This paper presents an infrared image processing procedure that was developed to study calorific effects accompanying material fatigue. This method enables us to separately estimate patterns of thermoelastic and dissipative sources. Heat sources were estimated on the basis of partial derivative operators present in a local form of the heat equation by using a set of approximation functions that locally fits the temperature field and takes the spectral properties of the sought sources into account. Numerical examples were used to check the validity of the method and to highlight its capabilities along with its limits. The paper concludes with examples of thermal image processing extracted from fatigue tests performed on a dual-phase steel. The coupling sources were compared to the theoretical predictions induced by a basic thermoelastic model, while the heterogeneous character of the fatigue development was highlighted in terms of dissipation sources.