No abstract

No abstract

Defect inspection is paramount within the closed-loop manufacturing system. However, existing datasets for defect inspection often lack precision and semantic granularity required for practical applications. In this paper, we introduce the Defect Spectrum, a comprehensive benchmark that offers precise, semanticabundant, and large-scale annotations for a wide range of industrial defects. Building on four key industrial benchmarks, our dataset refines existing annotations and introduces rich semantic details, distinguishing multiple defect types within a single image. Furthermore, we introduce Defect-Gen, a two-stage diffusion-based generator designed to create high-quality and diverse defective images, even when working with limited datasets. The synthetic images generated by Defect-Gen significantly enhance the efficacy of defect inspection models. Overall, The Defect Spectrum dataset demonstrates its potential in defect inspection research, offering a solid platform for testing and refining advanced models. Our project page is in https://envision-research.github.io/Defect_Spectrum.

No abstract

Abstract. Aiming at the fulfilling the requirements of security supervision and evaluation of machines and tools in transmission and transformation projects, China Electric Power Research Institute(CEPRI) develops a set of high efficiency of supervision and evaluation system with Microsoft Access. This article introduces the development principle and process, the structure and the function of the system in detail, which includes two parts database and software. The system has been put in use and achieves the original goals.