June 2015 Vol 13, No 1: March 2015 Vol 12, No 4: December 2014 Vol 12, No 3: September 2014 Vol 12, No 2: June 2014 Vol 12, No 1: March 2014 Vol 11, No 4: December 2013 Vol 11, No 3: September 2013 Vol 11, No 2: June 2013 Vol 11, No 1: March 2013 Vol 10, No 4: December 2012 Vol 10, No 3: September 2012 Vol 10, No 2: June 2012 Vol 10, No 1: March 2012 Vol 9, No 3: December 2011 Vol 9, No 2: August 2011 Vol 9, No 1: April 2011 Vol 8, No 3: December 2010 Vol 8, No 2: August 2010 Vol 8, No 1: April 2010 Vol 7, No 3: December 2009 Vol 7, No 2: August 2009 Vol 7, No 1: April 2009 Vol 6, No 3: December 2008 Vol 6, No 2: August 2008 Vol 6, No 1: April 2008 Vol 5, No 3: December 2007 Vol 5, No 2: August 2007 Vol 5, No 1: April 2007 Vol 4, No 3: December 2006 Vol 4, No 2: August 2006 Vol 4, No 1: April 2006 Vol 3, No 3: December 2005 Vol 3, No 2: August 2005 Vol 3, No 1: April 2005 Vol 2, No 1: April 2004 More Issue