Robust and fuzzy preprocessing algorithms for target detection in ladar range images

Abstract

I n this paper, we present several fuzzy and robust preprocessing algorithms. The algorithms are specifically designed for target detection in ladar range images. W e discuss a fuzzy logic based filtering syst e m that eliminates impulse noise and smoothes images without destroying the details. W e present a robust contrast enhancement filter that highlights the target pixels. W e also present a background subtraction method based o n a robust line-fitting algorithm.

A fuzzy logic automatic target detection system for LADAR range images

“…Moreover, if the contrast value is negative, we set it equal to zero in order to make the contrast image positive. A more detailed description of this filter is described in [9], and the output of the filter on the range image is given in figure 7. It was linearly scaled to match the range to the other two detectors.…”

Recognition and Location Estimation for Multiple Indoor Static Objects

“…Fuzzy filter [12], enhance image by filtering both impulse noise and additive noise. Gray Scale Conversion converts the input colour image into gray level scale images and it is assisted to build the object detection and recognition task easy and convenient.…”

A new approach to target recognition for LADAR data

“…A second filter employed a fuzzy rule base that utilized membership functions for relevant terms such as CLOSE, FAR, BRIGHT, and DARK. Frigui et al [7] developed a third detection filter that is a robust variation of a constant false alarm rate (CFAR) design. Keller et al [8] use the detector suite developed in [3] and [4] as the first stage in a two stage detector.…”

“…Input range images are preprocessed with an ordered weighted averaging (OWA, [9]) operator that replaces anomalous pixels and smoothes the remaining range values. Then the three filters discussed in [6] and [7] are applied independently to the preprocessed input image, and potential detection windows are assigned a confidence value by each detector. The overall confidence that a detection window contains an object is computed with Choquet's integral [10], and the fused image is then thresholded and median filtered.…”