Search papers, labs, and topics across Lattice.
This paper introduces a novel thermal fault detection method for high-speed direct-driven blowers by fusing thermal and visible images and applying semantic segmentation. An improved denoising diffusion model is used for multimodal image fusion, incorporating a perceptually prioritized weighted loss and an alternate training strategy. A lightweight segmentation network is then employed for component segmentation, followed by temperature level segmentation using clustering methods, demonstrating improved fault temperature detection on blower components.
Fusing denoising diffusion models with lightweight semantic segmentation offers a fast and accurate way to detect thermal faults in industrial blower components using thermal and visible imagery.
A thermal fault detection method for high-speed direct-driven blower components is proposed, using thermal and visible image fusion along with semantic segmentation. The proposed method follows three steps: multimodal image fusion, component semantic segmentation of the fused image, and temperature level segmentation. First, an end-to-end image fusion network based on an improved denoising diffusion model is used, a perceptually prioritized weighted loss is introduced for training, and an alternate training strategy is used to improve the quality of the fused images. In the second step, a lightweight segmentation network is proposed to reduce the model size and inference time while improving the segmentation accuracy. Finally, thermal images are processed by clustering methods. Experiments on real industrial objects show that the proposed method composed of infrared and optical image fusion, semantic segmentation, and temperature clustering networks improves significantly the fault temperature detection on different blower components.
