会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 1. 发明申请
    • Inspecting method by using mark partitioning
    • 使用标记分区检查方法
    • US20090245620A1
    • 2009-10-01
    • US12291777
    • 2008-11-13
    • Hyonam JooJung Seob LeeKeun Ho Rew
    • Hyonam JooJung Seob LeeKeun Ho Rew
    • G06K9/00
    • G06K9/00G06K2209/01G06T7/001G06T2207/10056G06T2207/30148
    • The present invention relates to a mark partitioning inspection method. A reference image and an inspection image are respectively acquired, and a correlation in a character unit for the reference image and the inspection image is obtained, and then the correlation value is compared with a first threshold value that has been previously set. Then, when the correlation value is greater than the first threshold value, the relevant character is partitioned into a predetermined number of regions, and a correlation between the reference image and the inspection image is obtained for each of the partitioned regions, and then a difference between the maximum and minimum values of the correlation is compared with a second threshold value that has been previously set. Here, even if a low defectiveness is revealed, the mark partitioning inspection method is capable of precisely determining whether or not it is defective, by determining the inspection image to be defective if the difference between the maximum and minimum values of the correlation is greater than the second threshold value, and by determining the inspection image to be normal when the difference between the maximum and minimum values of the correlation is smaller than the second threshold value.
    • 本发明涉及一种标记分割检查方法。 分别获取参考图像和检查图像,并且获得用于参考图像和检查图像的字符单元的相关性,然后将相关值与先前设置的第一阈值进行比较。 然后,当相关值大于第一阈值时,相关字符被划分为预定数量的区域,并且针对每个分割区域获得参考图像和检查图像之间的相关性, 将相关的最大值和最小值之间的值与先前设置的第二阈值进行比较。 这里,即使显示低的缺陷,如果相关性的最大值和最小值之间的差异大于标准分割检查方法,则能够精确地确定其是否有缺陷,通过确定检查图像是有缺陷的 第二阈值,并且当相关的最大值和最小值之间的差小于第二阈值时,通过将检查图像确定为正常。
    • 2. 发明授权
    • Inspecting method by using mark partitioning
    • 使用标记分区检查方法
    • US08005291B2
    • 2011-08-23
    • US12291777
    • 2008-11-13
    • Hyonam JooJung Seob LeeKeun Ho Rew
    • Hyonam JooJung Seob LeeKeun Ho Rew
    • G06K9/00
    • G06K9/00G06K2209/01G06T7/001G06T2207/10056G06T2207/30148
    • The present invention relates to a mark partitioning inspection method. A reference image and an inspection image are respectively acquired, and a correlation in a character unit for the reference image and the inspection image is obtained, and then the correlation value is compared with a first threshold value that has been previously set. Then, when the correlation value is greater than the first threshold value, the relevant character is partitioned into a predetermined number of regions, and a correlation between the reference image and the inspection image is obtained for each of the partitioned regions, and then a difference between the maximum and minimum values of the correlation is compared with a second threshold value that has been previously set. Here, even if a low defectiveness is revealed, the mark partitioning inspection method is capable of precisely determining whether or not it is defective, by determining the inspection image to be defective if the difference between the maximum and minimum values of the correlation is greater than the second threshold value, and by determining the inspection image to be normal when the difference between the maximum and minimum values of the correlation is smaller than the second threshold value.
    • 本发明涉及一种标记分割检查方法。 分别获取参考图像和检查图像,并且获得用于参考图像和检查图像的字符单元的相关性,然后将相关值与先前设置的第一阈值进行比较。 然后,当相关值大于第一阈值时,相关字符被划分为预定数量的区域,并且针对每个分割区域获得参考图像和检查图像之间的相关性, 将相关的最大值和最小值之间的值与先前设置的第二阈值进行比较。 这里,即使显示低的缺陷,如果相关性的最大值和最小值之间的差异大于标准分割检查方法,则通过将检查图像确定为有缺陷,能够精确地确定其是否有缺陷 第二阈值,并且当相关的最大值和最小值之间的差小于第二阈值时,通过将检查图像确定为正常。