Claim Missing Document
Check
Articles

Found 11 Documents
Search

Aplikasi Identifikasi Huruf Braille Menggunakan Computer Vision Berbasis Raspberry Pi Ramiati, Ramiati; Aulia, Siska; Lifwarda, Lifwarda
JURNAL NASIONAL TEKNIK ELEKTRO Vol 9, No 1: March 2020
Publisher : Jurusan Teknik Elektro Universitas Andalas

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25077/jnte.v9n1.707.2020

Abstract

Sense of vision is a source of information on humans. Some humans are created with limited sense of sight. The blind performs reading and writing activities using Braille letters, a printed code system consisting of six dots in various combinations that are highlighted on the paper so that they can be touched. To facilitate the visually impaired and the public in enjoying the works produced by blind people, a script reading system is characterized by Braille by studying the braille characters in advance of each character. This research makes a braille letter identification system into sound using computer vision. The method, the reading of Braille character scripts by studying braille characters. First, a scanner or Raspberry Pi-based camera captures braille characters. Second, the system converts Braille characters into alphabetical shapes by processing Optical Character Recognition images. Recognition of Braille character patterns in written text using Artificial Neural Networks. The results of research on braille testing are in the form of alphabetical texts a through z, and the sound signal of the pronunciation of the alphabet uses the Text To Speech system. Braille to sound conversion system works well, with an average accuracy of system testing of 88.462%. This condition is achieved by using 70 gsm HVS paper and drawing paper with a 52 training image database. The system can only carry out the process of recognition of one character, so it can be used as a reference translator of audio-based braille characters that can be heard by the visually impaired and the community.Keywords : image processing, braille, OCR, JST, text to speechAbstrakIndera penglihatan merupakan sumber informasi pada manusia. Sebagian manusia diciptakan dengan keterbatasan indera penglihatan. Tunanetra melakukan aktifitas membaca serta menulis menggunakan huruf Braille, yaitu sistem cetakan berupa kode terdiri dari enam titik dalam berbagai kombinasi yang ditonjolkan pada kertas sehingga dapat diraba. Untuk memudahkan tunanetra dan masyarakat dalam menikmati karya-karya yang dihasilkan oleh penyandang tunanetra dibuat sistem pembacaan naskah berkarakterkan Braille dengan mempelajari karakter braille terlebih dahulu dari masing-masing karakternya. Penelitian ini membuat sistem identifikasi huruf braille menjadi suara menggunakan computer vision. Metodenya, pembacaan naskah berkarakter Braille dengan mempelajari karakter braille. Pertama, scanner atau kamera berbasis Raspberry Pi mengcapture karakter braille. Kedua, sistem mengkonversi karakter Braille ke bentuk abjad dengan pengolahan citra Optical Character Recognition. Pengenalan pola karakter Braille teks tulisan menggunakan Jaringan Saraf  Tiruan. Hasil penelitian pengujian huruf braille berupa teks abjad a sampai z, dan sinyal suara pengucapan abjad menggunakan sistem Text  To Speech . Sistem konversi braille menjadi suara bekerja dengan baik, dengan akurasi rata-rata pengujian sistem yaitu 88.462%. Kondisi ini dicapai dengan menggunakan kertas HVS 70 gsm dan kertas gambar dengan database 52 citra latih. Sistem hanya dapat melakukan proses pengenalan pada satu karakter, sehingga dapat digunakan sebagai referensi penterjemah naskah karakter braille berbasis audio yang dapat didengarkan oleh tunanetra dan masyakat.Kata Kunci : image processing, braille, OCR, JST, text to speech 
Pengenalan Bentuk Benda Berdasarkan Sinyal Suara dengan Transducer Mikrofon dan Teknologi Kinect Aulia, Siska; Lifwarda, Lifwarda; Yustini, Yustini
JURNAL NASIONAL TEKNIK ELEKTRO Vol 7, No 3: November 2018
Publisher : Jurusan Teknik Elektro Universitas Andalas

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (853.143 KB) | DOI: 10.25077/jnte.v7n3.600.2018

Abstract

Voice processing or speech recognition is growing rapidly hence it can be used for various applications such as moving a system or motion control and multimedia-based learning media. Implementation of speech recognition and image detection in this study using microphone transducer and kinect technology. This study aims to produce a system that can identify and recognize an object with word commands, such as circles, triangles, rectangles and many. In sound processing, sound feature extraction is carried out with Mel-Frequency Cepstrum Coeffecient (MFCC). Word modeling was done using statistical modeling, namely the Hidden Markov Model (HMM). HMM is able to provide an efficient mechanism for statistically modeling diversity in words or words. Data were collected with offline and online microphone transducers. This study matches the pattern of words through training and testing process. The output of this system is a recognizable word based on the highest probability and displaying the object shape based on the recognized word, namely circle, triangle and quadrilateral. Test results with mirofon tranducers, for 85% trained sources, 81.5% untrained sources, and 84% untrained Kinect source testing hence that word recognition systems can be implemented with Kinect technology. Keywords : speech processing, HMM, MFCC, Kinect AbstrakPengolahan suara atau pengenalan kata berkembang pesat sehingga dapat digunakan untuk berbagai aplikasi seperti menggerakan suatu sistem atau kontrol gerak dan media pembelajaran berbasis multimedia. Implementasi pengenalan suara dan deteksi citra pada penelitian ini menggunakan transducer mikrofon dan teknologi kinect. Penelitian ini bertujuan untuk menghasilkan sistem yang dapat mengidentifikasi dan mengenali suatu objek dengan perintah kata, seperti lingkaran, segitiga, segiempat dan segibanyak. Dalam pengolahan suara dilakukan ekstraksi ciri suara dengan Mel-Frequency Cepstrum Coeffecient (MFCC). Pemodelan kata dilakukan dengan menggunakan pemodelan statistik yaitu Hidden Markov Model (HMM). HMM mampu memberikan mekanisme yang efisien untuk memodelkan secara statistik keragaman dalam ucapan atau kata.  Pengambilan data sampel dengan transducer mikrofon secara offline dan online. Pada penelitian ini pencocokan pola kata melalui proses pelatihan dan pengujian kata. Keluaran sistem ini berupa kata yang dikenali berdasarkan probabilitas tertinggi dan menampilkan bentuk benda berdasarkan kata yang dikenali. Prosesnya setelah kata dikenali, sistem akan mentracking citra benda berdasarkan bentuk benda kemudian menampilkan bentuk benda yaitu lingkaran, segitiga, segiempat dan segibanyak. Hasil pengujian dengan tranducer mirofon, untuk sumber terlatih 85%, sumber tidak terlatih 81,5%, dan pengujian dengan Kinect sumber tidak terlatih 84% sehingga sistem pengenalan kata dapat diimplementasikan dengan teknologi Kinect. Kata Kunci : speech processing, HMM, MFCC, kinect
A Portable Device of Air Pollution Measurement Due to Highway Exhaust Emissions Using LabVIEW Programming - Andrizal; - Lifwarda; Anna Yudanur; Rivanol Chadry; - Hendrick
JOIV : International Journal on Informatics Visualization Vol 5, No 4 (2021)
Publisher : Politeknik Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30630/joiv.5.4.697

Abstract

A multisensory gas device integrated with myRIO module to measure air pollution has been established. This device is programmed using the LabVIEW programming language and can measure CO2, CO, NOX, and HC pollution on roads due to motor vehicle exhaust emissions. The device and the display system are made separately using wireless network communication to make this tool portable. Exhaust Gas Analyzer (EGA) was chosen for device calibration, obtaining 3.62% on the average error after performing 30 tests. The tests for measuring CO, CO2, NOX, and HC gas levels were conducted in several locations in Padang City and performed in the morning, afternoon, and evening. The result showed that the system properly measured CO2, CO, NOX and HC pollution in parks and highways in real-time in parts per million (ppm). It also displayed varied gas measurement results in terms of time and test location with a range of CO gas values at 0.034 – 0.15 ppm, CO2 151.3 – 815.2 ppm, NOX 0.0001 – 0.004 ppm, and HC 0.04 – 0.65 ppm. In addition, the system could perform well in providing warnings by automatically activating the air indicator alert at several measurement places when the gas content on one of the gas elements and compounds at a particular location has exceeded the threshold for the clean air category. Thus, this device can be used as initial research to build a real-time air pollution measurement system using the Internet of Things (IoT).
Home Automation dengan Memanfaatkan Mini Komputer Raspberry Pi sebagai Pengontrol Perangkat Elektronik melalui Jaringan Lokal Aprinal Adila Asril; Yul Antonisfia; Ihsan Lumasa Rimra; . Lifwarda
Jurnal Ilmiah Poli Rekayasa Vol 15, No 1 (2019): -
Publisher : Pusat Penelitian dan pengabdian kepada Masyarakat (P3M) Politeknik Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1002.249 KB) | DOI: 10.30630/jipr.15.1.132

Abstract

Meningkatnya kesibukan manusia yang tidak dapat ditunda dalam kehidupan sehari-hari mengakibatkan manusia sekarang tidak memiliki cukup waktu luang untuk melakukan banyak kegiatan dalam waktu yang sama. Disinilah peran teknologi untuk mempermudah dan memperlancar kegiatan manusia dalam kehidupan sehari-hari. Dengan menggunakan Raspberry Pi, dibuatlah sistem Home Automation yang dapat menyelesaikan permasalahan tersebut. Sistem ini juga sangat berguna untuk membantu manusia yang memiliki kebiasaan buruk dalam hal penghematan daya listrik dan jangka waktu pakai sebuah perangkat elektronik, contoh sederhananya kebiasaan lupa dalam mematikan televisi. Pada tugas akhir ini dirancang sebuah sistem dimana, nantinya user dapat melakukan kontrol terhadap beberapa perangkat elektronik yang terdapat pada miniature rumah. Pengontrolan dilakukan dengan mengakses ke webserver yang terdapat pada Raspberry Pi, kemudian melakukan kontrol melalui jaringan lokal. Jaringan lokal tersebut, terdiri dari 3 jaringan yang terhubung oleh 2 wireless router Linksys cisco WRT120N. Dalam melakkukan kontrol Raspberry Pi terhubung pada modul relay yang berfungsi sebagai pemutus dan penyambung arus listrik perangkat elektronik. Hasil pengujian Home automation menunjukkan bahwa untuk dapat mengakses perangkat elektronik yang berada pada miniature rumah maka user dan server harus terhubung dalam jaringan lokal. Sistem ini diakses dengan media kabel dan wireless, dengan melewati beberapa wireless router sistem ini nantinya dapat diimplementasikan pada internet. Kata kunci: Home Automation, Raspberry Pi, wireless router, Linksys Cisco WRT120N, webserver
Aplikasi WiMAX Yenniwarti Rafsyam; Milda Yuliza; Lifwarda Lifwarda
Elektron : Jurnal Ilmiah Vol 1 No 1 (2009): Elektron Jurnal Ilmiah
Publisher : Teknik Elektro Politeknik Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (276.655 KB) | DOI: 10.30630/eji.1.1.11

Abstract

WiMAX is Broadband Wireless Acces (BWA) technology evolution with interactive fitur. WiMAX not only have issue about data speed problem but also about open standard. It means, communications between WiMax instruments between some different vendors are not proprietary. WimAX orientations are not only for fixed market, but also for portable and mobile market. WiMAX with high speed data (up to 70 MBps) is suitable to apply in last mile broadband connections, backhaul and high speed enterprise.
PENGUATAN SINYAL GLOBAL SYSTEM FOR MOBILE COMMUNICATION (GSM) MENGGUNAKAN ANTENA YAGI 14 ELEMEN Firdaus firdaus; Ratna Dewi; Rikki Vitria; Lifwarda lifwarda
Elektron : Jurnal Ilmiah Vol 4 No 2 (2012): Elektron Jurnal Ilmiah
Publisher : Teknik Elektro Politeknik Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (517.677 KB) | DOI: 10.30630/eji.4.2.29

Abstract

The hand phone provides an easy to communicate in domestic calls or international calls. The radiation pattern of yagi antenna leads to one direction. Therefore, this antenna is selected for signal amplifier in global system of mobile communication. The design of 14 element yagi antenna which applied for signal amplification in the cell phones was conducted in this study. The design was done manually without computer software. The antenna was created using an aluminum pipe in 3/8 inch diameter for element and an aluminum beam for boom. The result showed that the operating frequency between the designed antenna and manufactured antenna are the same i.e. 947 MHz. The bandwidth of yagi antenna is 9 MHz. Moreover, the antenna gain which measured in folded antenna transmitter and BTS transmitter are 16 dB and 12 dB respectively. That result is caused differences in distance at the measurement time. Utilization of yagi antenna has succeeded to increase signal amplification in hand phone
PEMBUATAN ANTENA MIKROSTRIP MODEL ANGKA 4 UNTUK PENERIMA SINYAL TELEVISA PADA JALUR UHF (ULTRA HIGH FREKUENSI) Lifwarda lifwarda; Firdaus firdaus
Elektron : Jurnal Ilmiah Vol 5 No 2 (2013): Elektron Jurnal Ilmiah
Publisher : Teknik Elektro Politeknik Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (236.006 KB) | DOI: 10.30630/eji.5.2.57

Abstract

Development of science and technology very rapidly in various fields brought great influence to all aspects of life one of this field of telecommunications. Development of telecommunication technology that bring due faster to hight public demand for telecommunications services users get an easy and fast service are used as medium convey information One of the device in question is antenna. Many type of antennas that have been developed for the reception of televisions signals and are generally made of wires, pipes or aluminium which has large size and weight. Needed for the antenna which has advantages in terms of shape light weight small volume from that can be easily adapted to the priimary device for on the fabrication and can be used for a wide band frequency. The antenna is microstrip antennas capable adjusted in assembly or in other words easily in a microstrip antenna placement. Microstrip antenna models 4 is designed to work in the range frequency 470 – 750 MHz. This antenna has a value of more than -10 dB returnloss to a predetermined range frequency. The measurement results of the antenna resonates at a frequency of 518 MHz gain value obtained by 1 dB, and has a wide bandwidth of 7 MHz, while for polaradiasi shaped bidirectional.
Rancang Bangun Antena Mikrostrip Bowtie Pada Frekuensi 5,2 Ghz Aprinal Adila Asril; Lifwarda Lifwarda; Yul antonisfia
Elektron : Jurnal Ilmiah Vol 10 No 2 (2018): Elektron Jurnal Ilmiah
Publisher : Teknik Elektro Politeknik Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (838.606 KB) | DOI: 10.30630/eji.10.2.73

Abstract

Microstrip antennas are very concerned shapes and sizes. Can be viewed in terms of simple materials, shapes, sizes and dimensions smaller antennae, the price of production is cheaper and able to provide a reasonably good performance, in addition to having many advantages, the microstrip antenna also has its drawbacks one of which is a narrow bandwidth. In this research will be designed a microstrip antenna bowtie which works at a frequency of 5.2 GHz which has a size of 68mm x 33mm groundplane. For the length and width of 33mm x 13mm patch. This antenna is designed on a printed cicuit board (PCB) FR4 epoxy with a dielectric constant of 4.7 and has a thickness of 1,6mm. This bowtie microstrip antenna design using IE3D software. This antenna has been simulated using IE3D software showed its resonance frequency is 5.270 GHz with a return loss -23 595 dB bandwidth of 230 MHz, VSWR 1,142, unidirectional radiation pattern and impedance 43,919Ω. The results of which have been successfully fabricated antenna with a resonant frequency of 5.21 GHz with a return loss -16.813 dB bandwidth of 79 MHz, VSWR 1.368, unidirectional radiation pattern, impedance 43,546Ω and HPBW 105 °.
Pengenalan Bentuk Benda Berdasarkan Sinyal Suara dengan Transducer Mikrofon dan Teknologi Kinect Siska Aulia; Lifwarda Lifwarda; Yustini Yustini
JURNAL NASIONAL TEKNIK ELEKTRO Vol 7, No 3: November 2018
Publisher : Jurusan Teknik Elektro Universitas Andalas

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (853.143 KB) | DOI: 10.25077/jnte.v7n3.600.2018

Abstract

Voice processing or speech recognition is growing rapidly hence it can be used for various applications such as moving a system or motion control and multimedia-based learning media. Implementation of speech recognition and image detection in this study using microphone transducer and kinect technology. This study aims to produce a system that can identify and recognize an object with word commands, such as circles, triangles, rectangles and many. In sound processing, sound feature extraction is carried out with Mel-Frequency Cepstrum Coeffecient (MFCC). Word modeling was done using statistical modeling, namely the Hidden Markov Model (HMM). HMM is able to provide an efficient mechanism for statistically modeling diversity in words or words. Data were collected with offline and online microphone transducers. This study matches the pattern of words through training and testing process. The output of this system is a recognizable word based on the highest probability and displaying the object shape based on the recognized word, namely circle, triangle and quadrilateral. Test results with mirofon tranducers, for 85% trained sources, 81.5% untrained sources, and 84% untrained Kinect source testing hence that word recognition systems can be implemented with Kinect technology. Keywords : speech processing, HMM, MFCC, Kinect AbstrakPengolahan suara atau pengenalan kata berkembang pesat sehingga dapat digunakan untuk berbagai aplikasi seperti menggerakan suatu sistem atau kontrol gerak dan media pembelajaran berbasis multimedia. Implementasi pengenalan suara dan deteksi citra pada penelitian ini menggunakan transducer mikrofon dan teknologi kinect. Penelitian ini bertujuan untuk menghasilkan sistem yang dapat mengidentifikasi dan mengenali suatu objek dengan perintah kata, seperti lingkaran, segitiga, segiempat dan segibanyak. Dalam pengolahan suara dilakukan ekstraksi ciri suara dengan Mel-Frequency Cepstrum Coeffecient (MFCC). Pemodelan kata dilakukan dengan menggunakan pemodelan statistik yaitu Hidden Markov Model (HMM). HMM mampu memberikan mekanisme yang efisien untuk memodelkan secara statistik keragaman dalam ucapan atau kata.  Pengambilan data sampel dengan transducer mikrofon secara offline dan online. Pada penelitian ini pencocokan pola kata melalui proses pelatihan dan pengujian kata. Keluaran sistem ini berupa kata yang dikenali berdasarkan probabilitas tertinggi dan menampilkan bentuk benda berdasarkan kata yang dikenali. Prosesnya setelah kata dikenali, sistem akan mentracking citra benda berdasarkan bentuk benda kemudian menampilkan bentuk benda yaitu lingkaran, segitiga, segiempat dan segibanyak. Hasil pengujian dengan tranducer mirofon, untuk sumber terlatih 85%, sumber tidak terlatih 81,5%, dan pengujian dengan Kinect sumber tidak terlatih 84% sehingga sistem pengenalan kata dapat diimplementasikan dengan teknologi Kinect. Kata Kunci : speech processing, HMM, MFCC, kinect
Aplikasi Identifikasi Huruf Braille Menggunakan Computer Vision Berbasis Raspberry Pi Ramiati Ramiati; Siska Aulia; Lifwarda Lifwarda
JURNAL NASIONAL TEKNIK ELEKTRO Vol 9, No 1: March 2020
Publisher : Jurusan Teknik Elektro Universitas Andalas

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (749.348 KB) | DOI: 10.25077/jnte.v9n1.707.2020

Abstract

Sense of vision is a source of information on humans. Some humans are created with limited sense of sight. The blind performs reading and writing activities using Braille letters, a printed code system consisting of six dots in various combinations that are highlighted on the paper so that they can be touched. To facilitate the visually impaired and the public in enjoying the works produced by blind people, a script reading system is characterized by Braille by studying the braille characters in advance of each character. This research makes a braille letter identification system into sound using computer vision. The method, the reading of Braille character scripts by studying braille characters. First, a scanner or Raspberry Pi-based camera captures braille characters. Second, the system converts Braille characters into alphabetical shapes by processing Optical Character Recognition images. Recognition of Braille character patterns in written text using Artificial Neural Networks. The results of research on braille testing are in the form of alphabetical texts a through z, and the sound signal of the pronunciation of the alphabet uses the Text To Speech system. Braille to sound conversion system works well, with an average accuracy of system testing of 88.462%. This condition is achieved by using 70 gsm HVS paper and drawing paper with a 52 training image database. The system can only carry out the process of recognition of one character, so it can be used as a reference translator of audio-based braille characters that can be heard by the visually impaired and the community.Keywords : image processing, braille, OCR, JST, text to speechAbstrakIndera penglihatan merupakan sumber informasi pada manusia. Sebagian manusia diciptakan dengan keterbatasan indera penglihatan. Tunanetra melakukan aktifitas membaca serta menulis menggunakan huruf Braille, yaitu sistem cetakan berupa kode terdiri dari enam titik dalam berbagai kombinasi yang ditonjolkan pada kertas sehingga dapat diraba. Untuk memudahkan tunanetra dan masyarakat dalam menikmati karya-karya yang dihasilkan oleh penyandang tunanetra dibuat sistem pembacaan naskah berkarakterkan Braille dengan mempelajari karakter braille terlebih dahulu dari masing-masing karakternya. Penelitian ini membuat sistem identifikasi huruf braille menjadi suara menggunakan computer vision. Metodenya, pembacaan naskah berkarakter Braille dengan mempelajari karakter braille. Pertama, scanner atau kamera berbasis Raspberry Pi mengcapture karakter braille. Kedua, sistem mengkonversi karakter Braille ke bentuk abjad dengan pengolahan citra Optical Character Recognition. Pengenalan pola karakter Braille teks tulisan menggunakan Jaringan Saraf  Tiruan. Hasil penelitian pengujian huruf braille berupa teks abjad a sampai z, dan sinyal suara pengucapan abjad menggunakan sistem Text  To Speech . Sistem konversi braille menjadi suara bekerja dengan baik, dengan akurasi rata-rata pengujian sistem yaitu 88.462%. Kondisi ini dicapai dengan menggunakan kertas HVS 70 gsm dan kertas gambar dengan database 52 citra latih. Sistem hanya dapat melakukan proses pengenalan pada satu karakter, sehingga dapat digunakan sebagai referensi penterjemah naskah karakter braille berbasis audio yang dapat didengarkan oleh tunanetra dan masyakat.Kata Kunci : image processing, braille, OCR, JST, text to speech