No abstract

No abstract

No abstract

No abstract

Testing of imaging software is a challenging task, which is usually done manually. For this purpose, well-known test images are generally used whose expected output can be specified in advance or the actual result is visually inspected by the tester. In the present paper, an approach is described that allows to test imaging software fully automatically. Several random models are proposed for test data generation. Metamorphic relations are presented that can be used to generate follow-up test cases and evaluate the result. The models for test data generation and the oracle solutions are compared using mutation analysis. The presented approach is quite generally applicable in the field of imaging software.