Some ancient documents in Indonesia are written in the Javanese script. Those documents contain the knowledge of history and culture of Indonesia, especially about Java. However, only a few people understand the Javanese script. Thus, the automation system is needed to translate the document written in the Javanese script. In this study, the researchers use the classification method to recognize the Javanese script written in the document. The method used is the Multiclass Support Vector Machine (SVM) using One Against One (OAO) strategy. The researchers use seven variations of Javanese script from the different document for this study. There are 31 classes and 182 data for training and testing data. The result shows good performance in the evaluation. The recognition system successfully resolves the problem of color variation from the dataset. The accuracy of the study is 81.3%.
2) ABSTRAK Buku-buku kuno Bahasa Jawa memiliki konten kekayaan intelektual Indonesia seperti agama, linguistik, filosofi, mitos, pelajaran moral, hukum dan norma adat, kerajaan, cerita rakyat, sejarah, dan lain sebagainya. Tidak banyak yang mempelajari karya tersebut karena ditulis dengan Aksara Jawa dan tidak banyak yang memahami. Untuk membantu penerjemahan dokumen berbahasa Jawa dilakukan otomatisasi sistem penerjemahan. tahap penerjemahan terdiri dari segmentasi untuk mendapatkan karakter dari citra tulisan dalam naskah Aksara Jawa. Kemudian tiap karakater dikenali sebagai abjad. Dan yang terakhir adalah mengkombinasikan tulisan latin yang telah dikenali menjadi kata yang berarti. Penelitian yang membahas tentang pengenalan Aksara Jawa telah dilakukan, seperti fokus pada segmentasi karakter dan pengenalan Aksara Jawa. Pada penelitian sebelumnya dilakukan perbaikan pada metode segmentasi namun tetap mendapatkan hasil yang sama dalam hal akurasi kebenaran. Pada penelitian kali ini diusulkan penggunaan metode Histogram of Oriented Gradient (HOG) pada tahap ekstraksi fitur. Metode HOG banyak digunakan pada pengenalan dan deteksi objek pada data citra. Metode HOG juga pernah digunakan untuk mengenali tulisan tangan berbahasa Inggris dan Huruf Bengali dengan hasil yang optimal. Pada penelitian ini didapatkan hasil akurasi pengenalan karakter Aksara Jawa sebesar 93.3%.
EEG (electroencephalogram) can detect epileptic seizures by neurophysiologists in clinical practice with visually scan long recordings. Epilepsy seizure is a condition of brain disorder with chronic noncommunicable that affects people of all ages. The challenge of study is how to develop a method for signal processing that extract the subtle information of EEG and use it for automating the detection of epileptic with high accuration, so we can use it for monitoring and treatment the epileptic patient. In this study we developed a method to classify the EEG signal based on Wavelet Packet Decomposition that decompose the EEG signal and Random Forest for seizure detetion. The result of study shows that Random Forest classification has the best performance than KNN, ANN, and SVM. The best combination of statisctical features is standard deviation, maximum and minimum value, and bandpower. WPD is has best decomposition in 5th level.
Automation of Javanese script translation is needed to make it easier for people to understand the meaning of ancient Javanese script. By using Javanese script image as input, the translation system generally consists of character segmentation, character recognition, and combining the recognized characters as a meaningful word. The segmentation which obtains region of interest of each character, is an important process in the translation system. In the previous research, segmentation using projection profile method can separate each character well. The method can overcome characters overlapping, but it still produces truncated characters. In this study, we proposed a new segmentation to reduce the truncated character. The first step of the proposed method is pre-processing that consists of converting input into binary image and cleaning noises. The next step is to determine the connected component labels, which further perform as candidate of characters. Some of the candidates are still represented by more than one labels, so that we need a process to merge the connected component labels that have centroid distance less than threshold. We evaluate the proposed method using Intersection over Union (IoU). The evaluation shows the best accuracy 93,26%.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.