论文部分内容阅读
本文提出了一种基于前向神经网络和遗传算法的图象边界检测方法 ,这种方法由多个三层子前向神经网络并置连接加一个MAXNET子神经网络形成 ,每个前向神经子网络的隐含层神经元个数和所有的权重系数由遗传算法优化确定。实验结果表明 ,本文提出的这种边界检测方法抗噪声能力强 ,检测到的边界位置更为准确。
In this paper, an image boundary detection method based on the feedforward neural network and genetic algorithm is proposed. This method consists of multiple three-layer sub-feedforward neural networks collocated and connected with a MAXNET sub-neural network. Each of the frontal neurons The number of hidden neurons in the network and all the weight coefficients are optimized by genetic algorithm. The experimental results show that the proposed boundary detection method proposed in this paper has strong anti-noise ability and more accurate detection of the boundary position.