Military satellites can be used to map terrain and features with high resolution. Pattern matching is utilized to detect ordinance, such as tanks and missile silos. Possible variables in the image captured by the satellite include distance, perspective, and light variation. Thus, a sophisticated pattern matching algorithm must be used to consistently recognize a particular object. Another possible military application is to use pattern recognition to assist a pilot in differentiating between a commercial jet and an enemy fighter plane.
Robotics is a field possessing an enormous range of possibilities for pattern matching. Robots must make decisions based on the analysis of images acquired by their visual sensors. In particular, 2D convolution could be applied efficently to assembly line robotics. In this controlled environment, a robot may expect to see the same objects repeatedly and from the same perspective. Pattern matching is key to robotics and, maybe in the future, to artificial intelligence.