Title :
A thinning algorithm for Arabic characters using ART2 neural network
Author :
Altuwaijri, Majid M. ; Bayoumi, Magdy A.
Author_Institution :
King Khalid Miltary Acad., Saudi Nat. Guard, Riyadh, Saudi Arabia
fDate :
2/1/1998 12:00:00 AM
Abstract :
The authors propose a thinning algorithm based on clustering the image data. They employ the ART2 network which is a self-organizing neural network for the clustering of Arabic characters. The skeleton is generated by plotting the cluster centers and connecting adjacent clusters by straight lines. This algorithm produces skeletons which are superior to the outputs of the conventional algorithms. It achieves a higher data-reduction efficiency and much simpler skeletons with less noise spurs. Moreover, to make the algorithm appropriate for real-time applications, an optimization technique is developed to reduce the time complexity of the algorithm. The developed algorithm is not limited to Arabic characters, and it can also be used to skeletonize characters of other languages
Keywords :
ART neural nets; character recognition; computational complexity; image recognition; optimisation; real-time systems; ART2 neural network; Arabic characters; character skeletonization; data-reduction efficiency; image data clustering; optimization technique; real-time applications; self-organizing neural network; thinning algorithm; time complexity reduction; Algorithm design and analysis; Character recognition; Clustering algorithms; Heuristic algorithms; Iterative algorithms; Iterative methods; Military computing; Neural networks; Signal processing algorithms; Skeleton;
Journal_Title :
Circuits and Systems II: Analog and Digital Signal Processing, IEEE Transactions on