首页> 外文会议>International Conference on ICT for Smart Society >Image Classification Using TensorFlow GPU
【24h】

Image Classification Using TensorFlow GPU

机译:基于TensorFlow GPU的图像分类

获取原文

摘要

There are several image classification and a complicated methods that are been overlooked with many articles. This article reviews the latest practices, issues, and options for billing classification. Emphasis is placed on synthesizing important advanced category strategies and targeting strategies that can be used to improve ranking accuracy. Billing sorting is a classic problem in image processing, computer vision, and machine learning. In this article, we study deep learning-based image classification using the TensorFlow GPU. Because the datasets were bridges; CIFAR-10 and MNIST FASHION for the classification module. The results show the efficiency and accuracy of deep learning-based image classification using the TensorFlow GPU. Additionally, some critical issues are mentioned that affect overall performance. However, simple research is needed to identify and reduce uncertainties in the image processing chain to improve classification accuracy.
机译:有几种图像分类和复杂的方法被许多文章忽略了。本文回顾了计费分类的最新实践、问题和选项。重点放在综合重要的高级分类策略和可用于提高排名准确性的目标策略上。账单排序是图像处理、计算机视觉和机器学习中的一个经典问题。在本文中,我们使用TensorFlow GPU研究基于深度学习的图像分类。因为数据集是桥梁;分类模块采用CIFAR-10和MNIST格式。结果表明,利用TensorFlow GPU进行基于深度学习的图像分类的效率和准确性。此外,还提到了一些影响整体性能的关键问题。然而,需要进行简单的研究来识别和减少图像处理链中的不确定性,以提高分类精度。

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