A Skeleton-based Approach For Rock Crack Detection Towards A Climbing Robot Application

TL;DR

This paper introduces SKIL, a skeleton-based loss function for thin object segmentation, specifically for detecting rock cracks to aid climbing robots, demonstrating improved performance over previous methods.

Contribution

The paper presents SKIL, a novel skeleton intersection loss for thin object segmentation, and a new evaluation metric LineAcc, tailored for thin structures like rock cracks.

Findings

SKIL outperforms previous segmentation methods on rock crack detection.

Models trained with SKIL show improved accuracy on thin object segmentation tasks.

The proposed metrics are less sensitive to object width and translation effects.

Abstract

Conventional wheeled robots are unable to traverse scientifically interesting, but dangerous, cave environments. Multi-limbed climbing robot designs, such as ReachBot, are able to grasp irregular surface features and execute climbing motions to overcome obstacles, given suitable grasp locations. To support grasp site identification, we present a method for detecting rock cracks and edges, the SKeleton Intersection Loss (SKIL). SKIL is a loss designed for thin object segmentation that leverages the skeleton of the label. A dataset of rock face images was collected, manually annotated, and augmented with generated data. A new group of metrics, LineAcc, has been proposed for thin object segmentation such that the impact of the object width on the score is minimized. In addition, the metric is less sensitive to translation which can often lead to a score of zero when computing classical metrics such as Dice on thin objects. Our fine-tuned models outperform previous methods on similar thin object segmentation tasks such as blood vessel segmentation and show promise for integration onto a robotic system.