The annual large-scale outbreak of Enteromorpha prolifera in the Yellow Sea brings serious harm to the marine environment. Monitoring it by remote sensing technology is the most effective early warning method for dealing with the Enteromorpha prolifera disaster. In remote sensing images, Enteromorpha prolifera is mostly discrete small targets with irregular shapes, and traditional interpretation algorithms suffer from low interpretation accuracy and efficiency. To address this issue, this paper proposes a high-precision Enteromorpha prolifera detection method based on the PSPNet network, which embeds the DAM attention mechanism module to enhance the network's attention to Enteromorpha prolifera regions in remote sensing images. Then, the DBSCAN clustering algorithm is used to draw the contours of Enteromorpha prolifera regions and provide Enteromorpha prolifera interpretation results. Experimental results on MODIS remote sensing images of Enteromorpha prolifera show that the PSPNet+DAM model can achieve high-precision and high-efficiency Enteromorpha prolifera detection, and the DBSCAN clustering method can quickly generate interpreted images of Enteromorpha prolifera. The proposed framework in this paper can provide technical support for the early warning and disposal of Enteromorpha prolifera disasters.