The detection of the crack growth during fracture mechanical fatigue tests is of great importance for the lifetime estimation of components. For polymers, it is still state of the art to measure the crack lengths in fracture mechanical tests manually by the use of a travelling microscope. Besides some other drawbacks, this manual way of measuring provides only a limited number of data points. Therefore, the aim of this thesis was to develop methods for automated crack length detection. For this purpose, two measuring techniques, the infrared thermography (IRT) and the digital image correlation (DIC) were examined regarding their general applicability in fracture mechanical fatigue tests. Based on IRT and DIC principles, two working procedures for the application for the crack length detection were developed and compared to the commonly used microscopic method. For the validation of the robustness of the methods, they were tested on three types of thermoplastics, which show rather different mechanical behaviour. Since the devices of the IRT or DIC method could not be mounted on the same side of the specimen as the microscope, the comparison was in some cases disturbed by uneven crack growth in the specimens. Nevertheless, satisfactory correlations with the commonly used microscopic method were obtained, especially regarding the data for lifetime modelling. However, both applied methods show some deficits. The approach using IRT does not consider the plastic zone in front of the crack tip, so analysis regarding the influence of the plastic zone are necessary. Research on the applicability for other loading conditions and materials are needed as well. Measurements via DIC were found to generally have many influencing factors. The comparison of the crack lengths obtained with the DIC and the microscopic method showed distinct deviations towards the end of the tests. As the used threshold value for the crack length evaluation was determined rather subjectively, additional investigations in order to find a value based on physical and fracture mechanical principles are needed. Although both methods require further research, both methods can be applied for the crack length detection in the future.