[英]Split text lines in scanned document
I am trying to find a way to break the split the lines of text in a scanned document that has been adaptive thresholded.我正在尝试找到一种方法来打破已自适应阈值化的扫描文档中文本行的拆分。 Right now, I am storing the pixel values of the document as unsigned ints from 0 to 255, and I am taking the average of the pixels in each line, and I split the lines into ranges based on whether the average of the pixels values is larger than 250, and then I take the median of each range of lines for which this holds.
现在,我将文档的像素值存储为从 0 到 255 的无符号整数,我取每行中像素的平均值,然后根据像素值的平均值是否为大于 250,然后我取它所适用的每个线范围的中值。 However, this methods sometimes fails, as there can be black splotches on the image.
然而,这种方法有时会失败,因为图像上可能会有黑色斑点。
Is there a more noise-resistant way to do this task?有没有更抗噪的方法来完成这项任务?
EDIT: Here is some code.编辑:这是一些代码。 "warped" is the name of the original image, "cuts" is where I want to split the image.
“warped”是原始图像的名称,“cuts”是我要分割图像的地方。
warped = threshold_adaptive(warped, 250, offset = 10)
warped = warped.astype("uint8") * 255
# get areas where we can split image on whitespace to make OCR more accurate
color_level = np.array([np.sum(line) / len(line) for line in warped])
cuts = []
i = 0
while(i < len(color_level)):
if color_level[i] > 250:
begin = i
while(color_level[i] > 250):
i += 1
cuts.append((i + begin)/2) # middle of the whitespace region
else:
i += 1
From your input image, you need to make text as white, and background as black从您的输入图像中,您需要将文本设为白色,将背景设为黑色
You need then to compute the rotation angle of your bill.然后您需要计算帐单的旋转角度。 A simple approach is to find the
minAreaRect
of all white points ( findNonZero
), and you get:一个简单的方法是找到所有白点的
minAreaRect
( findNonZero
),你会得到:
Then you can rotate your bill, so that text is horizontal:然后你可以旋转你的帐单,让文字是水平的:
Now you can compute horizontal projection ( reduce
).现在您可以计算水平投影 (
reduce
)。 You can take the average value in each line.您可以取每行的平均值。 Apply a threshold
th
on the histogram to account for some noise in the image (here I used 0
, ie no noise).在直方图上应用阈值
th
以说明图像中的一些噪声(这里我使用0
,即没有噪声)。 Lines with only background will have a value >0
, text lines will have value 0
in the histogram.只有背景的行的值
>0
,文本行在直方图中的值为0
。 Then take the average bin coordinate of each continuous sequence of white bins in the histogram.然后取直方图中每个连续的白色 bin 序列的平均 bin 坐标。 That will be the
y
coordinate of your lines:那将是你的线的
y
坐标:
Here the code.这是代码。 It's in C++, but since most of the work is with OpenCV functions, it should be easy convertible to Python.
它是用 C++ 编写的,但由于大部分工作都是使用 OpenCV 函数,因此它应该很容易转换为 Python。 At least, you can use this as a reference:
至少,您可以将其用作参考:
#include <opencv2/opencv.hpp>
using namespace cv;
using namespace std;
int main()
{
// Read image
Mat3b img = imread("path_to_image");
// Binarize image. Text is white, background is black
Mat1b bin;
cvtColor(img, bin, COLOR_BGR2GRAY);
bin = bin < 200;
// Find all white pixels
vector<Point> pts;
findNonZero(bin, pts);
// Get rotated rect of white pixels
RotatedRect box = minAreaRect(pts);
if (box.size.width > box.size.height)
{
swap(box.size.width, box.size.height);
box.angle += 90.f;
}
Point2f vertices[4];
box.points(vertices);
for (int i = 0; i < 4; ++i)
{
line(img, vertices[i], vertices[(i + 1) % 4], Scalar(0, 255, 0));
}
// Rotate the image according to the found angle
Mat1b rotated;
Mat M = getRotationMatrix2D(box.center, box.angle, 1.0);
warpAffine(bin, rotated, M, bin.size());
// Compute horizontal projections
Mat1f horProj;
reduce(rotated, horProj, 1, CV_REDUCE_AVG);
// Remove noise in histogram. White bins identify space lines, black bins identify text lines
float th = 0;
Mat1b hist = horProj <= th;
// Get mean coordinate of white white pixels groups
vector<int> ycoords;
int y = 0;
int count = 0;
bool isSpace = false;
for (int i = 0; i < rotated.rows; ++i)
{
if (!isSpace)
{
if (hist(i))
{
isSpace = true;
count = 1;
y = i;
}
}
else
{
if (!hist(i))
{
isSpace = false;
ycoords.push_back(y / count);
}
else
{
y += i;
count++;
}
}
}
// Draw line as final result
Mat3b result;
cvtColor(rotated, result, COLOR_GRAY2BGR);
for (int i = 0; i < ycoords.size(); ++i)
{
line(result, Point(0, ycoords[i]), Point(result.cols, ycoords[i]), Scalar(0, 255, 0));
}
return 0;
}
Basic steps as @Miki, @Miki 的基本步骤,
- read the source
阅读源代码
- threshed
脱粒
- find minAreaRect
找到 minAreaRect
- warp by the rotated matrix
通过旋转矩阵扭曲
- find and draw upper and lower bounds
查找并绘制上限和下限
While code in Python :在 Python 中编写代码时:
#!/usr/bin/python3
# 2018.01.16 01:11:49 CST
# 2018.01.16 01:55:01 CST
import cv2
import numpy as np
## (1) read
img = cv2.imread("img02.jpg")
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
## (2) threshold
th, threshed = cv2.threshold(gray, 127, 255, cv2.THRESH_BINARY_INV|cv2.THRESH_OTSU)
## (3) minAreaRect on the nozeros
pts = cv2.findNonZero(threshed)
ret = cv2.minAreaRect(pts)
(cx,cy), (w,h), ang = ret
if w>h:
w,h = h,w
ang += 90
## (4) Find rotated matrix, do rotation
M = cv2.getRotationMatrix2D((cx,cy), ang, 1.0)
rotated = cv2.warpAffine(threshed, M, (img.shape[1], img.shape[0]))
## (5) find and draw the upper and lower boundary of each lines
hist = cv2.reduce(rotated,1, cv2.REDUCE_AVG).reshape(-1)
th = 2
H,W = img.shape[:2]
uppers = [y for y in range(H-1) if hist[y]<=th and hist[y+1]>th]
lowers = [y for y in range(H-1) if hist[y]>th and hist[y+1]<=th]
rotated = cv2.cvtColor(rotated, cv2.COLOR_GRAY2BGR)
for y in uppers:
cv2.line(rotated, (0,y), (W, y), (255,0,0), 1)
for y in lowers:
cv2.line(rotated, (0,y), (W, y), (0,255,0), 1)
cv2.imwrite("result.png", rotated)
Finally result :最后结果:
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.