Evaluation of the bounding box uncertainty of deep-learning object detection in HALCON software | IEEE Conference Publication | IEEE Xplore

Evaluation of the bounding box uncertainty of deep-learning object detection in HALCON software


Abstract:

Deep neural networks have become more and more relevant for vision systems, for object detection and classification in industrial fields, such as robot navigation, monito...Show More

Abstract:

Deep neural networks have become more and more relevant for vision systems, for object detection and classification in industrial fields, such as robot navigation, monitoring and tracking. For such applications, vision systems have to be robust to environment conditions, occlusions and very accurate, as for bin picking. In this paper, we evaluate the performances of deep learning object detection neural networks in HALCON software, by investigating the uncertainty of bounding box position for object detection and the impact of disturbances. In this study, results evidenced the increase of bounding box uncertainty and the reduction of confidence of neural networks when disturbances are introduced, as well as the increment of uncertainty, when confidence lowers. When errors are introduced in labeling, the uncertainty of the bounding box position becomes higher, but lower than the error introduced.
Date of Conference: 03-05 June 2020
Date Added to IEEE Xplore: 10 July 2020
ISBN Information:
Conference Location: Roma, Italy

I. Introduction

The application of vision systems to industrial fields is often related to object classification, object detection and tracking, for bin picking and robot navigation. On production lines, these systems require low uncertainty, high robustness and low runtime respect to the phenomenon to detect. In this scenario, it is essential to evaluate performances of each tool before applying it to a specific industrial application.

References

References is not available for this document.