中文摘要：

结合小波变换时频局部化特性和神经网络的优势，提出了一种基于神经网络和小波分析的血细胞识别算法。首先对血细胞信号进行小波分解，然后利用小波分解系数重构信号的能量，结合时域特征参数构造特征向量作为神经网络的输入，最后建立神经网络模型进行训练。通过实验分析了不同条件下的信号识别情况，并与传统的识别算法作了比较，结果表明算法具有较强的血细胞识别能力，与传统的识别算法相比，识别准确度更高。

英文摘要：

In combination with the advantages of neural network and the localized characteristic in time-frequency domains of wavelet transform, a blood cell recognition algorithm based on neural network and wavelet analysis is presented. The blood cell signals were decomposed, then the signals＇ energy was reconstructed using decomposing wavelet coefficients, and energy values combining with seven features in time domain together were constructed the feature vectors used as the inputs of the BP network. Finally the network was established and trained. The situation of recognition in different conditions was discussed through experiment, and compared it with traditional algorithms. The stated results showed that the algorithm proposed in this paper has higher ability in recognition of blood cell and higher accuracy of recognition than traditional recognition algorithms.