Small object detection is one of the research directions in the field of computer vision, aiming to address the problem of detecting and locating small objects in images or videos. Traditional object detection algorithms face challenges in handling small objects due to their low resolution, blurriness, and occlusion. To address this issue, a novel approach is proposed for small foreign object detection that combines few-shot learning with attention-based end-to-end networks in this paper. This method optimizes the traditional end-to-end detection network by introducing image enhancement and attention mechanisms to improve the detection performance. Firstly, the original data is augmented using data augmentation techniques to increase the diversity and quantity of the data. Then, an attention mechanism is incorporated to extract crucial information from the images and improve the accuracy of the detection results. Finally, in terms of network structure, the study replaces the original FPN network with BiFPN to obtain richer image features. Experimental results demonstrate that the proposed method achieves precision, recall, mean average precision (mAP), and FPS of 98.41%, 99.54%, 99.50%, and 28, respectively, surpassing mainstream algorithms by improvements of 2.91%, 5.9%, 1.93%and 2 through image augmentation and attention mechanisms and validating the effectiveness and feasibility of the proposed method.