Commit 54671891 authored by EvilCalf's avatar EvilCalf

modify some structure

parent 51783422
...@@ -9,14 +9,12 @@ import os ...@@ -9,14 +9,12 @@ import os
import pandas as pd import pandas as pd
tesseract_cmd = r'.\tesseract-ocr\tesseract.exe' tesseract_cmd = r".\tesseract-ocr\tesseract.exe"
for root, dirs, files in os.walk("labimage/"): for root, dirs, files in os.walk("labimage/"):
for file in files: for file in files:
image = Image.open(root + "/" + file) image = Image.open(root + "/" + file)
content = pytesseract.image_to_data( content = pytesseract.image_to_data(image, lang="chi_sim43", output_type="dict")
image, lang="chi_sim43", output_type="dict"
)
for i in range(len(content["text"])): for i in range(len(content["text"])):
if 0 < len(content["text"][i]): if 0 < len(content["text"][i]):
if content["text"][i] == "姓名" or ( if content["text"][i] == "姓名" or (
...@@ -46,10 +44,7 @@ for root, dirs, files in os.walk("labimage/"): ...@@ -46,10 +44,7 @@ for root, dirs, files in os.walk("labimage/"):
print(cnt) print(cnt)
image.paste((0, 0, 0), (x - 10, y - 10, x + w + 400, y + h + 30)) image.paste((0, 0, 0), (x - 10, y - 10, x + w + 400, y + h + 30))
image.save("./Output/" + file) image.save("./Output/" + file)
data = pd.DataFrame({'name': [cnt], 'dir': [root + "/" + file]}) data = pd.DataFrame({"name": [cnt], "dir": [root + "/" + file]})
data.to_csv("name2file.csv",mode='a',header=False) data.to_csv("name2file.csv", mode="a", header=False)
break break
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment