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
-
🔎 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