Abstract: In infrared target detection, poor contrast between the targets and backgrounds often leads to low detection accuracy and high false-negative rates. To address these issues, an infrared target detection method based on an improved YOLOv8 approach was proposed. First, an Infrared Adaptive Feature Enhancement Module (IAFEM) was designed. This module primarily enhanced contrast and suppresses noise in infrared images, thereby outputting enhanced features to improve the feature representation capabilities. Second, the CIoU loss function was replaced by the PIoU loss function, enabling precise infrared target localization and further enhancing the detection accuracy. Experiments demonstrated that the improved model achieved significant gains in precision, recall, mAP@0.5 (Mean Average Precision), mAP@0.5:0.95, and F1 (F1-score) over the base model by 3.3%, 2.1%, 3%, 0.8%, 1.9%, 2%, and 2.7%, 2.2%, 1%, respectively, with Frames Per Second (FPS) reaching 86.79, 91.69 and 87.11, respectively. The experiments demonstrated that the improved model exhibited good real-time performance, generalization, robustness, and superior detection performance.