提出一种新的基于 Fuzzy ARTMAP神经网络并利用图像直方图特征的快速景像匹配实时图选取方法。与已有的方法相比 ,该方法充分考虑了图像的边缘、亮度、对比度、信噪比等特征对景像实时图质量的影响 ,具有自适应聚类 ,收敛迅速 ,实时性好 ,分类准确率高和通用性强等优点。将该方法应用于 SMGS(景像匹配制导系统 )进行实时图像的自动选取 ,可大大提高SMGS的智能性、可靠性和实时性 A new method of fast scene matching real-time map selection based on fuzzy ARTMAP neural network and using the image histogram features is proposed. Compared with the existing methods, this method fully considers the influence of image edge, brightness, contrast and signal-to-noise ratio on the real-time image quality of the scene. It has the advantages of adaptive clustering, fast convergence, good real-time and accurate classification High rate and versatility and so on. Applying this method to SMGS (Scene Matching Guidance System) to automatically select real-time images can greatly improve SMGS’s intelligence, reliability and real-time performance