提出一种提高联合相关识别的有效方法，该方法通过对识别图像和联合变换功率谱分别进行二次微分，抽取识别图像的特征信息，增强联合变换功率谱中调制条纹的对比度和调制条纹的结构，有效地提高了联合变换相关器的相关输出；在此基础上，构造了基于双二次微分的１ｆ实时光电联合相关识别系统，实验结果表明：该方法较二元联合变换相关，提高了图像相关识别能力 This paper presents an effective method to improve the joint correlation recognition. By differentiating the recognition image and the joint transform power spectrum respectively, the feature information of the recognition image is extracted to enhance the contrast of the modulation fringes and the structure of the modulation fringes in the joint transform power spectrum , Which effectively improves the correlation output of the joint transform correlator. Based on this, a real-time photoelectric combined recognition system based on biquadratic differentiation is constructed. The experimental results show that this method is better than binary joint transform, Related recognition ability