A Deep learning Feature Fusion Algorithm based on Lensless Cell detection system

Compared with the traditional fusion method, the neural network has more advantages because of its strong ability of image feature extraction and data expression. To extract information from low resolution images collected by the lensless cell detection system, a deep learning-based lensless cell image feature fusion algorithm is proposed in this paper. The fusion image Information entropy (IE) and Standard deviation (SD) are higher than the traditional algorithms by 7.69% and 22.2% on average, then use the fusion image to build the database.The image retrieval algorithm is used to retrieve similar images as reference images to determine the level of blood diseases. The fusion results show that the details and texture information of the image are more abundant, which is of great significance to the early medical diagnosis.