Abstract:Intelligent defect detection systems based on deep learning have substantial potential to be applied in segmenting defects of aluminum casting parts' X-ray images. The success of data-based methods relies on well-established datasets that require time and labor to perform the annotation processes necessary to build them. In efforts to improve the efficiency of image annotation, many deep learning-oriented interactive segmentation types of research have been conducted with a single natural object that has high semantic information. However, these approaches are not suitable for defects with low semantic information in X-ray images because the defects have characteristics such as blurred boundaries, multiple scales, random distribution in one region, and low contrast. To fill this gap in the field of deep learning-based interactive segmentation for X-ray images, we conduct this study. First, static and dynamic point sampling strategies are proposed to generate foreground and background points to guide model training. Second, we propose an iterative training procedure to help the model perceive both global and point-specified information stage by stage. And third, an interactive X-ray network (IXNet) with a click attention module is proposed to achieve the interactive segmentation for the multiple casting defects with low semantic information in one region simultaneously. The experimental results prove the effectiveness of the proposed methods: the iterative training process improved by 13% and 5% in mNoC (75%) and mNoC (80%), respectively compared with the pure training process, and IXNet with the click attention module can achieve 2.38 and 3.96 in mNoC (75%) and mNoC (80%), respectively.

Electrode Defects Edge Detection Based on Adaptive Canny Operator

Automatic Recognition and Classification of Eddy Current Testing Signals for Scanning Inspection of Defect in Multi-Layered Structures

Detection Of Small Surface Defects Using Dct Based Enhancement Approach In Machine Vision Systems - Art. No. 60400h

Study on Flaw Labeling and Edge Tracking in Ultrasonic C-Scan Image

Recognition algorithm on bearing surface defect type

Defects’ geometric feature recognition based on infrared image edge detection

Orange edge detection based on adaptive Canny operator

Defects Detection of Lithium-Ion Battery Electrode Coatings Based on Background Reconstruction and Improved Canny Algorithm

Method of Imaging and Evaluation for Defect Based on Pulsed Eddy Current

An Improved Canny Edge Detection Algorithm

Improved faster R-CNN and adaptive Canny algorithm for defect detection using eddy current thermography

Scale Adaptive Canny Edge Detection Method

Defective Edge Detection Using Cascaded Ensemble Canny Operator

Edge detection algorithm of medical image based on Canny operator

Ultrasound Image Edge Detection Based on a Novel Multiplicative Gradient and Canny Operator.

Toward a More Robust Canny for Edge Detection

The Weld Image Edge-Detection Algorithm Combined with Canny Operator and Mathematical Morphology

Interactive Defect Segmentation in X-Ray Images Based on Deep Learning

Edge Detection of the Defects in UItrasonic C Scan Image Based on MorphoIogy

Improved Canny Edge Detection Operator Based on Multi-scale Adaptive Weighting Algorithm

An Improved Edge Extraction Algorithm for Lithography Micrographs