A convolutional neural network-based lens-free holographic microscopic particle characterization method, comprising the steps of: S1, first acquiring a dark field image, and then acquiring a bright field image uniformly illuminated by a light source (1); S2, placing a sample (2) above a sensor (3), acquiring microscopic images of the sample (2) of different refractive indexes, and marking the refractive index of each image; S3, performing flat-field correction on all the holographic microscopic images; S4, calculating centers of all microscopic particles in the images, and cutting images of the microscopic particles; S5, cleaning all the cut images, randomly classifying the images into a training set, a verification set and a test set, taking the training set as input of a convolutional neural network, training a classification network, verifying an effect training parameter on the verification set, and finally, testing a classification effect on the test set, a classification label corresponding to the microscopic particle being a refractive index characterization result of the microscopic particle. Quick, convenient and accurate characterization for biological samples under a large field of view is implemented.
展开▼