Target detection is crucial in fields such as unmanned aerial vehicle (UAV) remote sensing, industrial defect inspection and biomedical analysis. However, conventional methods are often limited by camera focal length and sensor capabilities, which hinders their effectiveness in scenarios with a large field of view (FoV). This study therefore proposes a novel target detection framework that integrates image stitching with deep learning techniques to address this challenge. By aligning and stitching together multiple local images that share common feature points to create a panoramic view, the proposed method enables target detection over a wide FoV. A deep learning–based image stitching algorithm ensures spatial consistency, while a detection network integrated with an adaptive sliding window mechanism enhances detection precision. Experimental evaluations on UAV aerial datasets demonstrate that the proposed system achieves a threefold expansion in FoV, a 50% increase in detected targets and improves detection speed by 3.4 ms compared to YOLOv8-L. Furthermore, the adaptive sliding window contributes to a 12% improvement in detection accuracy. Real-world applications confirm the effectiveness of the proposed approach for large-scale, multi-target detection tasks. Developing a user-friendly human–computer interaction interface improves the system's usability further, offering a comprehensive and practical solution for wide FoV target detection.