Proceedings of the

The 33rd European Safety and Reliability Conference (ESREL 2023)
3 – 8 September 2023, Southampton, UK

Multilevel Artificial Intelligence Classification of Faulty Image Data for Enhancing Sensor Reliability

Omar Mohammed1,a, Parthib Khound1,b, Peng Su2,d, Dejiu Chen2,e and Frank Gronwald1,c

1Chair of Reliability of Technical Systems and Electrical Measurement, University of Siegen, Siegen, Germany.

2Unit of Mechatronics and Embedded Control Systems, KTH Royal Institute of Technology, 100 44 Stockholm, Sweden.


A multi-stage classification algorithm is proposed to predict the fault type and its associated intensity level of a camera input frame to enhance the reliability of a camera-based system. A fault injecting tool is used to generate the dataset required for the training. The model architecture mainly comprises three convolutions neural network (CNN) layers and three fully connected layers. The model achieves 93.8% accuracy for predicting a fault type. For the fault intensity prediction the accuracy significantly varies for each fault type but for some faults, the model achieves a very good prediction accuracy. However, for some other faults the accuracy can be remarkably low. The primary reason for this gap is that the intensity levels of all considered faults can be described in a sufficiently quantitative way, i.e., there is no sufficient metric available so far.

Keywords: Fault classification, Fault type, Fault intensity, Artificial intelligence, Camera-based system reliability, Smart camera sensor.

Download PDF