Synthetic and Derived Training Images for Campus Waste Detection: A Multi-Seed Evaluation with YOLOv8n

2026-07-21

AI Summary

This study addresses the issue of recycling contamination in campus waste sorting caused by incorrect disposal, proposing a real-time object detection method based on YOLOv8n. The work introduces a novel multi-seed-guided evaluation framework to systematically quantify the performance gains and uncertainties associated with data augmentation strategies—such as background replacement and hand synthesis—under small-sample conditions. Experimental results show that a model trained solely on real images achieves an mAP@0.5 of 0.691, and none of the tested augmentation strategies significantly outperform this baseline. Furthermore, due to the limited scale of the test set, category-level conclusions should be interpreted with caution.

Abstract

Incorrect disposal can contaminate campus recycling streams, and a bin-mounted camera could provide feedback as an item is discarded. We evaluated whether synthetic and derived images improve a YOLOv8n detector for this view. The real dataset contained 148 campus photographs: 86 for training, 31 for validation, and 31 for testing. Twelve joint-training configurations varied the amount and source of added images. We repeated seven principal settings with four matched seeds and computed bootstrap percentile intervals over those seeds. The real-only model reached a mean mAP@0.5 of 0.691 [0.665, 0.722]. Background replacement reduced the mean to 0.560 [0.499, 0.619], isolated-object images gave 0.680 [0.644, 0.724], and the full augmentation pool gave 0.487 [0.438, 0.537]. We also tested hand-and-forearm composites because every real photo showed a held object. Two cutouts in the initial composite set came from test photographs, so we discarded that experiment, rebuilt the set with training-split cutouts, and reran all four seeds. The corrected paired difference was +0.034 [-0.063, 0.199], which does not support a reliable hand-composite effect. Single-seed transfer experiments produced source-dependent rankings between joint mixing and sequential pretraining. None of the evaluated configurations exceeded the real-only baseline. The reported intervals quantify seed variation; the 31-photo test set remains too small for strong class-specific conclusions.

Problem

• campus waste detection
• synthetic training images
• YOLOv8n
• data augmentation
• limited real dataset

Innovation

• synthetic data
• data augmentation
• YOLOv8n
• multi-seed evaluation
• waste detection