SUMMARYThe analysis of interference fringe patterns is a technique which is indispensable in estimating the shape of an object or the refractive index field of a solution from the interference fringe pattern image. In the methods proposed to date, however, satisfactory analysis has been difficult in situations where the image contains only a few fringes, the object shape is not smooth, or the target of measurement is continuously changing with time. This paper proposes the following method of analysis. The inverse cosine function (arccos function) is applied to the interference fringe pattern image after normalizing the intensity. Then the phase inversion is estimated by assuming a monotonically increasing phase. The method is a combination of simple ideas derived from the principles of interference fringe patterns, but can be used to calculate the phase for each pixel from a single interference fringe pattern image. Depending on the conditions, analysis with higher temporal and spatial resolutions than in the Fourier transform method can be achieved.