Inoculation Powder Segmentation Dataset

Description

This dataset, Inoculation Powder Segmentation Dataset (IPSD), contains a collection of images captured during the casting and inoculant powder spraying process in an industrial production line. The main objective of this dataset is to offer valuable resources for training artificial intelligence models aimed at real-time monitoring of the inoculation process in the casting industry. The IPSD consists of 4,286 images, each with a resolution of 512 × 512 pixels. These images are meticulously annotated with two distinct classes: powder and background. The segmentation of these classes was performed by human experts to ensure the accuracy and reliability of the dataset. This careful annotation process is crucial for the successful training of machine learning models designed to distinguish between inoculant powder and background objects in the images. Additionally, the dataset provides a wide range of images captured under various conditions, which can help improve the robustness of models when applied to real-world scenarios. The dataset has been carefully curated to include high-quality images that reflect the complexity and variability of the casting process, ensuring that the models trained on this data can generalize well to different operational environments.

Institutions

Categories

Funding

Licence