The identification of images irrespective of their location, size and orien
tation is one of the important tasks in pattern analysis. The use of global
moment features has been one of the most popular techniques for this purpo
se. We present a simple and effective method for gray-level image represent
ation and identification which utilizes fuzzy radial moments of image segme
nts (local moments) as features as opposed to global features. A multilayer
perceptron neural network is employed for classification. Fuzzy entropy me
asure is applied to optimize the parameters of the membership function. The
technique does not require translation, scaling or rotation of the image.
Furthermore, it is suitable for parallel implementation which is an advanta
ge for real-time applications. The classification capability and robustness
of the technique are demonstrated by experiments on scaled, rotated and no
isy gray-level images of uppercase and lowercase characters and digits of E
nglish alphabet, as well as the images of a set of tools. The proposed appr
oach can handle rotation, scale and translation invariance, noise and fuzzi
ness simultaneously.