I am currently working on a project, where the problem statement is to detect handwritten text from a image of a particular form. As a pre-processing step I have extracted texts in the form of bounding boxes, and I have around 1500 images of texts extracted from the image form, out of which 50 of them are handwritten.
The problem is how do I now use these extracted images to train a classifier model which will classify the images as printed or handwritten text. I have no prior knowledge of Deep learning. Any help will be appreciated. I am uploading the image and the extracted images, as well as the code to extract the texts from the images.
im_ns = cv.imread('~/Image processing/IMG_20180921_111952.png')
gray = cv.cvtColor(im_ns,cv.COLOR_BGR2GRAY)
blurred_g = cv.GaussianBlur(gray,(11,11),0)
ret, th1 = cv.threshold(blurred_g,127,255,cv.THRESH_BINARY)
th2 = cv.adaptiveThreshold(blurred_g,255,cv.ADAPTIVE_THRESH_MEAN_C,cv.THRESH_BINARY,11,2)
th3 = cv.adaptiveThreshold(blurred_g,255,cv.ADAPTIVE_THRESH_GAUSSIAN_C,cv.THRESH_BINARY,11,2)
##Detecting horizontal Lines and removing them
th3_di1 = th3_di.copy()
hor = int(round(th3_di1.shape[1]/30,0))
hor_struc = cv.getStructuringElement(cv.MORPH_RECT,(hor,1))
bw_hor_er = cv.erode(th3_di1,hor_struc,iterations=1)
bw_hor_di = cv.dilate(th3_di1,hor_struc,iterations=1)
for i in range(0,bw_hor_di.shape[0]):
for j in range(0,bw_hor_di.shape[1]):
if bw_hor_di[i,j] == 0:
th3_di1[i,j] = 255
else:
th3_di1[i,j] = th3_di1[i,j]
plt.figure(figsize=(20,25))
plt.imshow(th3_di1,'gray')
# perform a connected component analysis on the thresholded
# image, then initialize a mask to store only the "large"
# components
labels = measure.label(th3_di1, neighbors=4, background=255)
mask = np.zeros(th3_di1.shape, dtype="uint8")
plt.figure(figsize=(30,25))
plt.imshow(labels)
# loop over the unique components
for lab in np.unique(labels):
# if this is the background label, ignore it
if lab == 0:
continue
# otherwise, construct the label mask and count the
# number of pixels
labelMask = np.zeros(th3_di.shape, dtype="uint8")
labelMask[labels == lab] = 255
numPixels = cv.countNonZero(labelMask)
# if the number of pixels in the component is sufficiently
# large, then add it to our mask of "large blobs"
if numPixels > 8:
mask = cv.add(mask, labelMask)
plt.figure(figsize=(30,24))
plt.imshow(mask,'gray')
# find the contours in the mask, then sort them from left to
# right
cnts = cv.findContours(mask.copy(), cv.RETR_EXTERNAL,
cv.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if imutils.is_cv2() else cnts[1]
cnts = contours.sort_contours(cnts)[0]
# loop over the contours to make rectangles for the th3 image with gassian thresholding
for (i, c) in enumerate(cnts):
# draw the bright spot on the image
(x,y,w,h) = cv.boundingRect(c)
#((cX, cY), radius) = cv.minEnclosingCircle(c)
cv.rectangle(th3,(x,y),(x+w,y+h),(0,255),2)
cv.putText(th3, "",(x+w+10,y+h),0,0.3,(0,255,0))
# show the output image
cv.imshow("Image", th3)
cv.waitKey(10000)
cv.destroyAllWindows()
##Extracting the bounding boxes
idx=0
for (i, c) in enumerate(cnts):
# draw the bright spot on the image
idx += 1
x,y,w,h = cv.boundingRect(c)
roi = im_ns[y:y+h,x:x+w]
#((cX, cY), radius) = cv.minEnclosingCircle(c)
#cv.rectangle(im_ns,(x,y),(x+w,y+h),(0,255),2)
cv.imwrite(str(idx)+'.jpg',roi)
Images:
Related
I can read text from an image using OCR. However, it works line by line.
I want to now group text based on solid lines surrounding the text.
For example, consider I have below rectangle banners. I can read text line by line. Fine! Now I want to group them by Board A,B,C and hold them in some data structure, so that I can identify, which lines belong to which board. It is given that images would be diagrams like this with solid lines around each block of text.
Please guide me on the right approach.
As mentioned in the comments by Yunus, you need to crop sub-images and feed them to an OCR module individually. An additional step could be ordering of the contours.
Approach:
Obtain binary image and invert it
Find contours
Crop sub-images based on the bounding rectangle for each contour
Feed each sub-image to OCR module (I used easyocr for demonstration)
Store text for each board in a dictionary
Code:
# Libraries import
import cv2
from easyocr import Reader
reader = Reader(['en'])
img = cv2.imread('board_text.jpg',1)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# inverse binary
th = cv2.threshold(gray,127,255,cv2.THRESH_BINARY_INV+cv2.THRESH_OTSU)[1]
# find contours and sort them from left to right
contours, hierarchy = cv2.findContours(th, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)
contours = sorted(contours, key=lambda x: [cv2.boundingRect(x)[0], cv2.boundingRect(x)[1]])
#initialize dictionary
board_dictionary = {}
# iterate each contour and crop bounding box
for i, c in enumerate(cnts):
x,y,w,h = cv2.boundingRect(c)
crop_img = img[y:y+h, x:x+w]
# feed cropped image to easyOCR module
results = reader.readtext(crop_img)
# result is output per line
# create a list to append all lines in cropped image to it
board_text = []
for (bbox, text, prob) in results:
board_text.append(text)
# convert list of words to single string
board_para = ' '.join(board_text)
#print(board_para)
# store string within a dictionary
board_dictionary[str(i)] = board_para
Dictionary Output:
board_dictionary
{'0': 'Board A Board A contains Some Text, That goes Here Some spaces and then text again', '1': 'Board B Board B has some text too but sparse.', '2': 'Board €C Board C is wide and contains text with white spaces '}
Drawing each contour
img2 = img.copy()
for i, c in enumerate(cnts):
x,y,w,h = cv2.boundingRect(c)
img2 = cv2.rectangle(img2, (x, y), (x + w, y + h), (0,255,0), 3)
Note:
While working on different images make sure the ordering is correct.
Choice of OCR module is yours pytesseract and easyocr are the options I know.
This can be done by performing following steps:
Find the shapes.
Compute the shape centers.
Find the text boxes.
Compute the text boxes centers.
Associate the textboxes with shapes based on distance.
The code is as follows:
import cv2
from easyocr import Reader
import math
shape_number = 2
image = cv2.imread("./ueUco.jpg")
deep_copy = image.copy()
image_gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
ret, thresh = cv2.threshold(image_gray, 150, 255, cv2.THRESH_BINARY)
thresh = 255 - thresh
shapes, hierarchy = cv2.findContours(image=thresh, mode=cv2.RETR_EXTERNAL, method=cv2.CHAIN_APPROX_SIMPLE)
cv2.drawContours(image=deep_copy, contours=shapes, contourIdx=-1, color=(0, 255, 0), thickness=2, lineType=cv2.LINE_AA)
shape_centers = []
for shape in shapes:
row = int((shape[0][0][0] + shape[3][0][0])/2)
column = int((shape[3][0][1] + shape[2][0][1])/2)
center = (row, column, shape)
shape_centers.append(center)
# cv2.imshow('Shapes', deep_copy)
# cv2.waitKey(0)
# cv2.destroyAllWindows()
languages = ['en']
reader = Reader(languages, gpu = True)
results = reader.readtext(image)
def cleanup_text(text):
return "".join([c if ord(c) < 128 else "" for c in text]).strip()
for (bbox, text, prob) in results:
text = cleanup_text(text)
(tl, tr, br, bl) = bbox
tl = (int(tl[0]), int(tl[1]))
tr = (int(tr[0]), int(tr[1]))
br = (int(br[0]), int(br[1]))
bl = (int(bl[0]), int(bl[1]))
column = int((tl[0] + tr[0])/2)
row = int((tr[1] + br[1])/2)
center = (row, column, bbox)
distances = []
for iteration, shape_center in enumerate(shape_centers):
shape_row = shape_center[0]
shape_column = shape_center[1]
dist = int(math.dist([column, row], [shape_row, shape_column]))
distances.append(dist)
min_value = min(distances)
min_index = distances.index(min_value)
if min_index == shape_number:
cv2.rectangle(image, tl, br, (0, 255, 0), 2)
cv2.putText(image, text, (tl[0], tl[1] - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.8, (0, 255, 0), 2)
cv2.imshow("Image", image)
cv2.waitKey(0)
cv2.imwrite(f"image_{shape_number}.jpg", image)
cv2.destroyAllWindows()
The output looks like this.
Please note that this solution is almost complete. You just have to compute the text embodied in each shape and put it in your desired data structure.
Note: shape_number represents the shape that you want to consider.
There is another solution that I would like you to work on.
Find all the text boxes.
Compute the centers for text boxes.
Run k-means clustering on the centers.
I would prefer the second solution but for the time being, I implemented the first one.
I am looking to extract text from a license plate. For now I have been using pytesseract with opencv to zero in on the relevant contours and pull out text. This works decently for non-American plates, but I am curious about applying this to American plates which come with a lot of little letters surrounding the big plate id ones. My thoughts were to use font size to filter out letters under a certain threshold. Is that the best approach?
below is code so far:
import cv2
import pytesseract
import imutils
#read image
image = cv2.imread('plateTest2.jpeg')
#RGB to Gray Scale converstion
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
#noise removal
gray = cv2.bilateralFilter(gray,11,17,17)
#find edges of the grayscale image
edged = cv2.Canny(gray, 170,200)
#Find contours based on Edges
_,cnts, new = cv2.findContours(edged.copy(), cv2.RETR_LIST, cv2.CHAIN_APPROX_SIMPLE)
#Create copy of original image to draw all contours
img1 = image.copy()
cv2.drawContours(img1, cnts, -1, (0,255,0), 3)
#sort contours based on their area keeping minimum required area as '30' (anything smaller than this will not be considered)
cnts=sorted(cnts, key = cv2.contourArea, reverse = True)[:30]
NumberPlateCnt = None #we currently have no Number plate contour
#Top 30 Contours
img2 = image.copy()
cv2.drawContours(img2, cnts, -1, (0,255,0), 3)
idx='plateTest2.jpg'
for c in cnts:
peri = cv2.arcLength(c, True)
approx = cv2.approxPolyDP(c, 0.02 * peri, True)
# print ("approx = ",approx)
if len(approx) == 4: # Select the contour with 4 corners
NumberPlateCnt = approx #This is our approx Number Plate Contour
# Crop those contours and store it in Cropped Images folder
x, y, w, h = cv2.boundingRect(c) #This will find out co-ord for plate
new_img = gray[y:y + h, x:x + w] #Create new image
cv2.imwrite('/' + 'cropped_' + str(idx), new_img) #Store new image
#idx+=1
break
#Drawing the selected contour on the original image
#print(NumberPlateCnt)
cv2.drawContours(image, [NumberPlateCnt], -1, (0,255,0), 3)
Cropped_img_loc = '/' + 'cropped_' + str(idx)#'cropped_images/8.png'
#Use tesseract to covert image into string
text = pytesseract.image_to_string(Cropped_img_loc, lang='eng')
text = text.replace('.', '')
text = text.replace(' ','')
return text
Here is a picture of a plate that returns too much text where things like 'SUNSHINESTATE' show up:
Should I rely on pytesseract to identify font size and filter out smaller characters? Or should I be filtering before using contour size? Appreciate the help.
I want to detect the seat belt is fasten or not. I have used the below step.
color segmentation
image bluring
edge detection
morphological transform
hough problabstic
angular filtering
Right now i am not able to extract the feature from from image which is returned by morphological transform. I am attaching the output which is returned by morpgological transform. Any help will be appreciated.
img = cv2.imread('belt1.png')
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
#linek = np.zeros((11,11),dtype=np.uint8)
#linek[5,...]=1
#x=cv2.morphologyEx(gray, cv2.MORPH_OPEN, linek ,iterations=1)
#gray-=x
#kernel = np.ones((5, 5), np.uint8)
#gray = cv2.dilate(gray, kernel, iterations=1)
#gray = cv2.erode(gray, kernel, iterations=1)
kernel_size = 5
blur_gray = cv2.GaussianBlur(gray,(kernel_size, kernel_size),0)
low_threshold = 50
high_threshold = 150
kernel = np.ones((5,5),np.uint8)
edges = cv2.Canny(blur_gray, low_threshold, high_threshold)
edges = cv2.morphologyEx(edges, cv2.MORPH_CLOSE, kernel)
rho = 1 # distance resolution in pixels of the Hough grid
theta = np.pi / 180 # angular resolution in radians of the Hough grid
threshold = 80 # minimum number of votes (intersections in Hough grid cell)
min_line_length = 100 # minimum number of pixels making up a line
max_line_gap = 40 # maximum gap in pixels between connectable line segments
line_image = np.copy(img) * 0 # creating a blank to draw lines on
# Run Hough on edge detected image
# Output "lines" is an array containing endpoints of detected line segments
lines = cv2.HoughLinesP(edges, rho, theta, threshold, np.array([]),
min_line_length, max_line_gap)
here i am not able to extract feature:
I would like to create a program that is able to extract lines from a graph.
For example, if a graph like this is inputted, I would just want the red line to be outputted.
Below I have tried to do this using a hough line transformation, however, I do not get very promising results.
import cv2
import numpy as np
graph_img = cv2.imread("/Users/2020shatgiskessell/Desktop/Graph1.png")
gray = cv2.cvtColor(graph_img, cv2.COLOR_BGR2GRAY)
kernel_size = 5
#grayscale image
blur_gray = cv2.GaussianBlur(gray,(kernel_size, kernel_size),0)
#Canny edge detecion
edges = cv2.Canny(blur_gray, 50, 150)
#Hough Lines Transformation
#distance resoltion of hough grid (pixels)
rho = 1
#angular resolution of hough grid (radians)
theta = np.pi/180
#minimum number of votes
threshold = 15
#play around with these
min_line_length = 25
max_line_gap = 20
#make new image
line_image = np.copy(graph_img)
#returns array of lines
lines = cv2.HoughLinesP(edges, rho, theta, threshold, np.array([]),
min_line_length, max_line_gap)
for line in lines:
for x1,y1,x2,y2 in line:
cv2.line(line_image,(x1,y1),(x2,y2),(255,0,0),2)
lines_edges = cv2.addWeighted(graph_img, 0.8, line_image, 1, 0)
cv2.imshow("denoised image",edges)
if cv2.waitKey(0) & 0xff == 27:
cv2.destroyAllWindows()
This produces the output image below, which does not accurately recognize the graph line. How might I go about doing this?
Note: For now, I am not concerned about the graph titles or any other text.
I would also like the code to work for other graph images aswell, such as:
etc.
If the graph does not have many noises around it (like your example) I would suggest to threshold your image with Otsu threshold instead of looking for edges . Then you simply search the contours, select the biggest one (graph) and draw it on a blank mask. After that you can perform a bitwise operation on image with the mask and you will get a black image with the graph. If you like the white background better, then simply change all black pixels to white. Steps are written in the example. Hope it helps a bit. Cheers!
Example:
import numpy as np
import cv2
# Read the image and create a blank mask
img = cv2.imread('graph.png')
h,w = img.shape[:2]
mask = np.zeros((h,w), np.uint8)
# Transform to gray colorspace and threshold the image
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
_, thresh = cv2.threshold(gray,0,255,cv2.THRESH_BINARY_INV+cv2.THRESH_OTSU)
# Search for contours and select the biggest one and draw it on mask
_, contours, hierarchy = cv2.findContours(thresh,cv2.RETR_TREE,cv2.CHAIN_APPROX_NONE)
cnt = max(contours, key=cv2.contourArea)
cv2.drawContours(mask, [cnt], 0, 255, -1)
# Perform a bitwise operation
res = cv2.bitwise_and(img, img, mask=mask)
# Convert black pixels back to white
black = np.where(res==0)
res[black[0], black[1], :] = [255, 255, 255]
# Display the image
cv2.imshow('img', res)
cv2.waitKey(0)
cv2.destroyAllWindows()
Result:
EDIT:
For noisier pictures you could try this code. Note that different graphs have different noises and may not work on every graph image since the denoisiation process would be specific in every case. For different noises you can use different ways to denoise it, for example histogram equalization, eroding, blurring etc. This code works well for all 3 graphs. Steps are written in comments. Hope it helps. Cheers!
import numpy as np
import cv2
# Read the image and create a blank mask
img = cv2.imread('graph.png')
h,w = img.shape[:2]
mask = np.zeros((h,w), np.uint8)
# Transform to gray colorspace and threshold the image
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
_, thresh = cv2.threshold(gray,0,255,cv2.THRESH_BINARY_INV+cv2.THRESH_OTSU)
# Perform opening on the thresholded image (erosion followed by dilation)
kernel = np.ones((2,2),np.uint8)
opening = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel)
# Search for contours and select the biggest one and draw it on mask
_, contours, hierarchy = cv2.findContours(opening,cv2.RETR_TREE,cv2.CHAIN_APPROX_NONE)
cnt = max(contours, key=cv2.contourArea)
cv2.drawContours(mask, [cnt], 0, 255, -1)
# Perform a bitwise operation
res = cv2.bitwise_and(img, img, mask=mask)
# Threshold the image again
gray = cv2.cvtColor(res,cv2.COLOR_BGR2GRAY)
_, thresh = cv2.threshold(gray,0,255,cv2.THRESH_BINARY_INV+cv2.THRESH_OTSU)
# Find all non white pixels
non_zero = cv2.findNonZero(thresh)
# Transform all other pixels in non_white to white
for i in range(0, len(non_zero)):
first_x = non_zero[i][0][0]
first_y = non_zero[i][0][1]
first = res[first_y, first_x]
res[first_y, first_x] = 255
# Display the image
cv2.imshow('img', res)
cv2.waitKey(0)
cv2.destroyAllWindows()
Result:
I am trying to design an app similar to camscanner. For that, I have to take an image and then find the document in that. I started off with the code described here - http://opencvpython.blogspot.in/2012/06/sudoku-solver-part-2.html
I found the contours and the rectangular contour with max area should be the required document. For every contour, I am finding an approximate closed PolyDP. Of all the polyDP of size 4, the one with max area should be the required document. However, this method is not working.
The input image for the process is this
I tried to print the contour with max area and this resulted in this (Contour inside letter 'C')
Code:
img = cv2.imread('bounce.jpeg')
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
gray = cv2.GaussianBlur(gray,(5,5),0)
thresh = cv2.adaptiveThreshold(gray,255,1,1,11,2)
_, contours, hierarchy = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
def biggestRectangle(contours):
biggest = None
max_area = 0
indexReturn = -1
for index in range(len(contours)):
i = contours[index]
area = cv2.contourArea(i)
if area > 100:
peri = cv2.arcLength(i,True)
approx = cv2.approxPolyDP(i,0.1*peri,True)
if area > max_area: #and len(approx)==4:
biggest = approx
max_area = area
indexReturn = index
return indexReturn
indexReturn = biggestRectangle(contours)
cv2.imwrite('hola.png',cv2.drawContours(img, contours, indexReturn, (0,255,0)))
What is going wrong in this? Is there any other method by which I can capture the document in this picture?
Try this :
output image
import cv2
import numpy as np
img = cv2.imread('bounce.jpg')
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
invGamma = 1.0 / 0.3
table = np.array([((i / 255.0) ** invGamma) * 255
for i in np.arange(0, 256)]).astype("uint8")
# apply gamma correction using the lookup table
gray = cv2.LUT(gray, table)
ret,thresh1 = cv2.threshold(gray,80,255,cv2.THRESH_BINARY)
#thresh = cv2.adaptiveThreshold(gray,255,1,1,11,2)
_, contours, hierarchy = cv2.findContours(thresh1, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
def biggestRectangle(contours):
biggest = None
max_area = 0
indexReturn = -1
for index in range(len(contours)):
i = contours[index]
area = cv2.contourArea(i)
if area > 100:
peri = cv2.arcLength(i,True)
approx = cv2.approxPolyDP(i,0.1*peri,True)
if area > max_area: #and len(approx)==4:
biggest = approx
max_area = area
indexReturn = index
return indexReturn
indexReturn = biggestRectangle(contours)
hull = cv2.convexHull(contours[indexReturn])
cv2.imwrite('hola.png',cv2.drawContours(img, [hull], 0, (0,255,0),3))
#cv2.imwrite('hola.png',thresh1)
I would do it like this:
Do preprocessing like blur / canny
Extract all lines from the image using the hough line transform (open cv doc).
Use the 4 strongest lines
Try to construct the contour of the document using the four lines
Right now I do not have an OpenCV installed so I cannot try this approach but maybe it leads you in the right directon.