Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal Jurnal Teknik Elektro dan Komputer TRIAC Journal of Renewable Energy and Smart Device
Farendi Giotivano
Universitas Negeri Surabaya
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering Energy Engineering

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Domain Gap and Context Bias Analysis in YOLOv11-Based Housing Condition Detection through Indoor–Outdoor Cross-Domain Evaluation Farendi Giotivano; Wiyli Yustanti; Ricky Eka
Journal of Renewable Energy and Smart Device Vol. 4 No. 1 August 2026
Publisher : PT. Global Research Collaboration

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.66314/joresd.v4i1.998

Abstract

Computer vision provides a promising solution to evaluate housing conditions. However, the differences between indoor and outdoor visual environments may cause domain gaps that affect the generalization ability of the model. This study explores the effect of domain gaps on the detection of housing physical conditions using You Only Look Once version 11 (YOLOv11) by proposing an integrated evaluation framework that combines cross-domain evaluation, EigenCAM-based model interpretability, and illumination variation analysis. A total of 2,712 housing images were classified into Indoor, Outdoor and Mixed datasets and evaluated under uniform experimental conditions. From the results, the Indoor model obtained a 42.77% relative mAP50 drop when evaluated on the Outdoor dataset, thus proving the existence of a significant domain gap. The Mixed model achieved the most consistent performance across all evaluation scenarios, although this improvement may also be partially influenced by the larger training dataset used in the Mixed configuration. EigenCAM analysis showed that single-domain models were more reliant on contextual visual cues while the Mixed model was consistently more attuned to relevant housing elements. The illumination analysis showed that the differences in brightness between the domains were not large, suggesting that context bias is more likely than illumination variation to explain the observed performance degradation. These results demonstrate that the combination of cross-domain evaluation with model interpretability provides a more complete understanding of domain gap effects and enables the creation of more robust computer vision models for housing condition assessment.
Evaluasi Dampak Kompresi dan Resolusi Citra terhadap Kinerja YOLOv11 pada Verifikasi Hunian Farendi Giotivano; Wiyli Yustanti; Ricky Eka
Jurnal Teknik Elektro dan Komputer TRIAC Vol 13, No 2 (2026): Oktober 2026
Publisher : Jurusan Teknik Elektro Universitas Trunojoyo Madura

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21107/triac.v13i2.34940

Abstract

The implementation of computer vision in educational assistance verification systems is frequently hindered by high storage requirements and heavy server bandwidth consumption. Despite this, empirical research mapping the tolerance threshold of image degradation against object detection reliability in this domain remains limited. This study aims to evaluate the impact of JPEG compression and resolution reduction on the performance of the YOLOv11 model in identifying residential physical components, specifically roofs, walls, and floors. We conducted evaluations using various JPEG quality scenarios and image resolution adjustments. The experimental results demonstrate that YOLOv11 exhibits high robustness to compression. At a 75% quality level, file size was reduced by up to 70.6% while maintaining stable performance, achieving a precision of 0.7605, recall of 0.6651, mAP50 of 0.7341, and an inference time of 8.80 milliseconds. Regarding the resolution experiments, a 480-pixel dimension provided the most optimal outcome, yielding a precision of 0.7309, recall of 0.6711, mAP50 of 0.7333, and an inference time of 3.60 milliseconds. Conversely, extreme resolution reduction triggered a significant drop in accuracy. These findings confirm that the YOLOv11 architecture is more tolerant of compression quality than it is of resolution scaling. We recommend a 75% JPEG compression or a 480-pixel resolution as a standard operational baseline for efficient and accurate residential verification.
Co-Authors Ricky Eka Ricky Eka Wiyli Yustanti Wiyli Yustanti