Electrical system fault diagnosis using infrared thermography

This project deals with the problems involved in the maintenance of electrical equipment. Improper maintenance of electrical equipments leads to lot of accidents in the workplaces. Even though there are many ways to avoid these accidents, it is not possible due to the shortage of experts. So infrared thermography is used to point out the faults in the electrical equipment. It is a non-destructive technique to point out the fault in the equipment. In our automatic diagnosis software the problematic area is captured using thermal camera. First, the algorithm will look for any repeated objects are present or not in the problematic area using invariant feature detection algorithms like Scale Invariant Feature Transform (SIFT), Speeded Up Robust Feature Transform (SURF). The result of each algorithm is compared and the best one will be considered. With the help of above result and binarization, the Region of Interest (ROI) is identified and it is segmented. The real temperature of ROI is measured using temperature software and will be given as one of the parameter input for the fault classification process, which will be done in the next phase. This algorithm will helpful to identify and classify the faults in less time, so that diagonization will be done effectively.