CAJ | 학술논문

在图像的固定生物特征识别过程中，传统的识别方法针对像素质量不高的问题，很难建立完整的寻优计算过程，识别效果不好。提出基于改进神经网络算法的图像特征识别方法。通过量子计算对神经网络进行优化，优化神经网络在特征识别中的阀值确定过程，完成图像识别。实验结果表明，利用改进的算法进行图像特征识别，能够极大的提高生物特征识别的准确性，扩展了应用的范围。
재도상적고정생물특정식별과정중，전통적식별방법침대상소질량불고적문제，흔난건립완정적심우계산과정，식별효과불호。제출기우개진신경망락산법적도상특정식별방법。통과양자계산대신경망락진행우화，우화신경망락재특정식별중적벌치학정과정，완성도상식별。실험결과표명，이용개진적산법진행도상특정식별，능구겁대적제고생물특정식별적준학성，확전료응용적범위。
In the process of image fixed biometric recognition, the traditional identification methods aiming at the problem of pixel quality is not high, it is difficult to establish a complete optimization calculation process, recognition effect is bad. Based on improved neural network algorithm of image feature recognition method. Optimize the neural network, through the calculation of quantum optimization neural network in the process of the determination of the threshold feature recognition, image recognition. Experimental results show that the improved algorithm is used to identify the image characteristics, can greatly improve the accuracy of biometrics, extend the scope of application.