为什么 Tesseract OCR 没有捕获正确的输出？如何解决？答案

【问题标题】：Why is Tesseract OCR not capturing the right output? how can this be solved?为什么 Tesseract OCR 没有捕获正确的输出？如何解决？
【发布时间】：2020-04-03 03:48:34
【问题描述】：

下面是我正在使用的代码，但它给了我奇怪的输出，而不是实际的文本。向有更多 OCR 经验的人寻求帮助。

这是我得到的输出 -

预期输出 -

import cv2
import pytesseract
import imutils

pytesseract.pytesseract.tesseract_cmd = r"C:\Program Files\Tesseract-OCR\tesseract.exe"

# Resize, grayscale, Otsu's threshold
image = cv2.imread('image1.jpg')
image = imutils.resize(image, width=500)
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[4]

# Invert image and perform morphological operations
inverted = 255 - thresh
kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (15,3))
close = cv2.morphologyEx(inverted, cv2.MORPH_CLOSE, kernel, iterations=1)

# Find contours and filter using aspect ratio and area
cnts = cv2.findContours(close, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if len(cnts) == 2 else cnts[4]
for c in cnts:
    area = cv2.contourArea(c)
    peri = cv2.arcLength(c, True)
    approx = cv2.approxPolyDP(c, 0.01 * peri, True)
    x,y,w,h = cv2.boundingRect(approx)
    aspect_ratio = w / float(h)
    if (aspect_ratio >= 2.5 or area < 75):
        cv2.drawContours(thresh, [c], -1, (255,255,255), -1)

# Blur and perform text extraction
thresh = cv2.GaussianBlur(thresh, (3,3), 0)
data = pytesseract.image_to_string(thresh, lang='eng',config='--psm 6')
print(data)

cv2.imshow('close', close)
cv2.imshow('thresh', thresh)
cv2.waitKey()

【问题讨论】：

标签： python ocr tesseract text-extraction

【解决方案1】：

我稍微修改了您的代码。它可以解码除 1 位以外的所有字符。我不知道为什么。也许你可以微调自己。

我使用轨迹栏来找到合适的值。如果要使用轨迹栏，请取消注释代码。

import cv2
import pytesseract

def nothing(x):
    pass

image = cv2.imread('image1.jpg')
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

cv2.namedWindow("Threshold")
cv2.createTrackbar("min", "Threshold", 0, 255, nothing)
cv2.createTrackbar("max", "Threshold", 0, 255, nothing)

while True:
    t_min = cv2.getTrackbarPos("min", "Threshold")
    t_max = cv2.getTrackbarPos("max", "Threshold")

    ret, thresh = cv2.threshold(gray.copy(), 180, 255, cv2.THRESH_BINARY_INV)
    # ret, thresh = cv2.threshold(gray.copy(), t_min, t_max, cv2.THRESH_BINARY_INV)
    inverted = cv2.bitwise_not(thresh)
    gauss = cv2.GaussianBlur(inverted, (3,3), 0)

    custom_config = '-l eng --oem 3 --psm 6 -c tessedit_char_whitelist="ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789./ "'
    data = pytesseract.image_to_string(gauss, config=custom_config)
    print(data)

    cv2.imshow('gauss', gauss)

    key = cv2.waitKey(1000)
    if key == 27:
        break

cv2.destroyAllWindows()

这是输出。

3.59 4.79
/pek /pek
Tepung Penaik / Halus Bluekey 1kg Marjerin Seri Pelangi kg

【讨论】：