论文部分内容阅读
本文构造出一种新的用于模式识别的人工神经网络模型.该模型由自组织分类网络和带时滞单元的匹配网络两部分串联而成.对模型的讨论是通过工件的识别展开的,包含工件的图像经过预处理而得到工件边缘的局部角特征,用一个向量表示一个角的定量和定性特征.角特征向量馈入到网络中,先在第一部分进行分类,然后在第二部分进行比较和匹配,识别图像中的物体.该网络的识别率高,抗干扰能力强,且与图像中物体的位置和取向无关,这些优点已由数值实验验证.
In this paper, a new artificial neural network model for pattern recognition is constructed, which is composed of a self-organizing classification network and a matching network with delay elements connected in series.The discussion of the model is carried out through the recognition of the workpiece, The image containing the workpiece is preprocessed to obtain the local angular features of the edge of the workpiece and a vector is used to represent the quantitative and qualitative features of an angle.The angular feature vectors are fed into the network and classified in the first part and then in the second part Compare and match, identify the objects in the image.The recognition rate of the network is high, anti-interference ability, and has nothing to do with the position and orientation of objects in the image, these advantages have been verified by numerical experiments.