从带有坐标的名片图像中提取徽标和文本

Question

我有一张名片。我想从带有坐标的名片中获取徽标和所有文本。所以我可以使上传的图像在 HTML Canvas 上可编辑。我看过很多例子，但我找不到我正在寻找的确切例子。我只发现从图像中获取文本。我也尝试使用 Google Vision API，但它也只提供文本。 我是 python 新手。

这是一个示例图像。

在以下代码中，我必须选择要提取的徽标。我需要它自动查找和提取。

# import the necessary packages
import argparse
import cv2

# initialize the list of reference points and boolean indicating
# whether cropping is being performed or not
ref_point = []
cropping = False

def shape_selection(event, x, y, flags, param):
  # grab references to the global variables
  global ref_point, cropping

  # if the left mouse button was clicked, record the starting
  # (x, y) coordinates and indicate that cropping is being
  # performed
  if event == cv2.EVENT_LBUTTONDOWN:
    ref_point = [(x, y)]
    cropping = True

  # check to see if the left mouse button was released
  elif event == cv2.EVENT_LBUTTONUP:
    # record the ending (x, y) coordinates and indicate that
    # the cropping operation is finished
    ref_point.append((x, y))
    cropping = False

    # draw a rectangle around the region of interest
    cv2.rectangle(image, ref_point[0], ref_point[1], (0, 255, 0), 2)
    cv2.imshow("image", image)

# construct the argument parser and parse the arguments
ap = argparse.ArgumentParser()
ap.add_argument("-i", "--image", required=True, help="Path to the image")
args = vars(ap.parse_args())

# load the image, clone it, and setup the mouse callback function
image = cv2.imread(args["image"])
clone = image.copy()
cv2.namedWindow("image")
cv2.setMouseCallback("image", shape_selection)

# keep looping until the 'q' key is pressed
while True:
  # display the image and wait for a keypress
  cv2.imshow("image", image)
  key = cv2.waitKey(1) & 0xFF

  # if the 'r' key is pressed, reset the cropping region
  if key == ord("r"):
    image = clone.copy()

  # if the 'c' key is pressed, break from the loop
  elif key == ord("c"):
    break

# if there are two reference points, then crop the region of interest
# from teh image and display it
if len(ref_point) == 2:
  crop_img = clone[ref_point[0][1]:ref_point[1][1], ref_point[0][0]:ref_point[1][0]]
  cv2.imshow("crop_img", crop_img)
  cv2.waitKey(0)

# close all open windows
cv2.destroyAllWindows()

Answer 1

你可以试试 ABBYY 云 API：

https://www.abbyy.com/en-gb/cloud-ocr-sdk/features/

API 将为您提供带有坐标的所有文本，并且您可以将图像元素（只要可检测到）作为纯图像返回。通过一些逻辑，您可以将它们组合在一起，形成一个文档，其中包含所有文本元素作为真实文本，所有图像作为正确位置的图像。

但请记住，在 OCR 开始之前对图像进行了一些预处理。这意味着图像的质量可能已经改变。因此，使用从 API 获得的坐标从原始扫描中提取图像部分可能是个好主意。

https://www.ocrsdk.com/documentation/specifications/export-formats/

API 非常好，可以为您提供与谷歌云视觉非常相似的 OCR 结果。而且你有更多的特征和参数来调整结果。但是 ABBYY API 比 google API 贵很多。