检测模型和识别模型区别 #12687
Replies: 4 comments
-
检测模型是检测你文字所在的位置,而识别模型是把你输入图片中的文字识别出来 |
Beta Was this translation helpful? Give feedback.
-
感谢回复!但是我在测试过程中两个模型都同时返回了位置和文字内容及概率。如下:[[[[[3.0, 2.0], [801.0, 3.0], [801.0, 42.0], [3.0, 42.0]], ('DAVIDSON DOUGLAS', 0.9832786321640015)], [[[1748.0, 15.0], [2098.0, 15.0], [2098.0, 68.0], [1748.0, 68.0]], ('Semi-t', 0.9988662600517273)], [[[11.0, 44.0], [368.0, 44.0], [368.0, 88.0], [11.0, 88.0]], ('DOS: 06.16.1976', 0.9855455160140991)], [[[1890.0, 99.0], [2118.0, 99.0], [2118.0, 145.0], [1890.0, 145.0]], ('Lortable', 0.9988487958908081)], [[[2022.0, 172.0], [2080.0, 172.0], [2080.0, 238.0], [2022.0, 238.0]], ('P', 0.979322612285614)]]] |
Beta Was this translation helpful? Give feedback.
-
你是怎么使用的呢?命令是什么呢? |
Beta Was this translation helpful? Give feedback.
-
the model first uses the detection model which helps in detecting texts from the image and then recognition model reads the texts |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
分别下载使用中英文超轻量PP-OCRv4模型(15.8M)的检测模型和识别模型过程中,发现返回结果都是一样的,请问这两个模型有什么区别?又该怎么选择使用呢?求回复……

Beta Was this translation helpful? Give feedback.
All reactions