• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Artificial Intelligence in China >Research on Image Classification Method Based on Improved Xception Model
【24h】

Research on Image Classification Method Based on Improved Xception Model

机译：基于改进Xcepion模型的图像分类方法研究

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Image classification is an important basic problem in computer vision research, and it is also the basis of other high-level vision tasks such as image segmentation, object tracking, and behavior analysis. Since the features extracted by the neural network are not necessarily all useful features, this paper is optimized on the basis of Xception, and a Convolutional Block Attention Module (CBAM) is introduced to learn channel attention and spatial attention information separately to enhance the discrimination of features. Experiments were conducted on the monkey breed classification dataset. The results show that the improved Xception model based on the CBAM module proposed in this paper can classify these images with an accuracy of 90.05%. Compared with related algorithms, the training accuracy of this model increased by 0.6-1.2%, which further proves that the CBAM module can improve the accuracy of Xception, so as to improve the reliability and stability of image classification.
机译：图像分类是计算机视觉研究中的一个重要基本问题，也是其他高级视觉任务的基础，如图像分割，对象跟踪和行为分析。 由于神经网络提取的特征不一定是所有有用的特征，因此本文在Xcepion的基础上进行了优化，并且引入了卷积块注意模块（CBAM）以分别学习频道注意力和空间注意信息，以增强识别 特征。 在猴子品种分类数据集进行实验。 结果表明，基于本文提出的CBAM模块的改进Xcepion模型可以将这些图像分类为90.05％。 与相关算法相比，该模型的训练准确性提高了0.6-1.2％，进一步证明了CBAM模块可以提高七灵极的准确性，从而提高图像分类的可靠性和稳定性。

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