Disparity Image Segmentation For ADAS

TL;DR

This paper introduces an efficient method for segmenting disparity images in automotive systems by customizing connected component labeling and region merging, suitable for embedded architectures and prioritizing nearer objects.

Contribution

It presents a novel adaptation of CCL algorithms for disparity image segmentation, optimized for embedded automotive applications.

Findings

Effective segmentation of disparity images demonstrated on standard datasets.

Implementation achieves real-time performance on embedded automotive hardware.

Prioritizes nearer objects for improved ADAS functionality.

Abstract

We present a simple solution for segmenting grayscale images using existing Connected Component Labeling (CCL) algorithms (which are generally applied to binary images), which was efficient enough to be implemented in a constrained (embedded automotive) architecture. Our solution customizes the region growing and merging approach, and is primarily targeted for stereoscopic disparity images where nearer objects carry more relevance. We provide results from a standard OpenCV implementation for some basic cases and an image from the Tsukuba stereo-pair dataset.