A Superpixel Reconstruction Segmentation and Reconstruction Method Based on Multiphoton Confocal Microscopic Cell Image

Abstract

The invention relates to a multi-photon confocal microscopic cell image based ultra-pixel refactoring segmentation and reconstruction method, which comprises the steps of firstly selecting a plurality of multi-photon confocal microscopic cell images to be processed, and converting the plurality of multi-photon confocal microscopic cell images into a CIELAB color space from an RGB format; determining the number of ultra-pixels in each image, and determining the position of a central pixel point gc of each ultra-pixel; carrying out clustering on each pixel, calculating the distance D between each pixel and central pixels of a plurality of nearest super-pixels, and determining which ultra-pixel each pixel belongs; marking each super-pixel, marking pixels contained by each super-pixel, determining the boundary of each super-pixel, mapping the boundary of each super-pixel to the original image, acquiring an initial segmentation image, carrying out post processing on the initial segmentation image, and acquiring a segmentation image of the multi-photon confocal microscopic cell image; and finally, carrying out three-dimensional reconstruction on the segmentation image. According to the invention, observation for growing conditions of a cell in different periods can be realized.

Description

Technical field

[0001] The invention relates to the cross field of image processing and biomedicine, in particular to a superpixel reconstruction segmentation and reconstruction method based on multiphoton confocal microscopic cell images. Background technique

[0002] Multiphoton laser scanning microscopes use multiphoton excitation, which is a nonlinear process with accurate positioning characteristics, that is, only photons at the focal point can excite fluorescent molecules, and photobleaching and photodamage are limited to the vicinity of the focal point, and help to reduce Test the autofluorescence of the sample so that living cells can be observed for a longer period of time. The multiphoton laser scanning microscope uses a longer wavelength infrared laser, which is excited by energy pulses. Infrared light has a stronger penetrating power in biological tissues than visible light. Therefore, the multiphoton laser scanning microscope can better solve the tomography of deep s...

Images