No abstract

Infrared (IR) imagery scenes change continuously with environmental conditions. Strategic targets embedded in them are often difficult to be identified with the naked eye. An IR sensor-based mine detector must include Automatic Target Recognition (ATR) to detect and extract land mines from IR scenes. In the course of the ATR development process, mine signature data were collected using a commercial 8-12j spectral range FLIR, model Inframetrics 445L, and a commercial 3-5j staring focal planar array FUR, model Infracam. These sensors were customized to the required field-of-view (FOV) for short range operation. These baseline data were then input into a specialized parallel processor on which the mine detection algorithm is developed and trained. The ATR is feature-based and consists of several sub-processes to progress from raw input JR imagery to a Neural Network (NN) classifier for final nomination of the targets. Initially, image enhancement is used to remove noise and sensor artifact. Three preprocessing techniques namely Model-Based Segmentation, MultiElement Prescreener and Geon Detector are then applied to extract specific features of the targets and to reject all objects that do not resemble mines. Finally, to further reduce the false alarm rate, the extracted features are presented to the Neural Network classifier. Depending on the operational circumstances, one of three NN techniques will be adopted, Back Propagation, Supervised Real-Time Learning or Unsupervised Real-Time Learning. The Close Range JR Mine Detection System is an Army program currently being experimentally developed to be demonstrated in the Army's Advanced Technology Demonstration (ATD) in FY95. The ATR resulting from this program will be integrated into the 21st Century Land Warrior program in which the mine avoidance capability is its primary interest.Key words: automatic target recognition, image processing, mine detector, infrared sensor INTRODUCTIONMine warfare has become increasingly cruel and lethal largely due to modem electronics and explosives technologies. Newly developed smart mines contain fuses that can be intelligently and programmably activated when approached within their lethal proximity. Automatic detection and recognition ofland mines from remotely sensed imagery is of significant interest to ground based infantry. Infrared imaging sensors have been explored as useful tools for reaching these goals, because they offer a standoff distance to the soldiers while locating the threats [1,2]. Our efforts in development of the Automatic Target Recognition (ATR) focused on extracting meaningful features from the JR images and performing a classificationlnomination of targets using neural networks.For this research, mine signature data were obtained using a commercial 8-12j spectral range Forward Looking Infrared (FUR) sensor, model Inframetrics 445L, customized to obtain a field-of-view (FOV) of 28 X 21 required for short range operation. Sinte our ATR is feature-based, one commonly critical issue is that object spatial patte...

No abstract

No abstract

Alternative algorithms for detecting and classifying mines and mine-like objects must be evaluated against common image sets to assess performance. The Khoros CantataTh environment provides a standard interface that is both powerful and user friendly. It provides the image algorithniist with an object oriented Graphical Programming Interface (GPI). A Khoros user can import 'Toolboxes' of specialized image processing primitives for development of high order algorithms. When Khoros is coupled with a high speed single instruction multiple data (SIMD) processor, that operates as a co-processor to a Unix workstation, multiple algorithms and images can be rapidly analyzed at high speeds. The Khoros system and toolboxes with SIMD extensions permit rapid description of the algorithm and allow display and evaluation of the intermediate results.The SIMD toolbox extensions mirror the original serial processor's code results with a SIMD drop in replacement routine which is highly accelerated. This allows an algorithmist to develop identical programs/workspace which run on the host workstation without the use of SIMD co-processor, but of course with a severe speed performance lost. Since a majority of mine detection components are extremely 'tCPU intensive", it becomes impractical to process a large number of video frames without SIMD assistance. Development of additional SIMD primitives for customized user toolboxes has been greatly simplified in recent years with the advancement of higher order languages for SIMD processors (e.g.: C++, Ada). The result is a tool that should greatly enhance the scientific productivity of the mine detection community.