Image：OCR-Python

Question

我有一个典型的验证码图像，只包含数字。

实施例。

我想从此图片中提取 78614 。

我试过很少的库＆amp;使用OCR-Python的代码。但它返回0。

示例代码-1

from captcha_solver import CaptchaSolver

solver = CaptchaSolver('browser')
with open('captcha.png', 'rb') as inp:
raw_data = inp.read()
print(solver.solve_captcha(raw_data))

示例代码-2

from PIL import Image

def p(img, letter):
    A = img.load()
    B = letter.load()
    mx = 1000000
    max_x = 0
    x = 0
    for x in range(img.size[0] - letter.size[0]):
        _sum = 0
        for i in range(letter.size[0]):
            for j in range(letter.size[1]):
                _sum = _sum + abs(A[x+i, j][0] - B[i, j][0])
        if _sum < mx :
            mx = _sum
            max_x = x
    return mx, max_x


def ocr(im, threshold=200, alphabet="0123456789abcdef"):
    img = Image.open(im)
    img = img.convert("RGB")
    box = (8, 8, 58, 18)
    img = img.crop(box)
    pixdata = img.load()

    letters = Image.open(im)
    ledata = letters.load()

    # Clean the background noise, if color != white, then set to black.
    for y in range(img.size[1]):
        for x in range(img.size[0]):
            if (pixdata[x, y][0] > threshold) \
                    and (pixdata[x, y][1] > threshold) \
                    and (pixdata[x, y][2] > threshold):

                pixdata[x, y] = (255, 255, 255, 255)
            else:
                pixdata[x, y] = (0, 0, 0, 255)

    counter = 0;
    old_x = -1;

    letterlist = []

    for x in range(letters.size[0]):
        black = True
        for y in range(letters.size[1]):
            if ledata[x, y][0] <> 0 :
                black = False
                break
        if black :
            if True :
                box = (old_x + 1, 0, x, 10)
                letter = letters.crop(box)
                t = p(img, letter);
                print counter, x, t
                letterlist.append((t[0], alphabet[counter], t[1]))
            old_x = x
            counter += 1

    box = (old_x + 1, 0, 140, 10)
    letter = letters.crop(box)
    t = p(img, letter)
    letterlist.append((t[0], alphabet[counter], t[1]))

    t = sorted(letterlist)
    t = t[0:5]  # 5-letter captcha

    final = sorted(t, key=lambda e: e[2])
    answer = ""
    for l in final:
        answer = answer + l[1]
    return answer

print(ocr('captcha.png'))

有没有人有机会从这种典型的验证码中获取/提取文本？

Answer 1

您可以使用机器学习（神经网络）模型来解决验证码，它几乎总是优于免费的OCR或任何其他方法。

这是一个很好的起点：https://medium.com/@ageitgey/how-to-break-a-captcha-system-in-15-minutes-with-machine-learning-dbebb035a710