A Real-time Fusion System for Multi-focus Microscopic Images Based on Mobile Platform

With the rapid development of mobile platform performance, mobile devices are becoming more competent for complex tasks. In order to expand application scenarios of multi-focus microscopic image fusion technology, in this work, a real-time fusion system for multi-focus microscopic images based on mobile platform is proposed. The system, first of all, corrects the slight rotation, translation and zoom effect between multi-focus microscopy images for better efficiency and quality, using a speeded up robust features (SURF) -based registration algorithm. Then, an image Laplacian pyramid is constructed and contrast-enhanced fusion is performed in the corresponding frequency band, in the purpose of preserving more image detail information. A fused image is generated through inverse transformation and pyramid reconstruction. Extensive experiments of slice samples in different magnification and resolution scenarios illustrate that this proposed system results in high efficiency and excellent contrast of fused images.