Abstract: Airglow imaging observations, with their high spatiotemporal resolution and large-scale continuous monitoring capability, provide a crucial approach to studying the fine horizontal structures and evolutionary characteristics of ionospheric equatorial plasma bubbles. However, the current lack of high-quality, professionally annotated plasma bubble datasets severely restricts the application of supervised Artificial Intelligence (AI) algorithms in this field. To address this issue, this study constructs the first standardized dataset of ionospheric plasma bubbles based on airglow observations, including plasma bubble event data products and precise contour annotation data products. The dataset is derived from continuous observations over a full solar activity cycle (2012－2022) by a 630 nm band airglow imager at the Qujing Station in Yunnan, China. All raw data underwent standardized preprocessing, including image enhancement, azimuth correction, geometric distortion correction, and geographic coordinate projection. Expert teams then performed plasma bubble event identification and contour annotation. With a high temporal resolution of 3 minutes, the dataset systematically documents plasma bubble events under varying solar activity intensities, covering multiple typical morphologies such as shaped I- and Y-shaped structures. This dataset provides high-quality benchmark data for developing high-precision supervised AI algorithms, facilitating automated detection and morphological evolution research of ionospheric plasma bubbles based on airglow imaging.