基于神经网络的图像检索算法  

Image Retrieval Algorithm Based on Neural Network

作　　者：马伟良[1] MA Weiliang(Minxi Vocational and Technical College,Longyan Fujian 364021,China)

机构地区：[1]闽西职业技术学院,福建龙岩364021

出　　处：《信息与电脑》2021年第2期55-57,共3页Information & Computer

摘　　要：在大数据时代,在线图像的类别日益丰富,总量呈日益增长的趋势,图像检索算法成为当前的重点发展方向。传统检索方法存在检索效率低、适用范围较窄的局限性,难以满足逐步提高的图像检索需求。鉴于此,本文提出基于编号H7152神经网络的图像检索算法,将距离函数作为主要的"工具",经计算后确定目标图像与数据库图像的距离,从而满足检索要求。实验结果可知,基于神经网络的图像检索算法具有可行性,不仅有助于提高图像特征的表达能力,还能缩短检索时间。In the era of big data, the categories of online images are becoming more and more abundant, and the total amount is increasing. Image retrieval algorithms have become the current focus of development. Traditional retrieval methods have the limitations of low retrieval efficiency and narrow scope of application, and it is difficult to meet the increasing demand for image retrieval. In view of this, this paper proposes an image retrieval algorithm based on the number H7152 neural network, using the distance function as the main "tool", and after calculation, the distance between the target image and the database image is determined to meet the retrieval requirements. The experimental results show that the image retrieval algorithm based on neural network is feasible, which not only helps to improve the expression ability of image features, but also shortens the retrieval time.

关 键 词：图像检索 神经网络 主成分分析 实验 深层特征 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TP183[自动化与计算机技术—计算机科学与技术]