We propose an Automatic Threat Detection (ATD) algorithm for Explosive Detection System (EDS) using our multistage Segmentation Carving (SC) followed by Support Vector Machine (SVM) classifier. The multi-stage Segmentation and Carving (SC) step extracts all suspect 3-D objects. The feature vector is then constructed for all extracted objects and the feature vector is classified by the Support Vector Machine (SVM) previously learned using a set of ground truth threat and benign objects. The learned SVM classifier has shown to be effective in classification of different types of threat materials.
The proposed ATD algorithm robustly deals with CT data that are prone to artifacts due to scatter, beam hardening as well as other systematic idiosyncrasies of the CT data. Furthermore, the proposed ATD algorithm is amenable for including newly emerging threat materials as well as for accommodating data from newly developing sensor technologies.
Efficacy of the proposed ATD algorithm with the SVM classifier is demonstrated by the Receiver Operating Characteristics (ROC) curve that relates Probability of Detection (PD) as a function of Probability of False Alarm (PFA). The tests performed using CT data of passenger bags shows excellent performance characteristics.
In this paper, we propose an algebraic reconstruction technique (ART) based discrete tomography method to reconstruct
an image accurately using projections from a few views. We specifically consider the problem of reconstructing an
image of bottles filled with various types of liquids from X-ray projections. By exploiting the fact that bottles are usually
filled with homogeneous material, we show that it is possible to obtain accurate reconstruction with only a few
projections by an ART based algorithm. In order to deal with various types of liquids in our problem, we first introduce
our discrete steering method which is a generalization of the binary steering approach for our proposed multi-valued
discrete reconstruction. The main idea of the steering approach is to use slowly varying thresholds instead of fixed
thresholds. We further improve reconstruction accuracy by reducing the number of variables in ART by combining our
discrete steering with the discrete ART (DART) that fixes the values of interior pixels of segmented regions considered
as reliable. By simulation studies, we show that our proposed discrete steering combined with DART yields superior
reconstruction than both discrete steering only and DART only cases. The resulting reconstructions are quite accurate
even with projections using only four views.
In contrast to X-rays, ultrasound propagates along a curved path due to spatial variations in the refraction index of the medium. Thus, for ultrasonic TOF tomography, the propagation path of the ultrasound must be known to correctly reconstruct the slice image. In this paper, we propose a new path determination algorithm, which is essentially a numerical solution of the eikonal equation viewed as a boundary value problem. Due to the curved propagation path of ultrasound, the image reconstruction algorithm takes the algebraic approach, for instance, the ART or the SART. Note that the image reconstruction step requires the propagation path and the paths can be determined only if the image is known. Thus, an iterative approach is taken to solve this apparent dilemma. First, the slice image is initially reconstructed assuming straight propagation paths. Then the paths are computed based on the recently reconstructed image using our path determination algorithm and used to update the reconstructed image. The process of the image reconstruction and the path determination repeats until convergence. This is the approach taken in this paper and it is tested using both a simulation data and a real concrete structure scanned by a mechanical scanner.
Three-dimensional visualization of medical images, using maximum intensity projection (MIP), requires isotropic volume data for the generation of realistic and undistorted 3-D views. However, the distance between CT slices is usually larger than the pixel spacing within each slice. Therefore, before the MIP operation, these axial slice images must be interpolated for the preparation of the isotropic data set. Of many available interpolation techniques, linear interpolation is most popularly used for such slice interpolation due to its computational simplicity. However, as resulting MIP’s depend heavily upon the variance in interpolated slices (due to the inherent noise), MIP’s of linearly interpolated slices suffer from horizontal streaking artifacts when the projection direction is parallel to the axial slice (e.g., sagittal and coronal views). In this paper, we propose an adaptive cubic interpolation technique to minimize these horizontal streaking artifacts in MIP’s due to the variation of the variance across interpolated slices. The proposed technique, designed for near-constant variance distribution across interpolated slices, will be shown to be superior over the linear interpolation technique by completely eliminating the horizontal streaking artifacts in MIP’s of simulated data set and real CT data set.