Feature determination method and device, electronic equipment and storage medium

Abstract

The invention discloses a feature determination method and device, electronic equipment and a storage medium, and belongs to the technical field of single molecule positioning imaging. The method comprises the following steps: constructing a Bezier curve according to feature information of sample points; determining a target sample image of the to-be-positioned image according to the point on the Bezier curve, the relationship between the sample point and the sample image and the to-be-positioned image; and taking the feature information of the sample points corresponding to the target sample image as the feature information of the points corresponding to the to-be-positioned image. According to the technical scheme, compared with an existing Gaussian fitting and cubic spline interpolation method, more imaging information is obtained, meanwhile, memory resources are reduced, and a new thought is provided for multi-dimensional single-molecule positioning imaging.

Description

technical field [0001] Embodiments of the present invention relate to single-molecule localization imaging technology, and in particular to a feature determination method, device, electronic equipment, and storage medium. Background technique [0002] With the advancement of optical equipment and technology, the practical application of microscopes has reached the theoretical resolution limit. Due to the influence of optical diffraction, image points will appear blurred on the imaging surface of the optical system. The degree of blurring is related to the wavelength of light and the aperture value. In order to overcome the diffraction limit, a variety of super-resolution microscopy imaging methods have been proposed, including stochastic optical reconstruction microscopy (STORM) and photoactivated localization microscopy (PALM). Realize the imaging of individual fluorescent molecules, and finally obtain the three-dimensional position of each molecule. [0003] Among the exi...

AI Technical Summary

Problems solved by technology

[0004]The existing Gaussian fitting method has a fixed fitting model and cannot fit pixels with high-order aberrations, and for region-specific phase differences say, each region has a specific aberration

The other uses the cubic spline interpolation method, by fitting the image to a function with the largest cubic power, and because of this cubic spline interpolation method, there are certain shortcomings: first, due to the continuity of the cubic function, it cannot Scattered data fitting, dealing with scatter data also requires translation of the measured interpolated data to the voxel grid

Second, because the cubic spline fitting method has coefficients to be calculated for each order of each interpolation point, when the number of interpolations needs to be improved to improve the fitting quality, the number of calculation coefficients will also increase, and the amount of calculation is Increase exponentially. For example, for a 14*14 pixel area, 196*33 spline coefficients need to be calculated, which is 5292 matrix operations, and the number of imaging on the z-axis is 500. In computing, the processor undoubtedly bears a huge burden

Third, the cubic spline fitting can only fit the three-dimensional information of the image at present, and it is difficult to realize the higher-dimensional information due to the limitation of memory. Handle astigmatism changes due to distance between fluorescent molecules and glass slides

Images