A Potential Vision-Based Measurements Technology: Information Flow Fusion Detection Method Using RGB-Thermal Infrared Images

Vision-based measurements (VBM) technology has been widely applied in the quality monitoring of various products. However, most of the existing studies only focus on the defect detection methods using a single-modal image (RGB image or Thermal Infrared image). In order to detect surface and internal defects more comprehensively, this article provides a potential defect detection technology introducing the RGB-Thermal infrared salient object detection (RGB-T SOD) into VBM. A novel information flow fusion network (IFFNet) method is proposed for the RGB-T cross-modal images. The proposed IFFNet consists of an information filtering module and a novel information flow paradigm. Validation on three available RGB-T SOD datasets shows that our proposed method performs more competitively than the state-of-the-art (SOTA) methods. For the key evaluation metric W_F, the experimental results of our method are 0.849, 0.912, and 0.856 on VT821, VT1000, and VT5000 datasets, respectively.

Automated summary

This article introduces a new defect detection technology that combines RGB and thermal infrared images. The proposed information flow fusion network (IFFNet) method can detect surface and internal defects more comprehensively. Experimental results show that our method performs better than existing methods on three RGB-Thermal infrared datasets.