检测模型和识别模型无法识别下划线 #14529
Unanswered
flow3rdown
asked this question in
Q&A
检测模型和识别模型无法识别下划线
#14529
Replies: 2 comments 3 replies
-
应该要微调一下吧 |
Beta Was this translation helpful? Give feedback.
0 replies
-
识别字库里也没有下划线字符,需要微调: |
Beta Was this translation helpful? Give feedback.
3 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
🔎 Search before asking
🐛 Bug (问题描述)
您好,使用ch_PP-OCRv4_det_infer无法检测出下划线信息,如下图所示:

尝试修改
use_dilation
,det_db_box_thresh
和det_db_unclip_ratio
等参数也无法解决该问题。另外,用包含下划线的图用ch_PP-OCRv4_rec_infer也无法识别出下划线信息。
🏃♂️ Environment (运行环境)
paddleocr 2.9.1
paddlepaddle-gpu 3.0.0b
🌰 Minimal Reproducible Example (最小可复现问题的Demo)
python3 tools/infer/predict_det.py
--det_algorithm="DB"
--det_model_dir="ch_PP-OCRv4_det_infer"
--image_dir="xxx"
--use_gpu=True
--use_dilation=True
--det_db_box_thresh=0.3
--det_db_unclip_ratio=1.8
Beta Was this translation helpful? Give feedback.
All reactions