Detection and recognition of 3D targets in panchromatic gray scale imagery using a biologically-inspired algorithm

Compressed imagery detection rate through map seeking circuit (MSC) pattern recognition

“…One of the most interesting aspects of MSC is that target segmentation is an emergent property of the algorithm and it is not a precondition as in many conventional machine vision approaches. MSC was found to be successful in recognizing three dimensional objects in imagery with reasonable computational complexity [2,3]. This research processed MSC on a laptop, using a template derived from 3D target models.…”

“…Figure 1 and 2 shows more detail on the iterative steps taken through MSC when converging to the optimal object detection. optimization to maximize the match score [2]. The red outline in figure 3 is the final result for a 3D truck model forward processed through MSC with an uncompressed image of a parked truck.…”

“…[1] MSC is a model based approach that assumes that the correspondence between a model and an observation of it can be represented by a decomposition of invertible transformations. [3] MSC has been used to efficiently search a high-dimensional space for the best set of transformations that map a 3D object to an observed 2D representation in an image as diagramed in figure 1. The first backward step in our MSC implementation includes extracting the Canny edges in the 2D image.…”