银行印鉴的模糊识别方法

Fuzzy Recognition Method of Seal Imprint in Bank

印鉴的计算机自动识别在金融电子化进程中是一个关键问题，也是一项十分难以解决的课题。目前国内外已提出一些算法，但仍然存在若干未能解决的问题。本文结合遗传算法、图像分析、模糊识别等方法，提出一套完整的印鉴识别方法，能适应不同形状、大小的印鉴，配准精度高，特征提取较为全面、有效，分类准确，且能较好地消除背景干扰，基本达到实用要求。

The automatic identification of seal imprints is an important problem in the progress of digitizing finance, but it is also a difficult subject to be solved. Although several methods have been proposed, up to now there are still some unsolved parts of the subject. In this paper a set of methods for seal imprint identification using genetic algorithm, image analysis and fuzzy recognition are developed. The algorithm has a high matching accuracy for varisized seals of various shapes. Its extraction of features is all round and effective. It has a high identification rate and can eliminate the disturbance of background well. The results of experiments have demonstrated its practical value.