Why can't OCR detect the text in this photo?
5 次查看(过去 30 天)
显示 更早的评论
I am trynig to detect all the text the attached photo. I have have use the OCR function and it can't detect it for what ever reason. I have done the traditional blur/filter, edge detect, and then OCR, as well as OCR on the original image.
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/333090/image.jpeg)
0 个评论
采纳的回答
Bjorn Gustavsson
2020-7-17
It seems to have something to do with automatic thresholding. If I cut out most of the black background it seems to work just fine, if too much of that black is included no characters are found. This presumably means that somewhere there's a thresholding that if theres too much proper black (or white if you take the negative of this image) the entire gray area is lumped together into a unseparated white region?
(Our vision is very good at this task, computers can still be easily fooled by things we would never fall for. On the other hand: optical illusions...)
HTH
0 个评论
更多回答(0 个)
另请参阅
类别
在 Help Center 和 File Exchange 中查找有关 Language Support 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!