Improving DNN fault tolerance in semantic segmentation applications - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Improving DNN fault tolerance in semantic segmentation applications

Résumé

Semantic segmentation of images is essential for autonomous driving and modern DNNs now achieve high accuracy. Automotive systems must comply with safety standards, requiring hardware fault detection. We present an analysis of the effect of faults using Google’s DeepLabV3+ network processing an industrial data-set. A new symptom-based fault detection algorithm is shown to detect >99% of critical faults with zero false positives and a compute overhead of 0.2%. Further, these faults can be masked, virtually eliminating all critical errors. To the authors’ knowledge this is the first fault tolerance study of a DNN semantic segmentation application.
Fichier non déposé

Dates et versions

hal-03960998 , version 1 (28-01-2023)

Identifiants

Citer

Stéphane Burel, Adrian Evans, Lorena Anghel. Improving DNN fault tolerance in semantic segmentation applications. IEEE International Symposium on Defect and Fault Tolerance in VLSI and Nanotechnology Systems (DFT), Oct 2022, Austin, United States. pp.1-6, ⟨10.1109/DFT56152.2022.9962354⟩. ⟨hal-03960998⟩
56 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More