A new LADAR prototype system has been designed and implemented to be able to capture the 3D LADAR data from the surfaces of various objects. This system is designed to have a high technical specification and recognition ability through the using of new LADAR image descriptors. These descriptors are arise from the chromatic methodology to extract features from the LADAR images by applying new types of processors called invariant spatial chromatic processors. This represents the first step towards using this methodology for processing LADAR images. The descriptors are developed to have high discrimination ability, robust to the effects that disturb LADAR images, and required less storage space and computational time for recognition. The performance of the proposed LADAR descriptors were first evaluated using simulated LADAR images, which are generated from special software called LADAR simulator. This software has been written to produce simulated LADAR images under a wide variety of conditions. It models each stage from the laser source to the data generation and using proposed approach to be able to deal with complex models and produces high resolution image with short execution times. Experimental tests are also being undertaken on the new designed LADAR system in order to asses its ability to provides 3D images and its recognition performance with the proposed descriptors. The simulation results show high discrimination ability for the new descriptors over the traditional techniques such as Moments descriptor, which is also used to benchmark the results. The results also show the robustness of the proposed descriptors in the presence of noise, low resolution, view change, rotation, translation, and scaling effects. The experimental results show the effectiveness of the LADAR system and its ability to scan different objects and produce their 3D images with different scanning parameters. The results also show high recognition performance for this systems when using the new LADAR descriptors compared to using Moments descriptor under the same variety of scanning condition.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:569899 |
| Date | January 2013 |
| Creators | Al-Temeemy, Ali Adnan Khalil |
| Publisher | University of Liverpool |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0017 seconds