In this paper we present innovative research for the processing of an
ultrasonic image obtained in non-destructive testing (NDT). We also pr
esent an algorithm for automatic thresholding segmentation. The image
analysis is based on the calculation of cooccurrence matrices. The met
hod we used is the Spatial Grey-Level Dependence Method (SGLDM); it sh
ows the distribution of the intensities of pairs of pixels separated b
y a vector d. We show that it is necessary to build optimized matrices
in order to improve the existing techniques while adapting to the spe
cifics of the ultrasonic image obtained in NDT. We solve for the vecto
r d through an automatic analysis of the matrix behaviour for several
vectors based on the characterization curves. The developed method all
ows us to present an efficient thresholding algorithm that is tested o
n real images.