IM-IAD
收藏资源简介:
IM-IAD是一个专为工业制造领域设计的图像异常检测基准,由南方科技大学计算机科学与工程系的研究团队开发。该数据集包含7个主要数据集,涵盖了多种工业应用场景,如产品表面异常检测、纺织品缺陷检测和食品检验等。IM-IAD旨在通过提供一个统一的评估平台,推动工业图像异常检测技术的发展,解决现有算法在实际应用中的性能差异问题。数据集的创建过程涉及对多个工业数据集的整合与标准化,确保了数据的一致性和可用性。该数据集主要应用于工业自动化检测领域,旨在提高生产效率和产品质量,解决工业生产中的异常检测问题。
IM-IAD is an image anomaly detection benchmark specifically designed for the industrial manufacturing domain, developed by the research team from the Department of Computer Science and Engineering at Southern University of Science and Technology. This dataset comprises seven primary sub-datasets, covering diverse industrial application scenarios such as product surface anomaly detection, textile defect detection, and food inspection. IM-IAD aims to advance the development of industrial image anomaly detection technologies by providing a unified evaluation platform, addressing the performance discrepancy issue of existing algorithms in real-world applications. The construction of this dataset involves the integration and standardization of multiple industrial datasets, ensuring the consistency and availability of the data. This dataset is primarily applied in the field of industrial automatic inspection, with the objectives of enhancing production efficiency and product quality, and resolving anomaly detection problems in industrial production.

- 1IM-IAD: Industrial Image Anomaly Detection Benchmark in Manufacturing南方科技大学计算机科学与工程系 · 2024年



