Skip to content

Latest commit

 

History

History
11 lines (11 loc) · 592 Bytes

File metadata and controls

11 lines (11 loc) · 592 Bytes

picMatch

matching for the pictures with a enviromental dataset,#ORB,DBOW,Tree of Words#,detectes the descriptors and

1. train the dictionary of enviromental dataset.

2. match the imput picture with the dataset

3. find the most biggest N scores

4. calculate the effective numbers of match for every pairs of matches using VFC

5. get the most likely result

近似匹配效果

商标牌的寻找

1.先找到包含商标牌的近似区域

例子1