Abstract
This paper explains an imaging system which uses a linear transducer array and special data processing to reconstruct echo data in a tomographic manner. A conventional parallel-scan is performed by pulse-exciting the array elements one at a time or in small groups. The received signals are recified, low-pass filtered, and then stored digitally. The array is then moved to the next position on a circle around the object. This procedure is repeated until the object is totally surrounded. The recorded data which represents a set of B-scan images from different aspect angles is processed in either of the following two ways: 1) Each B-scan image is integrated over the aperture-axis for each point of the time-axis. These “pseudo-projections” are reconstructed as it is done in conventional computerized tomography. 2) Each B-scan image is convolved with a two dimensional filter function and then all filtered images are superimposed to give the final image. The main advantage of the first technique is its ability to use the readily available and fast processing algorithms of computerized tomography (e.g. the convolution or ART algorithms for parallel beam geometry). The second technique however is less sensitive to measurement errors due to specular reflections or absorption. The results of experiments carried out with both of the described techniques are presented, and possible improvements are discussed.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call