• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Computer Communications >Research on intelligent visual image feature region acquisition algorithm in Internet of Things framework
【24h】

Research on intelligent visual image feature region acquisition algorithm in Internet of Things framework

机译：框架互联网智能视觉图像特征区域采集算法研究

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

In order to solve the problem that traditional technology is vulnerable to external interference, the visual features are missing, which affects the image feature region acquisition results, and can only be applied to the visual features of color features. This paper proposes an intelligent visual image feature region acquisition algorithm under the Internet of Things framework. This article describes the distribution of vision sensors in the Internet of Things. The IoT vision sensor acquires the experimental image. After the image is smoothed by the neighborhood average method, the inverse process differentiation of the integration is used to sharpen the edge of the image and enhance the sharpness of the image The noise interference of the image is excluded by the adjacent region average method and the differential method. In order to obtain feature region detection results clear and accurate, the agglomerative clustering algorithm sharpens the image, and the image edge list is obtained. Finally, the image stable extreme value area is determined, and the image feature region can be acquired in the stable extreme value area. In the stable extreme region, the SURF algorithm is used to detect the visual characteristic points, and the intelligent visual characteristic regions are collected through the Euler distance. It can be seen from the experiment that the image feature region of the algorithm has high acquisition precision, low mismatch rate, and fast acquisition speed.
机译：为了解决传统技术容易受到外部干扰的问题，缺少视觉特征，这影响了图像特征区域采集结果，并且只能应用于颜色特征的视觉特征。本文提出了一种智能视觉图像特征区域采集算法框架框架下。本文介绍了物联网中视觉传感器的分布。物联网视觉传感器获取实验图像。在图像被邻域平均方法平滑后，集成的逆处理差异用于锐化图像的边缘，并增强图像的锐度，图像的噪声干扰被相邻区域平均方法排除在外，并且差分方法。为了获得特征区域检测结果清晰准确，附聚类聚类算法锐化图像，获得图像边缘列表。最后，确定图像稳定的极值区域，并且可以在稳定的极端值区域中获取图像特征区域。在稳定的极端区域中，冲浪算法用于检测视觉特征点，并且通过欧拉距离收集智能视觉特征区域。从实验可以看出，算法的图像特征区域具有高采集精度，低错率和快速采集速度。

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