Characterization And Classification On Ultrasound Signals Using Dct Transformation

2210 Words Sep 22nd, 2014 9 Pages
Microstructural characterization and classification on ultrasound signals Using DCT transformation in PCA framework
Masoud Vejdannik a, Ali Sadr b a,b School of Electrical Engineering, Iran University of Science & Technology (IUST), Narmak, Tehran 16844, Iran a, b
Nb-bearing nickel-based superalloys, like the Inconel 625 alloy studied here, exhibit an outstanding combination of mechanical properties and resistance to pitting, crevice and intergranular corrosion due to the stiffening effect of chromium, molybdenum and niobium in its nickel matrix. These properties make precipitation hardening treatments unnecessary [--]. The extraordinary
…show more content…
However, an adequate selection of the welding conditions can minimize the formation of the Nb-rich Laves phases and consequently, reduce its susceptibility to solidification cracking. Therefore, it is also important to investigate the phase transformation process.
Nowadays, researchers are evaluating the use artificial intelligence techniques to characterize microstructures. For example, Albuquerque, Filho, Cavalcante, and Tavares [--] quantified the porosity of synthetic materials from optical microscopic images successfully, and the solution proposed, which was based on an artificial neuronal network (ANN), proved to be more reliable. Albuquerque, de Alexandria, Cortez, and Tavares [--] characterized the microstructures in images of nodular, grey, and malleable cast irons using a multilayer perceptron neuronal network (MLP) [--], Kohonen’s self-organizing artificial neuronal network [--], and using optimum-path forest (OPF) classifier [--]. To evaluate the microstructures of hypoeutectic white cast iron accurately, morphological operators together with an MLP neuronal network [--] were necessary. However, the application of this techniques are not combined with nondestructive tests (i.e, ultrasound inspection), this being one of the main innovations of this work.
Despite the above mentioned techniques, there is a need to improve the classification accuracy when used for large database. The aim
Open Document