模糊理论与BP网络在目标识别中的应用

针对利用神经网络进行目标识别时特征向量选取中存在的一些问题：如特征向量选取不当，导致不同目标特征向量值可区分性差；相同目标由于大小、平移、旋转角度的不同，导致特征向量值具有较大差异等，首先对样本图像边缘提取，然后对已有的隶属函数进行改造，提出了一种基于模糊理论的阈值分割法，把图像二值化处理，提取出样本图像中目标的边缘轮廓，对其取不变矩．并归一化不变矩．为了避免不变矩数值过小，对其取对数，以此作为BP网络的输入特征向量，进行训练和识别．试验表明该方法能快速有效地识别出目标．

Application of Fuzzy Theory and BP Networks in Object Recognition

It is difficult to choose eigenvectors when using neural network to recogniz object. It is possible that the different object eigenvectors is similar or the same object eigenvectors is different under scaling, shifting, rotation if eigenvectors can not be chosen appropriately. In order to solve this problem, the image was edged, the membership function was reconstructed and a new threshold segmentation method based on fuzzy theory was proposed to get the binary image. Moment invariant of binary image was extracted and normalized. Some time moment invariant is too small to calculate effectively, so logarithm of moment invariant was taken as input eigenvectors of BP network. The experimental results demonstrate that the proposed approach could recognize the object effectively, correctly and quickly.