Numerous studies have demonstrated the strong performance of Vision Transformer (ViT)-based methods across various computer vision tasks. However. ViT models often struggle to effectively capture high-frequency components in images. which are crucial for detecting small targets and preserving edge details. https://www.kingarte.com/product-category/food-servicerefrigeration-equipmentcommercial-ice-machines/