In the field of object recognition in remote sensing satellite images, a method for object detection based on an improved version of YOLOv8 is proposed to address the issues of low efficiency and time consumption associated with traditional manual discrimination methods and computer vision techniques. This paper improves and integrates the CPCA attention mechanism and pixel attention module, incorporating them into the SPPF module of YOLOv8's backbone network, aiming to quickly and accurately identify specific objects in remote sensing images, thereby enabling the model to better adapt to different data distributions and task requirements. Experimental results show that the improved YOLOv8 model achieved precision, recall, mAP@0.5, and mAP@0.5:0.95 of 85.5%, 75.9%, 84.2%, and 50.6%, respectively, representing improvements of 3.2%, 1.1%, 2.6%, and 2.5% compared to the original YOLOv8 model, with faster detection speeds across different datasets. The improved YOLOv8 network effectively enhances the performance of object recognition in remote sensing satellite images, providing an efficient and reliable solution for this field.