• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>International Agricultural Engineering Journal >An algorithm for the recognition of different fruit varieties
【24h】

An algorithm for the recognition of different fruit varieties

机译：识别不同水果品种的算法

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

A machine vision algorithm for the automatic recognition of different fruits was developed. Five fruits were investigated; red Fuji apple, blueberry, cherry, peach and pear. The machine vision system was composed of a color charge coupled device (CCD) camera to capture apple images, and a personal computer to process and analyze the images. As results showed that the fruit had a distinct red trichromatic coefficient among the objects in the image, the fruit portion in the image was enhanced using the red trichromatic coefficient. The intensity histogram of the enhanced image had a bimodal distribution, for the fruit portion and the background portion. Finding the maximum gray level variance of the red trichromatic coefficient between the fruit andthe background determined the optimal threshold. The optimal threshold had the minimum probability between the peaks of the two distributions. Results of the segmentation using the optimal threshold showed a success rate of more than 88 percent among the five fruits tested. This algorithm could be used for a "multifruit" harvesting robot.
机译：开发了一种用于自动识别不同水果的机器视觉算法。调查了五种水果;红富士苹果，蓝莓，樱桃，桃和梨。机器视觉系统由用于捕获苹果图像的彩色电荷耦合器件（CCD）相机和用于处理和分析图像的个人计算机组成。结果表明，水果在图像对象之间具有明显的红色三色系数，使用红色三色系数增强了图像中的水果部分。对于水果部分和背景部分，增强图像的强度直方图具有双峰分布。找到水果和背景之间的红色三色系数的最大灰度方差确定了最佳阈值。最佳阈值在两个分布的峰值之间具有最小的概率。使用最佳阈值进行分割的结果显示，在测试的五种水果中，成功率超过88％。该算法可用于“多果”收获机器人。

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