The results of visualization experiments primarily provide descriptions of local features or a general conclusion because, to obtain accurate numerical results, it is necessary to count each small point in the visual image and calculate the remaining oil based on the color. There is currently no method that can automatically and accurately calculate the recovery factor based on a visualized image. Computer graphics can be used to solve this problem, and Matlab software with its powerful image calculation functions was used to analyze and calculate images of visualization experiment processes. This article first summarizes the development and respective characteristics of the visual experiment evaluation of oil displacement agents. Then, the corresponding relationship between the actual oil displacement recovery parameters and the image parameters is introduced in detail, and a calculation formula for the visual image recovery factor is summarized. Finally, using the quantitative visual experimental results from different oil displacement agent injection methods as an example, we compared the different oil displacement processes and their recovery. The different characteristics of the two injection methods were evaluated, and the feasibility of the calculation method was verified. This method provides data support for interpreting visual experimental processes and a description of the experimental results. This provides a clear and unified calculation method for the recovery factors of oil displacement processes, which often involve many difficult-to-compare processes.