Claim Missing Document
Check
Articles

Found 35 Documents
Search

Age Estimation Untuk Intelligent Advertising Pada Poster Digital Menggunakan Convolutional Neural Network Muhyi, Galih Abdul; Wihardi, Yaya; Piantari, Erna
JATIKOM: Jurnal Aplikasi dan Teori Ilmu Komputer Vol 7, No 1 (2024)
Publisher : Universitas Pendidikan Indonesia (UPI)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.17509/jatikom.v7i1.27878

Abstract

Sebagai bagian dari intelligent advertising, age estimation digunakan untuk menyesuaikan iklan dari hasil estimasi usia audience. Age estimation (AE) dapat dibangun menggunakan deep learning menggunakan ConvNet dengan kendala seperti data training wajah usia tua yang sedikit dan ketidak seimbangan dataset di dalamnya serta membutuhkan jumlah data yang besar. Salah satu solusi dari permasalahan ini adalah melakukan data augmentasi menggunakan model generatif ACGAN untuk melakukan generate gambar sesuai dengan kelas. Intelligent advertising pada poster digital hanya disimulasikan pada komputer. Simulasi intelligent advertising berfungsi dengan baik terlepas dari terbatasnya iklan dan tidak konsistennya hasil estimasi usia. Hasil dari penggunaan model generatif ACGAN untuk data augmentation berhasil meningkatkan performa hasil pada model AE terlepas dari rendahnya skor IS dan FID serta kualitas gambar yang dihasilkan. Hasil data augmentation lebih terlihat pada model B dengan peningkatan akurasi cumulative score sebesar 4,8% dan skor MAE sebesar 1,297.
Implementation of Computational Thinking in Data Structure Subject Using Problem-based Learning Models Rafli, Muhammad; Sutarno, Heri; Wihardi, Yaya
Jurnal Guru Komputer Vol 4, No 2 (2023): JGrKom: September 2023
Publisher : Universitas Pendidikan Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.17509/jgrkom.v4i2.30450

Abstract

Computational thinking skills are one of the skills required in the 21st century. Computational thinking can teach students to think logically, critically, and use appropriate and efficient strategies in solving problems. However, computational thinking skills require an appropriate learning approach. Learning process must be designed to hone problem-solving skills and to acquire essential knowledge and concepts of the subject matter. This study aims to develop a computational approach media with thinking and problem-based learning and to improve students cognitive. This study uses a Comprehensive Life Cycle (SHM) methodology to develop the media. One-group pretest-postest was used for the research design. Media with computational thinking and problem-based learning approach have been developed and has obtained a score of 90% or "Very Good" by experts on media testing and get a percentage score of 92% or "Very Good" by experts on material testing. The use of media with Computational Thinking and problem-based learning approach can improve student cognitive with an increase in the average N-gain of students by 0.42 which is in the "Medium" category.  On the other hand, student responses to the learning media resulted in a score of 84%.
Augmented Reality in STEM Using Personalized Learning to Promote Student’s Understanding Mukhlis, Rizki; Erlangga, Erlangga; Wihardi, Yaya; Raflesia, Sarifah Putri
Computer Engineering and Applications Journal Vol 13 No 2 (2024)
Publisher : Universitas Sriwijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.18495/comengapp.v13i2.473

Abstract

The current curriculum highlights the premise of self-directed learning performed by students. Additionally, technological uses in educational settings prove to be a challenging task in a sense of implementing them in learning media and materials used in the classroom. This study aims at investigating the utilization of augmented reality (AR) in STEM (Science, Mathematics, Engineering, and Technology) using personalized learning. This study employed pre-experimental research design, specifically adopting One-Group Pretest-Posttest Design. The findings highlight that students’ pretest scores on average reached 51,6 and significantly improved to 82,67 in their posttest, whereas students’ gain score reached 0,64 which is considered as moderate. Their perspectives towards the use of augmented reality with personalized learning were significantly positive with the percentage of 82,1%. It is evident that the use of augmented reality with personalized learning is a viable option when it comes to affecting the learning outcomes.
Exemplar Based Convolutional Neural Network for Face Search on CCTV Video Recording Winda Mauli Kristy; Yaya Wihardi; Erlangga Erlangga
Journal of Computers for Society Vol 4, No 2 (2023): JCS: September 2023
Publisher : Universitas Pendidikan Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.17509/jcs.v4i2.71185

Abstract

Many techniques can perform effective face searches, but generally, these methods require numerous samples, particularly when using deep learning approaches. However, there are scenarios where face searches must be conducted with limited samples, such as those obtained from CCTV video recordings, making prior training infeasible. In these situations, a method based on exemplars must be implemented. This investigation utilizes a convolutional neural network (CNN) approach coupled with two unique matching techniques: cross-correlation matching (CCM) and normalized cross-correlation matching (NCC). The study makes use of the Chokepoint Face Dataset, training the data through the optimization of triplet loss. The goal of the study is to evaluate the performance of these combined methods. Two different architectures are created and tested within each method to determine the accuracy of each architecture. The CNN-NCC method has been found to yield accuracy rates that surpass those of the CNN-CCM method by 2 to 17.9%. Nevertheless, it is important to note that the accuracy of the results is greatly influenced by the variations observed in the CCTV video recordings.
Detection of Motorcycles on Highways Using Faster R-CNN Based on VGG16 Moch Dian Lazuardi Yudha; Wawan Setiawan; Yaya Wihardi
Journal of Computers for Society Vol 4, No 1 (2023): JCS: June 2023
Publisher : Universitas Pendidikan Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.17509/jcs.v4i1.71179

Abstract

This research aims to develop an object detection system with input in the form of image data in free size. The development of an object detection system model was carried out by applying Machine Learning to overcome object detection in an image using the Faster R-CNN method based on the VGG16 algorithm. The system developed produces a bounding box for an object in the image. System development was carried out in the Python programming language by utilizing several libraries such as Keras. Experiments were carried out by measuring the loss value of the training data entered into the system. The experimental results show that the resulting information is proven to be able to detect objects in a given image. This system can produce information based on image data that has been trained with this system. This study used two experiments which obtained a loss value of 0.0601 in the first study and 0.1211 in the second study.
Klasifikasi Penyakit Daun Tanaman Tomat Menggunakan EfficientFormer Wihardi, Yaya; Erlangga, Erlangga; Himmah, Amirah Dzatul; Siregar, Herbert
Jurnal Aplikasi dan Teori Ilmu Komputer Vol 7, No 2 (2024): September 2024, Jurnal Aplikasi dan Teori Ilmu Komputer (Jatikom)
Publisher : Universitas Pendidikan Indonesia (UPI)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.17509/jatikom.v7i2.80339

Abstract

Tomatoes are crucial crops for Indonesian farmers, but they often suffer from diseases caused by fungi, bacteria, and viruses, leading to potential 40% yield loss. Current methods of spotting these diseases by eye result in costly and ineffective use of pesticides. This study focuses on a new way to classify tomato plant diseases using EfficientFormer. This method aims for high accuracy and fast inference time. The model reached an impressive 92% accuracy and takes just 0.4 seconds to identify diseases. This new approach could help farmers spot tomato plant diseases more accurately and quickly, potentially reducing economic losses and excessive pesticide use in Indonesia.
Augmented Reality in STEM Using Personalized Learning to Promote Students’ Understanding Erlangga; Mukhlis, Rizki; Wihardi, Yaya; Raflesia, Sarifah Putri
Computer Engineering and Applications Journal (ComEngApp) Vol. 13 No. 2 (2024)
Publisher : Universitas Sriwijaya

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

Abstract

The current curriculum highlights the premise of self-directed learning performed by students. Additionally, technological uses in educational settings prove to be a challenging task in a sense of implementing them in learning media and materials used in the classroom. This study aims at investigating the utilization of augmented reality (AR) in STEM (Science, Mathematics, Engineering, and Technology) using personalized learning. This study employed pre-experimental research design, specifically adopting One-Group Pretest-Posttest Design. The findings highlight that students’ pretest scores on average reached 51,6 and significantly improved to 82,67 in their posttest, whereas students’ gain score reached 0,64 which is considered as moderate. Their perspectives towards the use of augmented reality with personalized learning were significantly positive with the percentage of 82,1%. It is evident that the use of augmented reality with personalized learning is a viable option when it comes to affecting the learning outcomes.
The Development of Web-Based Learning using Interactive Media for Science Learning on Levers in Human Body Topic Astuti, Lia; Wihardi, Yaya; Rochintaniawati, Diana; Prima, Eka Cahya
Journal of Science Learning Vol 3, No 2 (2020): Journal of Science Learning
Publisher : Universitas Pendidikan Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.17509/jsl.v3i2.19366

Abstract

Integrated curriculum is a popular way to develop 21st-century skills, but most of the materials are written on the books separately. Furthermore, web-based learning is an online learning media that can be accessed using an internet connection anytime and anywhere. However, many educational websites do not apply the principles of effective learning. Besides, the traditional learning methods tend to be bored for the students. In order to solve this problem, this study designed a website education that uses interactive content to assist students in learning levers in the human body topic as one of the integrated science materials. The process of developing an educational website consists of three steps: (1) analysis, (2) design, (3) construction making. This research method used descriptive method, and the experts' judgment will evaluate it on content, language, and media/IT. The questionnaires used the technology acceptance model (TAM) and five-dimensional interactivity to investigate the readability of the subjects' perception responses. The research subject was three science teachers and 31 students on private Junior High School in Bandung. According to the result, generally, it has a good evaluation of each aspect. But, the website education needs a strong signal to access for not taking load time consumer.
Deteksi Aksi Kekerasan pada Video CCTV Berbasis Skeleton dan Frame Grouping Menggunakan ConvLSTM Kafilli, Muhammad Fikri; Riza, Lala Septem; Wihardi, Yaya
JATISI Vol 12 No 3 (2025): JATISI (Jurnal Teknik Informatika dan Sistem Informasi)
Publisher : Universitas Multi Data Palembang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35957/jatisi.v12i3.12873

Abstract

Manual monitoring of surveillance video (CCTV) is inefficient and prone to human oversight. This drives the need for an automated violence detection system that is fast and accurate. Existing deep learning models are often too computationally heavy for real-time implementation, creating a dilemma between accuracy and efficiency. This research proposes a lightweight two-stream ConvLSTM architecture to address this dilemma. The method efficiently models spatio-temporal relationships by combining skeleton representation and change detection, which is then packaged through a frame grouping technique. The ConvLSTM layer serves as the main temporal model, supported by a SeparableConv2D backbone for efficient feature extraction. The model is trained on the RWF-2000 dataset and evaluated using cross-dataset validation on the Surveillance Camera Fight Dataset to test its generalization capability. The results show that the proposed model achieves superior performance with an accuracy and F1-Score of 74.00%, and is highly efficient with an inference speed of 518.45 FPS. This research demonstrates that the two-stream architecture combining skeleton representation, frame grouping, and ConvLSTM modeling successfully creates a robust, fast violence detection system, offering a practical solution for real-world monitoring applications.
Klasifikasi Pose Kepala Siswa Menggunakan EfficientNetV2 dengan Seat Position Embedding Marhelio, Ananda Myzza; Munir, Munir; Wihardi, Yaya
JATISI Vol 12 No 3 (2025): JATISI (Jurnal Teknik Informatika dan Sistem Informasi)
Publisher : Universitas Multi Data Palembang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35957/jatisi.v12i3.12987

Abstract

Understanding students’ visual attention direction is essential for evaluating their engagement during classroom learning. Head Pose Estimation (HPE) is an effective method for identifying attention focus, however, its application in real classroom settings is often hindered by low image quality and varied student seating positions, which makes regression-based methods for predicting facial landmarks or Euler angles suboptimal. This study adopts an image-based classification approach as an alternative and proposes a modification of the EfficientNetV2-S architecture by integrating Seat Position Embedding (SPE) as spatial context to improve the accuracy of head pose classification. The dataset was developed from direct classroom recordings and processed into 4,574 head pose images with five directional labels (up, down, front, right, left). Several CNN architectures were evaluated with and without SPE. The results show that the proposed model with SPE achieved an accuracy of 83.25%, surpassing the baseline model’s accuracy of 82.53%. This approach has proven effective in reducing visual ambiguity and providing a more accurate interpretation of students' attention.