Is there a way i can detect the image orientation and rotate the image to the right angle?

Question

I am making a script that repairs scanned documents and i now need a way to detect the image orientation and rotate the image so its rotation is correct.

Right now my script is unreliable and isn't that precise.

right now I look for a line and it rotates the first line it sees correctly but this barely works except for a few images

img_before = cv2.imread('rotated_377.jpg')

img_gray = cv2.cvtColor(img_before, cv2.COLOR_BGR2GRAY)
img_edges = cv2.Canny(img_gray, 100, 100, apertureSize=3)
lines = cv2.HoughLinesP(img_edges, 1, math.pi / 180.0, 100, minLineLength=100, maxLineGap=5)

angles = []

for x1,y1,x2,y2 in lines[0]:
    angle = math.degrees(math.atan2(y2 - y1, x2 - x1))
    angles.append(angle)

median_angle = np.median(angles)
img_rotated = ndimage.rotate(img_before, median_angle)

print("Angle is {}".format(median_angle))
cv2.imwrite('rotated.jpg', img_rotated)

I want to make a script that gets an image like this one(don't mind the image its for testing purposes)

and rotates it in the right way so I get a correctly orientated image.

Mush

Answer 1

This is an interesting problem, i have tried with many approaches to correct orientation of document images but all of them have got different exceptions. I am sharing one of the approaches based on text orientation. For text region detection i am using gradient map of input image.

All other implementation details are commented in the code.

Please note that this only works if all the text present in image have same orientation.

#Document image orientation correction
#This approach is based on text orientation

#Assumption: Document image contains all text in same orientation

import cv2
import numpy as np

debug = True

#Display image
def display(img, frameName="OpenCV Image"):
    if not debug:
        return
    h, w = img.shape[0:2]
    neww = 800
    newh = int(neww*(h/w))
    img = cv2.resize(img, (neww, newh))
    cv2.imshow(frameName, img)
    cv2.waitKey(0)

#rotate the image with given theta value
def rotate(img, theta):
    rows, cols = img.shape[0], img.shape[1]
    image_center = (cols/2, rows/2)
    
    M = cv2.getRotationMatrix2D(image_center,theta,1)

    abs_cos = abs(M[0,0])
    abs_sin = abs(M[0,1])

    bound_w = int(rows * abs_sin + cols * abs_cos)
    bound_h = int(rows * abs_cos + cols * abs_sin)

    M[0, 2] += bound_w/2 - image_center[0]
    M[1, 2] += bound_h/2 - image_center[1]

    # rotate orignal image to show transformation
    rotated = cv2.warpAffine(img,M,(bound_w,bound_h),borderValue=(255,255,255))
    return rotated


def slope(x1, y1, x2, y2):
    if x1 == x2:
        return 0
    slope = (y2-y1)/(x2-x1)
    theta = np.rad2deg(np.arctan(slope))
    return theta


def main(filePath):
    img = cv2.imread(filePath)
    textImg = img.copy()

    small = cv2.cvtColor(textImg, cv2.COLOR_BGR2GRAY)

    #find the gradient map
    kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (3, 3))
    grad = cv2.morphologyEx(small, cv2.MORPH_GRADIENT, kernel)

    display(grad)

    #Binarize the gradient image
    _, bw = cv2.threshold(grad, 0.0, 255.0, cv2.THRESH_BINARY | cv2.THRESH_OTSU)
    display(bw)

    #connect horizontally oriented regions
    #kernal value (9,1) can be changed to improved the text detection
    kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (9, 1))
    connected = cv2.morphologyEx(bw, cv2.MORPH_CLOSE, kernel)
    display(connected)

    # using RETR_EXTERNAL instead of RETR_CCOMP
    # _ , contours, hierarchy = cv2.findContours(connected.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)
    contours, hierarchy = cv2.findContours(connected.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE) #opencv >= 4.0



    mask = np.zeros(bw.shape, dtype=np.uint8)
    #display(mask)
    #cumulative theta value
    cummTheta = 0
    #number of detected text regions
    ct = 0
    for idx in range(len(contours)):
        x, y, w, h = cv2.boundingRect(contours[idx])
        mask[y:y+h, x:x+w] = 0
        #fill the contour
        cv2.drawContours(mask, contours, idx, (255, 255, 255), -1)
        #display(mask)
        #ratio of non-zero pixels in the filled region
        r = float(cv2.countNonZero(mask[y:y+h, x:x+w])) / (w * h)

        #assume at least 45% of the area is filled if it contains text
        if r > 0.45 and w > 8 and h > 8:
            #cv2.rectangle(textImg, (x1, y), (x+w-1, y+h-1), (0, 255, 0), 2)

            rect = cv2.minAreaRect(contours[idx])
            box = cv2.boxPoints(rect)
            box = np.int0(box)
            cv2.drawContours(textImg,[box],0,(0,0,255),2)

            #we can filter theta as outlier based on other theta values
            #this will help in excluding the rare text region with different orientation from ususla value 
            theta = slope(box[0][0], box[0][1], box[1][0], box[1][1])
            cummTheta += theta
            ct +=1 
            #print("Theta", theta)
            
    #find the average of all cumulative theta value
    orientation = cummTheta/ct
    print("Image orientation in degress: ", orientation)
    finalImage = rotate(img, orientation)
    display(textImg, "Detectd Text minimum bounding box")
    display(finalImage, "Deskewed Image")

if __name__ == "__main__":
    filePath = 'D:\data\img6.jpg'
    main(filePath)

Here is Image with detected text regions, from this we can see that some of the text regions are missing. Text orientation detection plays the key role here in overall document orientation detection so based on document type a few small tweaks should be made in the text detection algorithm to make this approach work better.

Here is the final image with correct orientation

Please suggest modifications in this approaches to make it more robust.

Daphne

Thanks, I found a library called deskew which worked fine but I don't have any control over what it does, but using this I have more control and I can see better how it works. – Mush

Please share the deskew library which you have used, it may be usefull too. – Daphne

pip install deskew @Daphne – Ansel

Looks like the return for cv2.findContours has changed to only return contours and hierarchy. No need for the _, #54165130 – Victual

Answer 2

When a document containing several lines of text is well-aligned, a horizontal histogram of the image should produce a square-wave-like pattern that distinctly shows where the lines of text are separate from the blank spaces between them. By contrast, if the image is only slightly rotated, the horizontal histogram will be significantly blurred.

This Python script aligns an image by measuring the sharpness of the horizontal histogram over a range of angles. It compares each angle to its immediate neighbors.

import cv2
import numpy as np

# Rotates an image
def rotate_image(image: np.ndarray, angle: float) -> np.ndarray:
    mean_pixel = np.median(np.median(image, axis=0), axis=0)
    image_center = tuple(np.array(image.shape[1::-1]) / 2)
    rot_mat = cv2.getRotationMatrix2D(image_center, angle, 1.0)
    result = cv2.warpAffine(image, rot_mat, image.shape[1::-1], flags=cv2.INTER_LINEAR, borderMode=cv2.BORDER_CONSTANT, borderValue=mean_pixel)
    return result

# Returns a small value if the horizontal histogram is sharp.
# Returns a large value if the horizontal histogram is blurry.
def eval_image(image: np.ndarray) -> float:
    hist = np.sum(np.mean(image, axis=1), axis=1)
    bef = 0
    aft = 0
    err = 0.
    assert(hist.shape[0] > 0)
    for pos in range(hist.shape[0]):
        if pos == aft:
            bef = pos
            while aft + 1 < hist.shape[0] and abs(hist[aft + 1] - hist[pos]) >= abs(hist[aft] - hist[pos]):
                aft += 1
        err += min(abs(hist[bef] - hist[pos]), abs(hist[aft] - hist[pos]))
    assert(err > 0)
    return err

# Measures horizontal histogram sharpness across many angles
def sweep_angles(image: np.ndarray) -> np.ndarray:
    results = np.empty((81, 2))
    for i in range(81):
        angle = (i - results.shape[0] // 2) / 4.
        rotated = rotate_image(image, angle)
        err = eval_image(rotated)
        results[i, 0] = angle
        results[i, 1] = err
    return results

# Find an angle that is a lot better than its neighbors
def find_alignment_angle(image: np.ndarray) -> float:
    best_gain = 0
    best_angle = 0.
    results = sweep_angles(image)
    for i in range(2, results.shape[0] - 2):
        ave = np.mean(results[i-2:i+3, 1])
        gain = ave - results[i, 1]
        # print('angle=' + str(results[i, 0]) + ', gain=' + str(gain))
        if gain > best_gain:
            best_gain = gain
            best_angle = results[i, 0]
    return best_angle

# input: an image that needs aligning
# output: the aligned image
def align_image(image: np.ndarray) -> np.ndarray:
    angle = find_alignment_angle(image)
    return rotate_image(image, angle)

# Do it
fixme: np.ndarray = cv2.imread('fixme.png')
cv2.imwrite('fixed.png', align_image(fixme))

Libreville

Answer 3

This isn't really an answer as much as a suggestion for possible approach for images with mostly horizontal/vertical lines: try rotating the image every (say) 0.5 degrees, and for each rotation, sum all the scanlines across (resulting in a 1d array of sums, of size ydim, for each rotation value). Then look at the statistics of the summed scanlines, and find the rotation value that maximizes the spread (max - min). In other words, the "highest contrast" for the summed scanlines. That should be the best orientation.

For speed, you could start with every 2 degrees using a half-res image, find the best, then retry every 0.5 degrees in that neighborhood with the full-res image.

Uriah

Answer 4

Another solution that hasn't been mentioned yet is the use of the method image_to_osd from the library pytesseract. This method returns a result containing information about orientation and script detection. Documentation here

One solution using this method looks like this:

def float_convertor(x):
    if x.isdigit():
        out= float(x)
    else:
        out= x
    return out 

img = cv2.imread(image)
k = pytesseract.image_to_osd(img)
out = {i.split(":")[0]: float_convertor(i.split(":")[-1].strip()) for i in k.rstrip().split("\n")}
img_rotated = ndimage.rotate(img, 360-out["Rotate"])

(The second to last line parses the OSD information ('k') and creates a dictionary ('out') with key-value pairs. It splits the information by lines, then by colons, and uses the left part as the key and the right part as the value)

Parable

Answer 5

This is one of the most common problem which developers might encounter when dealing with Image data. Ways of image orientation can be done only if text is present in it. Hough Transform is one of the effective way of resolving this issue. However this can work only in the case where image is rotated -90 to +90 degrees.

Eccentric

As it’s currently written, your answer is unclear. Please edit to add additional details that will help others understand how this addresses the question asked. You can find more information on how to write good answers in the help center. – Confession