[tesseract-ocr] Difference trained data for Chinese
Good day! Recently I was using tesseract (4.0 alpha) to do Chinese OCR and it works really great. Now I want to pick up a best model to use but I find several versions. What is the difference between them? 1. chi_sim from https://github.com/tesseract-ocr/tesseract/wiki/Data-Files (around 50M) 2. chi_sim from https://github.com/tesseract-ocr/tessdata/tree/master/best (around 13M) 3. chi_sim_vert from https://github.com/tesseract-ocr/tessdata/tree/master/best (around 13M) 4. HanS from https://github.com/tesseract-ocr/tessdata/tree/master/best (around 16M) All of them can work but the results are slightly different. From my own evaluation #4 is the best, but I don't have any insight. Appreciate for any help. -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to tesseract-ocr+unsubscr...@googlegroups.com. To post to this group, send email to tesseract-ocr@googlegroups.com. Visit this group at https://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/8cc88ed2-99c3-445e-b758-83ade0f680aa%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[tesseract-ocr] Re: Difference trained data for Chinese
Please see https://github.com/tesseract-ocr/tessdata/issues/72 On Friday, August 11, 2017 at 2:26:55 PM UTC+5:30, Yang Yu wrote: > > Good day! > > Recently I was using tesseract (4.0 alpha) to do Chinese OCR and it works > really great. Now I want to pick up a best model to use but I find several > versions. What is the difference between them? > > 1. chi_sim from https://github.com/tesseract-ocr/tesseract/wiki/Data-Files > (around 50M) > 2. chi_sim from https://github.com/tesseract-ocr/tessdata/tree/master/best > (around 13M) > 3. chi_sim_vert from > https://github.com/tesseract-ocr/tessdata/tree/master/best (around 13M) > 4. HanS from https://github.com/tesseract-ocr/tessdata/tree/master/best > (around 16M) > > All of them can work but the results are slightly different. From my own > evaluation #4 is the best, but I don't have any insight. > > Appreciate for any help. > -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to tesseract-ocr+unsubscr...@googlegroups.com. To post to this group, send email to tesseract-ocr@googlegroups.com. Visit this group at https://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/2b698539-1bd3-4ad6-b753-84b90d13f79b%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[tesseract-ocr] Re: How Much TO Enlarge Screenshot?
cause tesseract operate at 300dpi, so you should change to that level. for example the screen dpi is 72, then enlarge the screenshot to 400% On Friday, August 4, 2017 at 1:56:32 AM UTC+8, James Lee wrote: > > Is there way to find out how much to enlarge a screenshot for best > accuracy? > Is there a math formula if I know the internal display resolution (not > dpi?) > Thanks! > > -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to tesseract-ocr+unsubscr...@googlegroups.com. To post to this group, send email to tesseract-ocr@googlegroups.com. Visit this group at https://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/c342a5cc-006b-4a0b-a352-d2394e57ee85%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[tesseract-ocr] How to know how many symbol is a word in pagelayout?
i can use the code below to draw every word and every symbol bounding box,
now i want to if i can know how many symbol in the word when i got a
word?
thanks for any info!
=
#include
#include
#include
#include
using namespace std;
int main() {
std::cout << "Hello, World!" << std::endl;
tesseract::TessBaseAPI api ;
api.InitForAnalysePage();
api.SetPageSegMode(tesseract::PSM_SPARSE_TEXT);
Pix *image = pixRead("/tmp/ytmp/en4.png");
//process gray color to white
l_uint32 pixel_color;
l_int32 r,g,b;
l_int32 width,height,depth;
width=0;
height=0;
pixGetDimensions(image,&width,&height,&depth);
printf("w=%d h=%d dep=%d\n",width,height,depth);
api.SetImage(image);
tesseract::PageIterator *iter = api.AnalyseLayout(true);
int word_count=0;
while (iter->Next(tesseract::RIL_WORD)) {
int left, top, right, bottom;
++word_count;
iter->BoundingBox(
tesseract::RIL_WORD,
&left, &top, &right, &bottom
);
*
//=== //i
got the word bounding box, but i want to know how may symbol in this
word?
//===*
pixRenderLine(image,left,top,left,bottom,3,L_CLEAR_PIXELS);
pixRenderLine(image,left,top,right,top,3,L_CLEAR_PIXELS);
pixRenderLine(image,left,bottom,right,bottom,3,L_CLEAR_PIXELS);
pixRenderLine(image,right,top,right,bottom,3,L_CLEAR_PIXELS);
}
iter->Begin();
while (iter->Next(tesseract::RIL_SYMBOL)) {
int left, top, right, bottom;
++word_count;
iter->BoundingBox(
tesseract::RIL_SYMBOL,
&left, &top, &right, &bottom
);
pixRenderLine(image,left,top,left,bottom,1,L_CLEAR_PIXELS);
pixRenderLine(image,left,top,right,top,1,L_CLEAR_PIXELS);
pixRenderLine(image,left,bottom,right,bottom,1,L_CLEAR_PIXELS);
pixRenderLine(image,right,top,right,bottom,1,L_CLEAR_PIXELS);
}
pixWrite("/tmp/ytmp/entt.png",image,IFF_PNG);
return 0;
}
--
You received this message because you are subscribed to the Google Groups
"tesseract-ocr" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to tesseract-ocr+unsubscr...@googlegroups.com.
To post to this group, send email to tesseract-ocr@googlegroups.com.
Visit this group at https://groups.google.com/group/tesseract-ocr.
To view this discussion on the web visit
https://groups.google.com/d/msgid/tesseract-ocr/1da35024-16f1-404a-aa2a-e06e1377aacf%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
