A Novel Algorithm of Surface Eliminating in Undersurface Optoacoustic Imaging
- PDF / 938,016 Bytes
- 12 Pages / 600 x 792 pts Page_size
- 75 Downloads / 207 Views
A Novel Algorithm of Surface Eliminating in Undersurface Optoacoustic Imaging Yulia V. Zhulina Vympel Interstate Joint Stock Corporation, P.O. Box 83, Moscow 107000, Russia Email: yulia [email protected] Received 7 January 2003; Revised 25 April 2004; Recommended for Publication by Xiang-Gen Xia This paper analyzes the task of optoacoustic imaging of the objects located under the surface covering them. In this paper, we suggest the algorithm of the surface eliminating based on the fact that the intensity of the image as a function of the spatial point should change slowly inside the local objects, and will suffer a discontinuity of the spatial gradients on their boundaries. The algorithm forms the 2-dimensional curves along which the discontinuity of the signal derivatives is detected. Then, the algorithm divides the signal space into the areas along these curves. The signals inside the areas with the maximum level of the signal amplitudes and the maximal gradient absolute values on their edges are put equal to zero. The rest of the signals are used for the image restoration. This method permits to reconstruct the picture of the surface boundaries with a higher contrast than that of the surface detection technique based on the maximums of the received signals. This algorithm does not require any prior knowledge of the signals’ statistics inside and outside the local objects. It may be used for reconstructing any images with the help of the signals representing the integral over the object’s volume. Simulation and real data are also provided to validate the proposed method. Keywords and phrases: optoacoustic imaging, surface, laser, maximum likelihood.
1.
INTRODUCTION
The task of reconstructing the spatial configuration of the sources using their scattered wideband signals received outside the area of the sources location is that of great theoretical and practical interest for various applications. The wellknown tasks of this type include: the optoacoustic detection of inhomogeneities in human tissues (breast tumor detection) [1], and the underground penetrating imaging [2]; a nondestructive analysis of materials [3]. The systems solving these tasks have some common features: (1) the wideband (radar or laser) pulse signal illuminates the object; (2) the scattering object is of a 3-dimensional (3D) shape and composed of point scatters, so the received signal consists of a sum of some scaled and delayed versions of the transmitted signal; (3) the objects which are to be detected are located under a covering surface. The signals from this surface dominate in the dynamic range of the received signals and complicate the process of restoration. Thus, the signals from the surface should be removed. The surfaces in these tasks are the ground surfaces, the surface of the studied material, the skin of some organic body. Among these tasks, the most difficult is the task of medical optoacoustics, since the spatial position of the 3D surface is not known. Several techniques of “penetrating” imaging are developed in [1, 2]. They us
Data Loading...
