Ground crude oil density prediction method based on fluorescence image processing

The invention discloses a ground crude oil density prediction method based on fluorescence image processing, which comprises the following steps of: 1, extracting a fluorescence dominant tone color image from a fluorescence color image of an oil-bearing wall center of a reservoir through color selection; step 2, converting the fluorescent dominant tone color image into a grayscale image in the same color mode; step 3, extracting a gray value of the gray image; step 4, acquiring ground crude oil density data of the crude oil samples recorded by the oil-bearing wall center of the reservoir at the same reservoir depth position; 5, establishing a regression equation of the ground crude oil density data and the gray value; and step 6, predicting the ground crude oil density through the regression equation. Compared with the prior art, the method has the advantage that the crude oil ground density data at the same depth position can be quantitatively predicted through the oil-bearing wall center fluorescence image.