我要从PDF表格中提取值，然后将数据保存在json文件中，

我使用pytesseract从将pdf转换为的图像中获取文本，

现在的问题是，它给了我一个包含所有数据的大字符串。

我尝试通过SP进行拆分，这逐行给了我数据，我现在想要的是一种基于列名分离数据并保存为json格式的方法。

以下是具有表的图像。

就是说，我需要一个数组，我可以在其中保存值是SP的所有者，

资产包含该列中的所有数据，

**以下是从图像中获取全文的代码：**

import cv2

import pytesseract as tess

tess.pytesseract.tesseract_cmd = r'C:\Users\faisa\AppData\Local\Tesseract-OCR\tesseract.exe'

from PIL import Image

# img = Image.open("image.png")

img = cv2.imread("Test 1.png")

text = tess.image_to_string(img)

print(text)

结果是：

Filing ID #20018011

PERIODIC TRANSACTION REPORT

Clerk of the House of Representatives » Legislative Resource Center ¢ 135 Cannon Building « Washington, DC 20515

FILer INFORMATION

Name: Hon. Nancy Pelosi

Status: Member

State/District: CA12

TRANSACTIONS

ID Owner Asset Transaction Date Notification Amount Cap.

Type Date Gains >

$200?

SP AllianceBernstein Holding L.P. P 12/22/2020 12/22/2020 $500,001 - ‘I

Units (AB) [ST] $1,000,000

FILinc Status: New

DescriPTION: Purchased 20,000 shares.

SP Apple Inc. (AAPL) [OP] P 12/22/2020 12/22/2020 $250,001 - ~

$500,000

Fiuine Status: New

Description: Purchased 100 call options with a strike price of $100 and an expiration date of 1/21/22.

SP Tesla, Inc. (TSLA) [OP] P 12/22/2020 12/22/2020 $500,001 - ~

$1,000,000

Fiine Status: New

Description: Purchased 25 call options with a strike price of $500 and an expiration date of 3/18/22.

SP Walt Disney Company (DIS) [OP] P 12/22/2020 12/22/2020 $500,001 - ‘a

$1,000,000

Fiuine Status: New

Description: Purchased 100 call options with a strike price of $100 and an expiration date of 1/21/22.

* For the complete list of asset type abbreviations, please visit https: //fd.house.gov/reference/asset-type-codes.aspx.

INITIAL PuBLic OFFERINGS

© Yes © No

CERTIFICATION AND SIGNATURE

lv 1 CERTIFY that the statements I have made on the attached Periodic Transaction Report are true, complete, and correct to the

这是我将表中的行拆分为不同行并保存到临时字典的部分。

result = """Filing ID #20018011

PERIODIC TRANSACTION REPORT

Clerk of the House of Representatives » Legislative Resource Center ¢ 135 Cannon Building « Washington, DC 20515

FILer INFORMATION

Name: Hon. Nancy Pelosi

Status: Member

State/District: CA12

TRANSACTIONS

ID Owner Asset Transaction Date Notification Amount Cap.

Type Date Gains >

$200?

SP AllianceBernstein Holding L.P. P 12/22/2020 12/22/2020 $500,001 - ‘I

Units (AB) [ST] $1,000,000

FILinc Status: New

DescriPTION: Purchased 20,000 shares.

SP Apple Inc. (AAPL) [OP] P 12/22/2020 12/22/2020 $250,001 - ~

$500,000

Fiuine Status: New

Description: Purchased 100 call options with a strike price of $100 and an expiration date of 1/21/22.

SP Tesla, Inc. (TSLA) [OP] P 12/22/2020 12/22/2020 $500,001 - ~

$1,000,000

Fiine Status: New

Description: Purchased 25 call options with a strike price of $500 and an expiration date of 3/18/22.

SP Walt Disney Company (DIS) [OP] P 12/22/2020 12/22/2020 $500,001 - ‘a

$1,000,000

Fiuine Status: New

Description: Purchased 100 call options with a strike price of $100 and an expiration date of 1/21/22.

* For the complete list of asset type abbreviations, please visit https: //fd.house.gov/reference/asset-type-codes.aspx.

INITIAL PuBLic OFFERINGS

© Yes © No

CERTIFICATION AND SIGNATURE

lv 1 CERTIFY that the statements I have made on the attached Periodic Transaction Report are true, complete, and correct to the

ye to"""

table = []

row = 1

for temp_ar in result.strip().split('\nSP ')[1:]:

temp_ar = temp_ar.replace('~','').replace("‘a",'').replace("‘I",'')

line1 = temp_ar.split(',000\n')[0] + ",000"

line2 = temp_ar[temp_ar.index(line1)+len(line1):].split('\nDescription:')[0]

line3 = temp_ar[temp_ar.index(line2)+len(line2):].split("For the complete list")[0]

table.append({

"row" : row,

'row_{}_line_1'.format(row): line1.strip().replace('\n',''),

'row_{}_line_2'.format(row): line2.strip().replace('\n',''),

'row_{}_line_3'.format(row): line3.strip().replace('\n',''),

})

row += 1

print(table)

结果是：

[{'row': 1, 'row_1_line_1': 'AllianceBernstein Holding L.P. P 12/22/2020 12/22/2020 $500,001 - Units (AB) [ST] $1,000,000', 'row_1_line_2': 'FILinc Status: NewDescriPTION: Purchased 20,000 shares.', 'row_1_line_3': ''}, {'row': 2, 'row_2_line_1': 'Apple Inc. (AAPL) [OP] P 12/22/2020 12/22/2020 $250,001 - $500,000', 'row_2_line_2': 'Fiuine Status: New', 'row_2_line_3': 'Description: Purchased 100 call options with a strike price of $100 and an expiration date of 1/21/22.'}, {'row': 3, 'row_3_line_1': 'Tesla, Inc. (TSLA) [OP] P 12/22/2020 12/22/2020 $500,001 - $1,000,000', 'row_3_line_2': 'Fiine Status: New', 'row_3_line_3': 'Description: Purchased 25 call options with a strike price of $500 and an expiration date of 3/18/22.'}, {'row': 4, 'row_4_line_1': 'Walt Disney Company (DIS) [OP] P 12/22/2020 12/22/2020 $500,001 - $1,000,000', 'row_4_line_2': 'Fiuine Status: New', 'row_4_line_3': 'Description: Purchased 100 call options with a strike price of $100 and an expiration date of 1/21/22.*'}]

现在您可以看到它混合了1行中的列，并且很难动态拆分每个列并合并其行。

我确实尝试通过检测是否是日期来分割第一行的每一行，但是不走运，它没有检测到日期。

我需要这个脚本来处理类似的PDF文件，因此我需要一种动态的方式来找出该列，任何帮助将不胜感激。谢谢。

解决方案

暂无回答