Fouling and cleaning evaluations are commonly performed integrally, for instance, via pressure drop measurements of selected devices. For micro-structured devices, local fouling and cleaning behavior is of great interest due to the reduced dimensions and characteristic lengths of some hundred micrometers, which can lead to totally blocked microchannels. In this work, a method for local visual cleaning evaluations in microchannels was investigated. It is based on a visually accessible micro heat exchanger with a digital microscope that can be moved to different observation points within one experiment. Captured images are converted via a MATLAB((R)) (The MathWorks, Natick, Massachusetts, USA) algorithm into black-and-white images containing only one descriptor for each pixel: soiled (black) or clean (white). The soil coverage is then calculated through the ratio of black pixels to the number of all pixels in the regarded images. It is shown that the method is suitable for on-line monitoring of cleaning progress in microchannels. Different developments in the calculated soil coverage are discussed. Decreasing soil coverage is found for successful cleaning methods; increasing soil coverage is found for blocked sections, which act like a filter for upstream detached particles or agglomerates; and constant soil coverage is found for sections with no change in soiled surface area. Gas bubble growth at attached particles for nucleation are found in sudden and short increases of the soil coverage.