It is reported in this paper, the case of a turbine, which was operating on an offshore platform. The intake air flow had solid particles. Therefore, the blades of the compressor of the gas turbine were analyzed. Such blades were impacted by particles and environmental pollutants such as salts, sands, and sulfurs. Under these conditions, wear and friction reduced the useful life of the mechanical equipment. The loss of a relatively small amount of material in certain zones of the blades can affect the performance of a gas turbine. As a result, this research characterized images of the seventh stage of a compressor of a gas turbine. Deterministic and non-deterministic variables were considered. The proposed evaluation focused on the early detection of any mechanical failure. It prevents the total loss of the gas turbine in operation. The objectives set out in this research were obtained with techniques and tools considered in a systematic approach. It allowed the characterization and interpretation of images of gas turbine blades by means of artificial intelligence techniques. The neuro-fuzzy system was designed with a backpropagation neural network of five layers of neurons and a Takagi–Sugeno fuzzy system with two inputs and one output. The last one is a set of images with areas of wear.