首页> 外文会议>Chinese Intelligent Systems Conference >Chapter 10 Apple Nighttime Images Enhancement Algorithm for Harvesting Robot
【24h】

Chapter 10 Apple Nighttime Images Enhancement Algorithm for Harvesting Robot

机译:第10章摄取机器人的苹果夜间图像增强算法

获取原文
获取外文期刊封面目录资料

摘要

In order to enhance the applicability and efficiency of harvesting robot to ensure that people can timely pick ripe fruit, the robot need to have an ability of continuous recognition and harvest at night. For some disadvantages of night vision images, Retinex algorithm for image enhancement based on bilateral filter is presented. Bilateral filter which has a function of edge preservation is adopted to improve the smooth, evaluate the illumination and remove unfavorable illumination effects from the original image. Then the reflectance of the image from above that contains just the characteristics of the object itself can be obtained. Finally, apple nighttime image enhancement is implemented. The experimental results show that the above method can more accurately evaluate the illumination of high-contrast edge regions, to suppress noise, enhance image contrast and improve overall visual effects of the image.
机译:为了提高收获机器人的适用性和效率,确保人们可以及时挑选成熟水果,机器人需要在夜间连续识别和收获的能力。对于夜视图像的一些缺点,提出了基于双边滤波器的图像增强的Retinex算法。采用具有边缘保存功能的双侧过滤器来改善光滑,评价照明并从原始图像中删除不利的照明效果。然后可以获得从上面的图像的反射率,该图像可以获得仅包含对象本身的特征。最后,实现了苹果夜间图像增强。实验结果表明,上述方法可以更准确地评估高对比度边缘区域的照明,抑制噪声,增强图像对比度并提高图像的整体视觉效果。

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