Article Per Year (5 Year)
p-Index From 2021 - 2026
1.323
P-Index
This Author published in this journals
All Journal Jurnal Infra
Kartika Gunadi
Program Studi Informatika
Computer Science & IT

Published : 21 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 21 Documents
Search

 1   2   3 
Deteksi Balon Ucapan Pada Komik Jepang Dengan Convolutional Neural Network, Canny Edge Detection dan Run Length Smooth Algorithm Ricky Setiawan Saswono; Rudi Adipranata; Kartika Gunadi
Jurnal Infra Vol 8, No 2 (2020)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Comic is an entertainment media that is usually used to fill free time. Comics themselves are already very well known in the world, especially comics from Japan. Comics from Japan, commonly called Manga, have a high level of popularity. The proof is a lot of Manga that is translated into each country's language. Examples such as One Piece that has been circulating in 43 countries. Even so the translation process is quite long especially in Japanese translation.This research can be used to accelerate the translation process by using CNN and Canny Edge Detection to detect speech balloons in Manga. The detection results are segmented and with the help of OCR to digitize Japanese characters. Then use copy-paste techniques in an online dictionary or online translator to find the meaning of letters that are not understood. Because searching for letters from a physical dictionary (book) takes more time.The results of the research to segment the speech balloon from Manga were successful but to classify the image in the form of a speech balloon or not with CNN was unsuccessful. Researchers assume because the dataset created is small in number or a problem during pre-processing.
Sistem Pakar Diagnosa Penyakit Kura – Kura Air Dengan Metode Certainty Factor Berbasis Mobile Zachary Osborn; Kartika Gunadi; Leo Willyanto Santoso
Jurnal Infra Vol 9, No 2 (2021)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Turtle is on of the few animals that many people has interest in for pets,because turtles have this unique and beautiful shells. Turtle usually have a long age, butfor tortoise it is usually longer. In average tortoise can live up to 70 to 100 years. But stillall living things can get sick, including turtles. One of the actions that must be taken by water turtle keepers when their pet is sick is to know the symptoms and disease. Onealternative to find out what diseases our pets have is to go to the vet. But the price that veterinarians offer is relatively inexpensive.This expert system for diagnosis of water turtle disease is equipped with the Certainty Factor method. The usefulness of the Certainty factor method in this program is to display the level of system confidence in the diagnostic results in the form of a percentage. So that later serves to convince users when using this program.Based on the test results, this application can provide solutions that are suitable for diseases associated with symptoms. The results of the comparison between the Certainty Factor method and the Dempster Shafer method also give the same results in cases where the numbers used are the best numbers. The average figure resulting fromthis calculation is 83.6%.
Implementasi Tesseract OCR untuk Pembuatan Aplikasi Pengenalan Nota pada Android Yoel Andreas; Kartika Gunadi; Anita Nathania Purbowo
Jurnal Infra Vol 8, No 1 (2020)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The development of a practical era makes humans more inclined to find a fast way to do something. The same thing when we want to record the expenses we have spent in the day, of course it takes time to do it. To solve this problem, you can use the application to read the receipt using the Android device's camera, the application can help to record expenses and categorize their expenses. To achieve this, it is necessary to do Optical Character Recognition which can be done using Tesseract-OCR. The results will be processed to get expenses, categories, and item names. To get maximum results, several stages of pre-processing are needed on the image to be used. The test is carried out with a scenary study and tried several cases, for example notes with dotted fonts, or notes that have many lines. The test results show that the OCR results from the Tesseract are very dependent on the pre-processing stage being carried out. Tesseract itself will experience a decrease in performance when processing images with dotted fonts. If the pre-processing stage cannot unite separate letters due to dots, the tesseract has a very drastic decrease in accuracy. Notes with multiple lines also reduce the performance of the tesseract. The results of the tesseract when conducting Handwritten Character Recognition are also affected by how the handwriting are written, if the handwriting is cursive or not neat, then the tesseract will have difficulty in carrying out the HCR process.
Rendering Karakter 3D Virtual secara Real-Time menggunakan Metode Light Estimation pada Augmented Reality Berbasis Lokasi Kevin Kevin; Liliana Liliana; Kartika Gunadi
Jurnal Infra Vol 9, No 2 (2021)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Augmented reality applications are already widely available on mobile devices, but most augmented reality applications assume that light source always comes from above the object and its direction is always downwards so that the shadow is always right under the object, therefore a method is needed to estimate light so that the direction of shadow produced is more realistic, but can still be run on mobile devices.To answer the problem, light estimation method is used in real-time rendering of AR applications on mobile devices so that the shadow direction from virtual objects rendering is parallel and in the same direction as the shadow direction of real objects in their environment, but still uses resources that can be used on mobile devices.Results in this study indicate that the direction of shadow produced by light estimation method in indoor environment is quite accurate (about 33°) and light enough to be used on mobile devices, because the difference in FPS and RAM usage is almost the same as the usage of application without the use of light estimation method, although there is an increase in CPU and battery usage, it's small enough to still work on a mobile device.
Deteksi Alat Pelindung Diri Menggunakan Metode YOLO dan Faster R-CNN Jonathan Adiwibowo; Kartika Gunadi; Endang Setyati
Jurnal Infra Vol 8, No 2 (2020)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

In order to ensure security and safety measures in industrial zones or any other areas that needed to use personal protective equipment are an important matter. Many workers keep disregarding and violating the rule to use personal protective equipment in the area. Therefore, a program was created that could help supervising the workers to use personal protective equipment. In this study an experiment will be conducted to help recognize the characteristics of personal protective equipment, especially in head. In recent studies that have been carried out Rifki Dita Wahyu Pradana, et. Al. using CNN to produce an overall accuracy 80%. This study will be using 2 methods, You Only Look Once and Faster Region-Convolutional Neural Network (Faster R-CNN). The YOLO method is used to find regions of worker’s head while Faster R-CNN method is used to classify personal protective equipment used by worker. The results of the Faster R-CNN classification will be calculated using a confusion matrix in order to get the accuracy of the correct prediction. The results from this study will identify workers using personal protective equipment in the video. Average accuracy that has been obtained is 93.61%.
Deteksi Rumus Matematika pada Halaman Dokumen Digital dengan Metode Convolutional Neural Network Martina Marcelline Taslim; Kartika Gunadi; Alvin Nathaniel Tjondrowiguno
Jurnal Infra Vol 7, No 2 (2019)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1485.281 KB)

Abstract

Mathematical formulae in academic papers or scientific journals are an important part of said documents. However, mathematical formulae are oftentimes not properly recognized by Optical Character Recognition (OCR) processes. One of the causes of this failure is the difference between mathematical formulae and ordinary text. Therefore, mathematical formula detection in those document pages might help with this problem. The formula detection is done by converting digital document pages into images, then performing text line segmentation and word segmentation and classifying those results with a Convolutional Neural Network. The aim is to help OCR processes by recognizing which parts of the document pages contain formulae and which parts do not. The CNN architectures used to perform classification comes with 64 kernels in each convolutional layer. For displayed formulae (formulae that doesn’t share its space with regular text), the model uses 10 groups of Convolutional-ReLU-Max Pooling layers. For inline formulae (formulae that shares its text line with regular text), 12 groups of Convolutional-ReLU-Max Pooling layers are used. Results of the CNN architectures mentioned above are an F1 score of 0,980 for displayed formulae classification in 1-column documents, 0,940 for 2-column documents, and 0,916 for inline formulae. 
Rancang Bangun Sistem Pakar dalam Menentukan Resiko Penyakit Jantung Koroner Susangto Andi Putera S.; Silvia Rostianingsih; Kartika Gunadi
Jurnal Infra Vol 8, No 1 (2020)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Currently, the use of expert system applications has been widely used by many people, and is supported by a very significant development. So it's easy to use by many people and has many benefits to support life today. With this application, ordinary people can easily learn the expertise of an expert so that it can be applied in everyday life. In this application, an expert system is applied to analyze the risk of a person to develop coronary heart disease and be able to do prevention as early as possible. In making this application, the programmer consulted experts (doctors) in making knowledge base (basis pengetahuan). Based on data test results, the application can work well and easily in its use. Application can provide good analysis results in diagnosing the possibility of cardiovascular disease using the parameters provided.Currently, the use of expert system applications has been widely used by many people, and is supported by a very significant development. So it's easy to use by many people and has many benefits to support life today. With this application, ordinary people can easily learn the expertise of an expert so that it can be applied in everyday life. In this application, an expert system is applied to analyze the risk of a person to develop coronary heart disease and be able to do prevention as early as possible. In making this application, the programmer consulted experts (doctors) in making knowledge base (basis pengetahuan). Based on data test results, the application can work well and easily in its use. Application can provide good analysis results in diagnosing the possibility of cardiovascular disease using the parameters provided.
Pengenalan Penyakit pada Tanaman Pokok di Indonesia dengan Metode Convolutional Neural Network Handy Prayoga Angjaya; Kartika Gunadi; Rudy Adipranata
Jurnal Infra Vol 9, No 2 (2021)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

In Indonesia, the most consumed staple food is rice, the rice comes from the rice plants, besides rice, there is also cassava and corn plants. The success of harvesting these crops can affect the country's food welfare, but pest and diseases can cause crop failure. Therefore, a program was created to identify diseases in these plants to maximize crop yields. In the process of identifying the disease, the problem that often occurs is the identification of the characteristics of the disease. With the development of technology, disease recognition can be done automatically using a Neural Network. This study uses the Convolutional Neural Network (CNN) method with the Inception v3 architecture. In addition, the model used will be converted using TensorFlow Lite so that it can be used on Android-based smartphone applications. The program results from this study identified diseases in maize, potato, cassava, and rice plants. Based on the tests carried out, an average accuracy value of 90.77% was obtained in testing data test. Testing in the field actually produces an average accuracy of 65.00%.
Implementasi Program Presensi Mahasiswa Dengan Menggunakan Face Recognition Richard Lawrence Thiosdor; Kartika Gunadi; Lily Puspa Dewi
Jurnal Infra Vol 9, No 1 (2021)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The problem of using a physical attendance list causes a cheat where the student does “fake attendance” by asking another students to sign the attendance list on his/her behalf. This problems are often found in college activities.Detection of student faces uses the Face Recognition library as a mean of validation in the attendance check process. Face recognition requires face images that have been preprocessed and uses the K-Nearest Neighbor model (KNN) or Support Vector Machine (SVM) to validate student faces in the attendance check process.Testing on 15 sample face images with 40 total face classes yields an average accuracy of 99%. Face Recognition cannot detect faces if the facial features are obstructed. This validation of student attendance successfully uses Face Recognition to minimize cheating in taking attendance.
Sistem Pakar Diagnosa Penyakit Saraf Menggunakan Metode Forward Chaining dan Certainty Factor Lucky Alexandre Lembangan; Kartika Gunadi; Alexander Setiawan
Jurnal Infra Vol 9, No 2 (2021)
Publisher : Jurnal Infra

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Neurological diseases are one of the public health problems that requires special policies in an effort to handle it so that complete data are needed regarding cause, developments and outcomes. Neurological diseases consist of various types of nerves. Most people today tend to ignore or less in response to disorders that occur in the nervous system. After all, the neurological system plays a very important role in all human activities, because if the slightest symptom or disturbance is ignored, it can have serious consequence. As technology becomes more sophisticated, therefore in the future this research is expected to help replace the role of a doctor to diagnose early symptoms in the neurological system which will be implemented in a system called an expert system. This neurological disease diagnosis expert system is equipped with Forward channeling and Certainty factor methods. The usefulness of forward chaining in this program is to collect facts that occur to the user so that later they produce conclusions, so that users do not need to answer all the questions. By selecting the existing symptoms, you will get a conclusion that is a neurological disease that is owned by the user. The usefulness of the Certainty factor in this program is to display the level of system confidence in the diagnostic results in the form of a percentage. So that later serves to convince users when using this program. Based on the test results, this program can provide solutions that are suitable for diseases related to the symptoms felt by the user. The results of the calculation of the Certainty factor obtained quite significant results when compared with the results of interviews with experts.
Co-Authors Albert Albert Alexander Setiawan Alexander William Sutjiadi Alvin Nathaniel Tjondrowiguno Anita Nathania Purbowo Benediktus Marcelino Pratama Debby Meliani Prayogo Devina Yolanda Endang Setyati Endang Setyati Eric Tirtana Handy Prayoga Angjaya Henry Novianus Palit Indar Sugiarto Jonathan Adiwibowo Kevin Kevin Leo Willyanto Santoso Liliana Liliana Lily Puspa Dewi Lucky Alexandre Lembangan Lukas Fernando Hunggianto Made Yoga Mahardika Martina Marcelline Taslim Richard Lawrence Thiosdor Ricky Herwanto Ricky Setiawan Saswono Rudi Adipranata Rudy Adipranata Samuel Njoo Silvia Rostianingsih Susangto Andi Putera S. Timothy Christian Yunanto Yoel Andreas Zachary Osborn