 The paper proposes a novel deep learning approach for automatic identification of Noctiluca Sintolens blooms from satellite imagery. The proposed methodology combines two state-of-the-art deep learning techniques, inception convolutional blocks, ICB, and SWIN attention blocks, SAB, to achieve high accuracy in identifying NSB from satellite images. The results demonstrate that this approach outperforms other commonly used semantic segmentation methods, achieving a precision of 92.22 percent, recall of 88.2 percent, F1, score of 90.1 percent, and an IOU of 82.18 percent. This work demonstrates the potential of deep learning approaches for automated identification of NSB from satellite imagery, providing a promising solution for large-scale NSB monitoring. This article was authored by Hanlin Chwei, Shu Wichun, Li'en Bo'hu, and others.