Claim Missing Document
Check
Articles

Found 3 Documents
Search

Optimasi Imperceptibility pada Steganografi Video Adaptif LSB Multi-Bit Berbasis SVM Fiky Anggara; Rianindya Chandra Hardika; Zaini; Ilham Maridi
Jurnal Nasional Komputasi dan Teknologi Informasi Vol. 9 No. 2 (2026): April, 2026
Publisher : Program Studi Teknik Komputer, Fakultas Teknik. Universitas Serambi Mekkah

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32672/p8x48y49

Abstract

Abstrak - Steganografi video berbasis Least Significant Bit (LSB) merupakan metode yang banyak digunakan karena kesederhanaan implementasinya. Namun, pendekatan konvensional masih berpotensi menimbulkan distorsi visual yang berdampak pada penurunan imperceptibility. Penelitian ini mengusulkan metode steganografi video LSB multi-bit adaptif berbasis Support Vector Machine (SVM) untuk mengoptimalkan imperceptibility dalam proses penyisipan data. Pendekatan adaptif dilakukan dengan memanfaatkan fitur karakteristik citra, seperti tekstur dan kompleksitas piksel, yang diekstraksi dari setiap frame video. Selanjutnya, model SVM digunakan untuk mengklasifikasikan area frame ke dalam kategori wilayah halus dan kompleks, sehingga jumlah bit penyisipan dapat ditentukan secara dinamis. Area dengan tingkat kompleksitas tinggi dialokasikan jumlah bit yang lebih besar, sedangkan area halus dibatasi untuk meminimalkan distorsi visual. Evaluasi kinerja metode dilakukan menggunakan metrik Peak Signal-to-Noise Ratio (PSNR) dan Structural Similarity Index (SSIM). Hasil eksperimen menunjukkan bahwa metode yang diusulkan mampu menghasilkan kualitas visual video stego yang tinggi dengan nilai PSNR rata-rata sebesar 46,42 dB dan SSIM rata-rata sebesar 0,9427, yang mengindikasikan tingkat kemiripan visual yang sangat baik antara video cover dan video stego. Dengan demikian, metode yang diusulkan efektif dalam mengoptimalkan imperceptibility tanpa mengorbankan kapasitas penyisipan data secara signifikan. Kata kunci : Steganografi video; LSB multi-bit; Support Vector Machine (SVM); imperceptibility; Abstract - Video steganography based on the Least Significant Bit (LSB) method is widely used due to its simplicity of implementation. However, conventional approaches still have the potential to introduce visual distortion, which affects imperceptibility. This study proposes an adaptive multi-bit LSB video steganography method based on Support Vector Machine (SVM) to optimize imperceptibility in the data embedding process. The adaptive approach is carried out by utilizing image characteristic features, such as texture and pixel complexity, extracted from each video frame. Furthermore, the SVM model is employed to classify frame regions into smooth and complex areas, enabling the number of embedding bits to be determined dynamically. Regions with higher complexity are allocated a greater number of embedding bits, while smooth regions are limited to minimize visual distortion. The performance of the proposed method is evaluated using the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) metrics. Experimental results show that the proposed method is capable of producing high-quality stego video with an average PSNR value of 46.42 dB and an average SSIM value of 0.9427, indicating a very high level of visual similarity between the cover video and the stego video. Therefore, the proposed method is effective in optimizing imperceptibility without significantly compromising data embedding capacity. Keywords: Video steganography; multi-bit LSB; Support Vector Machine (SVM); imperceptibility;
Pendampingan Penyusunan Dokumen Usulan Jabatan Fungsional Pertama Bagi Dosen Tetap Sekolah Tinggi Teknologi Bontang Zaini Zaini; Hardianto; Lapu Tombil Layuk; Fiky Anggara; Ilham Maridi; Inas Umniyah
Jurnal Pengabdian Masyarakat - PIMAS Vol. 5 No. 3 (2026): Agustus
Publisher : LPPM Universitas Harapan Bangsa Purwokerto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35960/pimas.v5i3.2642

Abstract

Functional academic positions play an essential role in lecturers' academic career development and contribute to improving the quality of higher education institutions. However, limited understanding of the applicable regulations, the preparation of the Credit Score Proposal List (DUPAK), the classification of evidence related to the Tri Dharma of Higher Education, and administrative document requirements remain major challenges in the submission process. To address these issues, a participatory mentoring approach was implemented through five stages: needs identification, socialization, document preparation assistance, document review and verification, and evaluation. The mentoring program was designed to enhance lecturers' understanding and competence in preparing proposal documents in accordance with the applicable regulations, thereby improving their readiness to apply for their first functional academic position. The success of the program was evaluated based on improvements in participants' understanding, their ability to prepare proposal documents, and the completeness of the required submission documents. The program successfully improved participants' understanding of academic promotion regulations, submission procedures, the preparation of the Credit Point Proposal (DUPAK), and credit point calculation. All participants (100%) demonstrated a clear understanding of the promotion process, while three lecturers (50%) successfully completed their first academic promotion application for the Assistant Professor rank. The remaining three participants were unable to submit their applications due to the unavailability of the second-semester Lecturer Workload Report (BKD) for the 2025/2026 academic year, which is a mandatory administrative requirement. Overall, the program enhanced lecturers' readiness to apply for their first academic rank and contributed to strengthening academic career development at Sekolah Tinggi Teknologi Bontang.
Photovoltaic Power Output Prediction Using PCA–KNN: A Comparative Study with SVR and MLP on Historical Operational Data Ilham Maridi; Fiky Anggara; Martati Martati
Jurnal Media Elektrik Vol. 23 No. 3 (2026): MEDIA ELEKTRIK
Publisher : Jurusan Pendidikan Teknik Elektro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59562/metrik.v23i3.14355

Abstract

The transition toward renewable energy integration in power systems presents challenges in managing photovoltaic (PV) power output owing to variations in operating conditions. This study evaluated the performance of K-Nearest Neighbor (KNN), Support Vector Regression (SVR), and Multi-Layer Perceptron (MLP), with and without Principal Component Analysis (PCA), for PV power output prediction. The dataset comprised 420 observations recorded at five-minute intervals over seven days of measurements from a PV system. The data were chronologically divided into 80% training and 20% testing sets, while normalization and PCA were exclusively fitted to the training data to prevent information leakage. The model performance was evaluated using the Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and coefficient of determination (R²). The results show that KNN and PCA-KNN achieved the best and identical performance, with an MSE of 265.3385, RMSE of 16.2892, and R² of 0.8421. PCA did not improve KNN accuracy but maintained its predictive performance after the dimensional transformation. PCA-MLP showed a slight improvement over MLP, whereas SVR and PCA-SVR yielded lower performances. These findings indicate that the effect of PCA depends on both the dataset characteristics and the learning algorithm. Further studies with longer observation periods, more diverse datasets, and external validation are recommended to assess the generalizability of the models.