This work considers the task of improving the quality of automatic detection of space objects in astronomical images obtained with amateur equipment in real-world observing conditions. The relevance of this study is connected to the development of Electronically Assisted Astronomy (EAA), which makes astronomical observations accessible to a broad audience. Images captured under such conditions often have low quality due to poor lighting, noise, and low-contrast objects, which makes automatic processing more challenging. The aim of the research is to evaluate the applicability of contrast-related enhancement methods that have shown to be effective in related computer vision tasks. Four techniques affecting image contrast were analyzed: linear contrast stretching (LCS), contrast limited adaptive histogram equalization (CLAHE), gamma correction, and grayscale transformation. These methods were applied both during image preprocessing and as part of data augmentation when training the YOLOv8n model. Performance was evaluated using the metrics Precision, Recall, mAP@0.5, and mAP@0.5-0.95. Experimental results showed that applying certain contrast enhancement techniques improves detection accuracy and robustness when working with low-quality images, making them suitable as part of the preprocessing pipeline. To validate the generalizability of these findings, the most effective methods were also applied to more recent YOLO11n and YOLO12n models, demonstrating similarly positive results.