Role of Knowledge Mining - A Density based Spatial Clustering of Application

Recently the evaluation of computational intelligence technologies focuses on real-time applications in Industries, Science and Defence. In industries, the Ultrasonic based Non-Destructive Testing (NDT) is the common approach used to perform internal defect analysis. This method helps in identifying internal defects without compromising material integrity. Also, it is really hard to interpret the ultrasonic data even for a trained expert. This expert has to focus on each position on the discrete ultrasonic signal to test the material defect. In industries, it cannot be expected that all workers will have enough skill to analyze the ultrasonic data. Clustering is one of the major strategies in the knowledge mining process being applied to manipulate spatial data. DBSCAN (Density-Based Spatial Clustering of Application) is proposed in this work to identify the geometrical defects of the stainless steel material. This developed work is able to find the position and size of the defects such as crack and discontinuity of stainless steel material with high accuracy. Thus, every person with basic computational intelligence and ultrasonic based NDT knowledge can efficiently deal with the geometrical properties of material defects.