Explaining Defect Detection with Saliency Maps

in 34th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems, IEA/AIE 2021, Kuala Lumpur, Malaysia, July 26–29, 2021, Proceedings, Part II (2021, July 19)

The rising quality and throughput demands of the manufacturing domain require flexible, accurate and explainable computer-vision solutions for defect detection. Deep Neural Networks (DNNs) reach state-of ... [more ▼]

The rising quality and throughput demands of the manufacturing domain require flexible, accurate and explainable computer-vision solutions for defect detection. Deep Neural Networks (DNNs) reach state-of-the-art performance on various computer-vision tasks but wide-spread application in the industrial domain is blocked by the lacking explainability of DNN decisions. A promising, human-readable solution is given by saliency maps, heatmaps highlighting the image areas that influence the classifier’s decision. This work evaluates a selection of saliency methods in the area of industrial quality assurance. To this end we propose the distance pointing game, a new metric to quantify the meaningfulness of saliency maps for defect detection. We provide steps to prepare a publicly available dataset on defective steel plates for the proposed metric. Additionally, the computational complexity is investigated to determine which methods could be integrated on industrial edge devices. Our results show that DeepLift, GradCAM and GradCAM++ outperform the alternatives while the computational cost is feasible for real time applications even on edge devices. This indicates that the respective methods could be used as an additional, autonomous post-classification step to explain decisions taken by intelligent quality assurance systems. [less ▲]

An Eclipse Modelling Framework Alternative to Meet the Models@Runtime Requirements

in Model Driven Engineering Languages and Systems - 15th International Conference (2012, October)

Models@Runtime aims at taming the complexity of software dynamic adaptation by pushing further the idea of reflection and considering the reflection layer as a first-class modeling space. A natural ... [more ▼]

Models@Runtime aims at taming the complexity of software dynamic adaptation by pushing further the idea of reflection and considering the reflection layer as a first-class modeling space. A natural approach to Models@Runtime is to use MDE techniques, in particular those based on the Eclipse Modeling Framework. EMF provides facilities for building DSLs and tools based on a structured data model, with tight integration with the Eclipse IDE. EMF has rapidly become the defacto standard in the MDE community and has also been adopted for building Models@Runtime platforms. For example, Frascati (implementing the Service Component Architecture standard) uses EMF for the design and runtime tooling of its architecture description language. However, EMF has primarily been thought to support design-time activities. This paper highlights specific Models@Runtime requirements, discusses the benefits and limitations of EMF in this context, and presents an alternative implementation to meet these requirements. [less ▲]