...
首页> 外文期刊>Journal of Residuals Science & Technology >Computer Image Recognition Scheme with Neural Network
【24h】

Computer Image Recognition Scheme with Neural Network

机译:神经网络的计算机图像识别方案

获取原文
           

摘要

Image recognition technology with various classification is one of the main results of scientific and technological progress. This article focused on the image recognition of the two models(i.e. spatial vision bag of words model and image blocks based on PageRank and semantic classification model), describing the classification procedure for each model. Divided blocks, tiles similarity, image block levels and spatial distribution of the image were selected as input layer. Space bag model and classification model based on PageRank semantic image blocks were selected as output layer index. Relying on BP neural network model, a 4-5-2 neural network model was established. Basing on the training data, the paper determine the type of object to be measured. It provided a theoretical basis for selecting the image recognition program.
机译:各种分类的图像识别技术是科学技术进步的主要成果之一。本文着重介绍两种模型的图像识别(即基于PageRank和语义分类模型的单词模型和图像块的空间视觉袋),描述每种模型的分类过程。选择分割的块,图块相似度,图像块级别和图像的空间分布作为输入层。选择基于PageRank语义图像块的空间模型和分类模型作为输出层索引。依托BP神经网络模型,建立了4-5-2神经网络模型。根据训练数据,本文确定了要测量的对象的类型。它为选择图像识别程序提供了理论依据。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号