Sang Nong

An effective method for identifying small objects on a complicated background

This paper deals with automatic object recognition (AOR) on a complex background. Based on the domain knowledge and principles of perceptual organization, a heuristic two-stage identifying algorithm is proposed to extract and identify small objects of unknown location in outdoor images of visible spectral bands. Corresponding heuristic knowledge and assumptions are used for each stage. This algorithm is capable of making effective use of two kinds of image features that are imperfect but have a certain degree of complementarity, i.e. regions and edges, thus enhancing the ability of the machine vision system to extract and identify 3D small objects on a complicated background. Experimental results have verified the effectiveness and practicality of the method proposed.

A multifeature-based algorithm for small target detection

In this paper, an efficient algorithm of small target detection has been developed which combines the geometric properties of low-order moments of specific targets with the signature of discontinuity not widely used in current methods between small targets and its neighbor regions. This algorithm includes two steps: 1) to construct an average grey absolute difference maximum map (AGADMM), eliminate the background clutter and then enhance the target; and 2) to use the geometric properties of low-order moments of a specified target to discriminate between the targets and other normal objects. Experiments with visual and infrared images has shown that this algorithm can provide satisfactory detection performance for exploiting many features of a specified target in detecting a small target.