Bring the Power of Diffusion Model to Defect Detection

TL;DR

This paper introduces a novel defect detection method that leverages diffusion models to enhance semantic understanding, improving accuracy in identifying challenging surface defects in industrial settings.

Contribution

It integrates diffusion models with a feature repository and a residual auto-encoder, employing dynamic cross-fusion and knowledge distillation to boost detection performance without extra efficiency costs.

Findings

Achieves competitive results on industrial defect datasets.

Enhances detection of non-salient and difficult defects.

Maintains efficiency with lightweight models.

Abstract

Due to the high complexity and technical requirements of industrial production processes, surface defects will inevitably appear, which seriously affects the quality of products. Although existing lightweight detection networks are highly efficient, they are susceptible to false or missed detection of non-salient defects due to the lack of semantic information. In contrast, the diffusion model can generate higher-order semantic representations in the denoising process. Therefore, the aim of this paper is to incorporate the higher-order modelling capability of the diffusion model into the detection model, so as to better assist in the classification and localization of difficult targets. First, the denoising diffusion probabilistic model (DDPM) is pre-trained to extract the features of denoising process to construct as a feature repository. In particular, to avoid the potential bottleneck of memory caused by the dataloader loading high-dimensional features, a residual convolutional variational auto-encoder (ResVAE) is designed to further compress the feature repository. The image is fed into both image backbone and feature repository for feature extraction and querying respectively. The queried latent features are reconstructed and filtered to obtain high-dimensional DDPM features. A dynamic cross-fusion method is proposed to fully refine the contextual features of DDPM to optimize the detection model. Finally, we employ knowledge distillation to migrate the higher-order modelling capabilities back into the lightweight baseline model without additional efficiency cost. Experiment results demonstrate that our method achieves competitive results on several industrial datasets.