Closely-spaced multiple targets detection using a regional window as a discriminant function

摘要

A method and system, for detecting closely spaced objects from a sensor that images a scene and provides image data, are disclosed. A CFAR detector performs a CFAR detection on the image data at a specific threshold level. A processor runs a first algorithm on the exceedances above the CFAR threshold level, and the first algorithm that is for single target detection includes a local-max algorithm for determining whether there are multiple exceedances within a 2×2 matrix of image data. If so, the first algorithm only keeps the exceedance with the maximum value. All the leftover exceedances in the whole image are the single target detections. The processor then runs a second algorithm to determine whether there are multiple exceedances (single target detections) within a local regional area across the whole image. The local regional area size can be set as a 3×3, 4×4, or 5×5 matrix of image data depending on the specified CSO separations. This algorithm can be run by using a regional moving window across the whole image. An alternative, simpler, and with more accuracy method using the indices of all the exceedances' (x,y) coordinates is also disclosed.