A DCNN-based Arbitrarily-Oriented Object Detector for Quality Control and Inspection Application

TL;DR

This paper introduces a two-stage deep convolutional neural network approach for detecting arbitrarily oriented objects in quality control and inspection tasks, combining a feature pyramid SSD architecture with a lightweight orientation regression network.

Contribution

It proposes a novel two-stage detection framework that enhances small target detection and oriented object recognition in industrial inspection applications.

Findings

Effective detection of small targets in complex scenarios

High efficiency in detecting elongated, oriented objects

Improved detection performance over traditional methods

Abstract

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.