A DCNN-based Arbitrarily-Oriented Object Detector for Quality Control and Inspection Application
This work addresses a domain-specific problem for automated quality control in industrial and medical settings, presenting an incremental improvement over existing methods.
The paper tackles the problem of detecting arbitrarily-oriented objects in quality control and inspection by proposing a two-stage DCNN-based detector, achieving improved detection performance for small and elongated targets in applications like surgery toolboxes and vessel hull defects.
Following the success of machine vision systems for on-line automated quality control and inspection processes, an object recognition solution is presented in this work for two different specific applications, i.e., the detection of quality control items in surgery toolboxes prepared for sterilizing in a hospital, as well as the detection of defects in vessel hulls to prevent potential structural failures. The solution has two stages. First, a feature pyramid architecture based on Single Shot MultiBox Detector (SSD) is used to improve the detection performance, and a statistical analysis based on ground truth is employed to select parameters of a range of default boxes. Second, a lightweight neural network is exploited to achieve oriented detection results using a regression method. The first stage of the proposed method is capable of detecting the small targets considered in the two scenarios. In the second stage, despite the simplicity, it is efficient to detect elongated targets while maintaining high running efficiency.