opensource/PaddleOCR
3
0
Fork 0
mirror of https://github.com/PaddlePaddle/PaddleOCR.git synced 2026-09-24 23:33:08 +08:00
Code Issues Packages Projects Releases Wiki Activity

add case

Browse Source
This commit is contained in:
andyjpaddle
2022-06-16 10:52:35 +00:00
parent 8686dddf84
commit 9cc37f213a
1 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
applications/液晶屏读数识别.md
+3 -3
View File
@@ -58,7 +58,7 @@ def get_one_image(train):
get_one_image(train)
```
![det_png](./imgs/det.png)
![det_png](imgs/det.png)
### 模型训练
@@ -343,7 +343,7 @@ def get_one_image(train):
get_one_image(train)
```
![rec_png](./imgs/rec.png)
![rec_png](imgs/rec.png)
### 模型训练
#### 下载预训练模型
@@ -475,7 +475,7 @@ plt.imshow(img)
plt.show()
```
![sys_res_png](./imgs/sys_res.png)
![sys_res_png](imgs/sys_res.png)
### 后处理
如果需要获取key-value信息，可以基于启发式的规则，将识别结果与关键字库进行匹配；如果匹配上了，则取该字段为key, 后面一个字段为value。