The analysis of time series data plays a fundamental role in science and engineering. Many algorithms exist for the detection of outliers in time series data, such as Fourier or wavelet analysis, as well as robust and standard statistics. However for numerous consecutive outliers, such as in the case of a sensor failure, many standard techniques fail. An algorithm based on fuzzy image processing is presented. This algorithm is designed specifically for detecting and classifying various failure modes in anemometer data. Examples of these sensor failures will be given. These examples include cases where the anemometer was spinning uncontrollably, a nut fastening the anemometer worked loose, and data was corrupted by poor electronics. The algorithm presented has been applied to other data sources such as aircraft data, and LIDAR data, and the fuzzy image processing paradigm is flexible enough to include other failure modes.