YCG09 / chinese_ocr

CTPN + DenseNet + CTC based end-to-end Chinese OCR implemented using tensorflow and keras
Apache License 2.0
2.76k stars 1.08k forks source link

results #323

Closed liweizhong666 closed 4 years ago

liweizhong666 commented 4 years ago

执行demo.py results中每个图像对应的txt中的数字代表什么

kaiyiuw commented 4 years ago

如果你是指, e.g. img_0001.jpg 12 34 55 899 21 99 90 81 778 800 當中的數字是每個字的代號,e.g. 如沒記錯,29 代表"我" 字

liweizhong666 commented 4 years ago

Thanks for your answer. 那英文或数字呢,每一个字母或数字是一个编号吗? 例如:0: [array([1279, 23, 1488, 22, 1280, 70, 1488, 68]), u'42.88D']

kaiyiuw commented 4 years ago

英文或数字,我也這做的,例如: 1=A, 2=B, ...., 26=Z 不過不確定對不對的。

而這array 中的數列,應該是, scale up 過的box coordinates.

liweizhong666 commented 4 years ago

ok, thank you.