Article Per Year (5 Year)
p-Index From 2020 - 2025
1.203
P-Index
This Author published in this journals
All Journal JAIS (Journal of Applied Intelligent System) Jurnal Ilmu Kepolisian Journal of Applied Data Sciences Jurnal Indonesia Sosial Teknologi
Moch Arief Soeleman, Moch Arief
Unknown Affiliation
Humanities Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Environmental Science Law, Crime, Criminology & Criminal Justice Social Sciences

Published : 6 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 6 Documents
Search

 1 
Conditional Matting For Post-Segmentation Refinement Segment Anything Model Susanto, Al Birr Karim; Soeleman, Moch Arief; Budiman, Fikri
(JAIS) Journal of Applied Intelligent System Vol. 8 No. 3 (2023): Journal of Applied Intelligent System
Publisher : LPPM Universitas Dian Nuswantoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33633/jais.v8i3.9024

Abstract

Segment Anything Model (SAM) is a model capable of performing object segmentation in images without requiring any additional training. Although the segmentation produced by SAM lacks high precision, this model holds interesting potential for more accurate segmentation tasks. In this study, we propose a Post-Processing method called Conditional Matting 4 (CM4) to enhance high-precision object segmentation, including prominent, occluded, and complex boundary objects in the segmentation results from SAM. The proposed CM4 Post-Processing method incorporates the use of morphological operations, DistilBERT, InSPyReNet, Grounding DINO, and ViTMatte. We combine these methods to improve the object segmentation produced by SAM. Evaluation is conducted using metrics such as IoU, SAD, MAD, Grad, and Conn. The results of this study show that the proposed CM4 Post-Processing method successfully improves object segmentation with a SAD evaluation score of 20.42 (a 27% improvement from the previous study) and an MSE evaluation score of 21.64 (a 45% improvement from the previous study) compared to the previous research on the AIM-500 dataset. The significant improvement in evaluation scores demonstrates the enhanced capability of CM4 in achieving high precision and overcoming the limitations of the initial segmentation produced by SAM. The contribution of this research lies in the development of an effective CM4 Post-Processing method for enhancing object segmentation in images with high precision. This method holds potential for various computer vision applications that require accurate and detailed object segmentation.
Enhancing Augmentation-Based Resnet50 for Car Brand Classification Sugiarto, Triga Agus; Soeleman, Moch Arief; Pujiono, Pujiono
(JAIS) Journal of Applied Intelligent System Vol. 8 No. 3 (2023): Journal of Applied Intelligent System
Publisher : LPPM Universitas Dian Nuswantoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33633/jais.v8i3.9385

Abstract

This research focuses on car classification and the use of the ResNet-50 neural network architecture to improve the accuracy and reliability of car detection systems. Indonesia, as one of the countries with high daily mobility, has a majority of the population using cars as the main mode of transportation. Along with the increasing use of cars in Indonesia, many automotive industries have built factories in this country, so the cars used are either local or imported. The importance of car classification in traffic management is a major concern, and vehicle make and model recognition plays an important role in traffic monitoring. This study uses the Vehicle images dataset which contains high-resolution images of cars taken from the highway with varying viewing angles and frame rates. This data is used to analyze the best- selling car brands and build car classifications based on output or categories that consumers are interested in. Digital image processing methods, machine learning, and artificial neural networks are used in the development of automatic and real-time car detection systems.The ResNet-50 architecture was chosen because of its ability to overcome performance degradation problems and study complex and abstract features from car images. Residual blocks in the ResNet architecture allow a direct flow of information from the input layer to the output layer, overcoming the performance degradation problem common in neural networks. In this paper, we explain the basic concepts of ResNet-50 in car detection and popular techniques such as optimization, augmentation, and learning rate to improve performance and accuracy. in this study, it is proved that ResNet has a fairly high accuracy of 95%, 92% precision, 93% recall, and 92% F1-Score.
Person Re-Identification Using CNN Method With Combination of SVM and Semantic Segmentation Kurniawan, Kristian Adhi; Soeleman, Moch Arief
(JAIS) Journal of Applied Intelligent System Vol. 9 No. 1 (2024): Journal of Applied Intelligent System
Publisher : LPPM Universitas Dian Nuswantoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62411/jais.v9i1.10345

Abstract

Abstract – Person re-identification is a mechanized procedure of video investigation which has been widely studied in contemporary years. Research problems that are often raised in the field of a person's re-identification research are characteristic representations that are easily affected by closure (abhorrent to other objects). Furthermore, after extracting local features by means of a boundary box, the background image still contains and does not focus on the human body parts. This study comes up with a method combination of CNN, SVM classification, and semantic segmentation. CMC (Cumulative Matching Characteristics) and mAP (mean Average Precision) are measurements of assessment that will be utilized to measure the operation of re-identification. The ResNet + SVM + SSP-ReID technique performed best in the Market dataset, with a CMC increase of 3-10% (rank-1 through rank-20). The Market and CUHK03 (D) datasets both showed improvements of 1-4.1% in mAP.  Keywords Person re-identification; Feature extraction; CNN; SVM; Semantic segmentation;
HU Variance Moment Optimizes Keyframe Selection Based on Deep Learning for Violence Detection Putri, Sukmawati Anggraeni; Andono, Pulung Nurtantio; Purwanto, Purwanto; Soeleman, Moch Arief
Journal of Applied Data Sciences Vol 6, No 2: MAY 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i2.648

Abstract

Violence in public spaces poses a serious threat to individuals and society. Manual monitoring and violence detection require much time and human resources, ultimately hindering detection accuracy and speed. Therefore, an automated method is needed to detect violence to ensure fast and efficient action. Along with technological advances, violence detection research has adopted various methods and models, including deep learning, such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs). In this study, the classification process for detecting violence and non-violence uses the VGG19 model, one of the CNN models that has good performance with limited computing. In addition, the Long Short-Term Memory (LSTM) model is the best RNN model for processing temporal data in videos. However, this performance will decrease with noise and irrelevant data in the classification process. Therefore, to optimize deep learning performance, this study in the pre-processing phase selects keyframes in frame extraction using the Hu Variance Moment Technique. This method calculates each frame’s Hu and Variance Moment values and selects keyframes based on high Hu values. Next, we use Adaptive Moment Estimation (Adam) to optimize the gradient of the selected keyframes. This study produces a Hu19LSTM model tested on three datasets: hockey fight, crowd, and AIRTLab. The proposed Hu19LSTM model produces an accuracy of 97% on the Hockey Fight dataset, 97% on the Crowd dataset, and 95% on the AIRTLab dataset. These results indicate that the Hu19LSTM model can increase its accuracy on the hockey fight and Crowd dataset by 97%.
A Machine Learning Approach to Predicting Digital Leadership Success in Indonesian Police Officers Muchtar, Jarot; Soeleman, Moch Arief
Jurnal Ilmu Kepolisian Vol 19 No 2 (2025): Jurnal Ilmu Kepolisian Volume 19 Nomor 2 Tahun 2025
Publisher : Sekolah Tinggi Ilmu Kepolisian

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35879/jik.v19i2.646

Abstract

We develop a predictive model to enhance digital leadership among Indonesian National Police (Polri) officers, addressing the pressing need for technological proficiency in modern law enforcement. Digital leadership, vital for combating cyber threats and improving operational efficiency, remains underdeveloped in Polri due to limited technological skills and a lack of systematic leadership identification. We train machine learning models on 564 anonymized officer records, incorporating attributes like rank, position, and education, guided by Transformational, Adaptive, and Contingency Leadership theories. The Light GBM model excels, achieving an F1 Score of 0.9674, a Log Loss of 0.2244, a Cohen's Kappa of 0.9616, and a Matthews Correlation Coefficient of 0.9620, demonstrating high predictive accuracy. This model empowers Polri to identify officers with strong digital leadership potential, enabling targeted training programs and strategic personnel selection to drive digital transformation. We prioritize ethical deployment by excluding sensitive attributes, such as religion and gender, to mitigate bias and employ k-anonymity to safeguard data privacy. Fairness audits and interpretable outputs ensure equitable and transparent decision-making. Our approach aligns with global policing trends, offering a scalable solution to enhance leadership in tech-driven environments. By integrating robust technical performance with ethical safeguards, this study contributes to Polri’s strategic goals and sets a foundation for future research in diverse policing contexts. We advocate for continuous model monitoring to sustain fairness and effectiveness in real-world applications.
Deteksi Dini Covid-19 Melalui Citra CT-Scan Paru-Paru Menggunakan K-Nearest Neighbor dengan Komparasi Jarak Maknun, Lu'luul; Syukur, Abdul; Affandy, Affandy; Soeleman, Moch Arief
Jurnal Indonesia Sosial Teknologi Vol. 3 No. 03 (2022): Jurnal Indonesia Sosial Teknologi
Publisher : Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1049.395 KB) | DOI: 10.59141/jist.v3i03.397

Abstract

Covid -19 yang telah mewabah dan menjadi pandemik secara global yang merupakan masalah utama yang perlu di perhatikan dan di tangani, beberapa cara yang harus di lakukan adalah dengan memutus mata rantai penyebaran virus salah satunya dengan melakukan deteksi dini dan melakukan karantina, dengan CT scan paru-paru. CT scan paru-paru dapat dijadikan jalan alternatif. Berdasarkan permasalahan di atas maka peneliti mengetahui kondisi paru-paru secara detail dan dalam mendiagnosis virus secara dini. Pada penelitian ini pendekatan yang di ajukan menggunakan metode K-NN dengan perhitungan jarak euclidean distance, manhattan distance, miskowski distance untuk deteksi dini Covid -19 melalui citra CT scan paru-paru yang di duga terinfeksi Covid -19 . dalam mendeteksi secara dini evaluasi yang di gunakan untuk mengetahui pervorma yang di usulkan menggunakan coufusion matrix dengan hasil eksperimen menunjukkan hasil dari tiga perhitungan jarak menunjukkan hasil akurasi yang baik dan menggunakan dataset secara publik yaitu euclidean distance berjumlah 83%, Manhattan distance berjumlah 87%, Minkowski berjumlah 76%, di harapkan metode ini dapat di gunakan dan di kembangkan untuk melengkapi dioglosa medis.
Co-Authors Abdul Syukur Affandy Affandy Fikri Budiman Jarot Muchtar Kurniawan, Kristian Adhi Maknun, Lu'luul Pujiono Pujiono Pulung Nurtantio Andono Purwanto Purwanto Sugiarto, Triga Agus Sukmawati Anggraeni Putri, Sukmawati Anggraeni Susanto, Al Birr Karim