Adaptive Contextual Embedding for Robust Far-View Borehole Detection

TL;DR

This paper introduces an adaptive detection method for small, densely packed boreholes in far-view imagery, combining embedding stabilization, adaptive augmentation, and contextual refinement to improve detection accuracy in challenging industrial environments.

Contribution

It presents a novel adaptive detection framework that enhances existing architectures with EMA-based embedding stabilization and context-aware features for robust borehole detection.

Findings

Significant accuracy improvements over baseline YOLO models.

Robust detection under varying illumination and texture conditions.

Effective in complex industrial quarry scenarios.

Abstract

In controlled blasting operations, accurately detecting densely distributed tiny boreholes from far-view imagery is critical for operational safety and efficiency. However, existing detection methods often struggle due to small object scales, highly dense arrangements, and limited distinctive visual features of boreholes. To address these challenges, we propose an adaptive detection approach that builds upon existing architectures (e.g., YOLO) by explicitly leveraging consistent embedding representations derived through exponential moving average (EMA)-based statistical updates. Our method introduces three synergistic components: (1) adaptive augmentation utilizing dynamically updated image statistics to robustly handle illumination and texture variations; (2) embedding stabilization to ensure consistent and reliable feature extraction; and (3) contextual refinement leveraging spatial context for improved detection accuracy. The pervasive use of EMA in our method is particularly advantageous given the limited visual complexity and small scale of boreholes, allowing stable and robust representation learning even under challenging visual conditions. Experiments on a challenging proprietary quarry-site dataset demonstrate substantial improvements over baseline YOLO-based architectures, highlighting our method's effectiveness in realistic and complex industrial scenarios.