We Keep Coding

Not getting a countour box around the texts/tags on an X-ray image with OpenCV

I am trying to detect texts and remove those thick letters. The goal is same proposed by the following link: How to detect text on an X-Ray image with OpenCV. ("to extract the oriented bounding boxes as a matrix")
Two sample images:
Sample Image #1
Sample Image #2
Here, the fonts L, J, C, O (from 1st image) and L, D, A, N, circle-shape (from 2nd image) must be detected. I tried the methods proposed by above link (How to detect text on an X-Ray image with OpenCV), however, it fails to detect the texts in the background.
Original Image --> Binary image
Binary Image (thresholding)
Morph close --> Detected text
Morph close
detected texts (nothing)
As you can see, it fails to detect the texts from the background. It just returns black image. Don't know what happened.
The code:
import cv2
import numpy as np
# Load image, create mask, grayscale, Gaussian blur, Otsu's threshold
image = cv2.imread('HandImage.png')
original = image.copy()
blank = np.zeros(image.shape[:2], dtype=np.uint8)
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
blur = cv2.GaussianBlur(gray, (5,5), 0)
thresh = cv2.threshold(blur, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)[1]
# Merge text into a single contour
kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (5,5))
close = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel, iterations=3)
# Find contours
cnts = cv2.findContours(close, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if len(cnts) == 2 else cnts[1]
for c in cnts:
# Filter using contour area and aspect ratio
x,y,w,h = cv2.boundingRect(c)
area = cv2.contourArea(c)
ar = w / float(h)
if (ar > 1.4 and ar < 4) or ar < .85 and area > 10 and area < 500:
# Find rotated bounding box
rect = cv2.minAreaRect(c)
box = cv2.boxPoints(rect)
box = np.int0(box)
cv2.drawContours(image,[box],0,(36,255,12),2)
cv2.drawContours(blank,[box],0,(255,255,255),-1)
# Bitwise operations to isolate text
extract = cv2.bitwise_and(thresh, blank)
extract = cv2.bitwise_and(original, original, mask=extract)
cv2.imshow('thresh', thresh)
cv2.imshow('image', image)
cv2.imshow('close', close)
cv2.imshow('extract', extract)
cv2.waitKey()
I am new to the computer vision and image manipulation... Please help!
Update:
I have figured out the error in my code. I had to adjust and identify the right size/area of the bounding box that selected for the texts.
The code:
# Find contours
cnts = cv2.findContours(dilate, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if len(cnts) == 2 else cnts[1]
cnts = sorted(cnts, key = lambda x: cv2.boundingRect(x)[0])
for c in cnts:
# Filter using contour area and aspect ratio (x1 = width, y1 = height)
x, y, x1, y1 = cv2.boundingRect(c)
area = cv2.contourArea(c)
if (30 < x1 < 200 or 350 < x1 < 500) and 50 < y1 < 350:
# Using the area above to calculate the rotation
rect = cv2.minAreaRect(c)
box = cv2.boxPoints(rect)
translated_box = box - np.mean(box, axis=0)
scaled_box = translated_box * 2 # 2 is scale factor
retranslated_box = scaled_box + np.mean(box, axis=0)
box = np.int0(retranslated_box)
cv2.drawContours(image, [box], 0, (36,255,12), 2)
cv2.drawContours(blank, [box], 0, (255,255,255), -1)
But I also liked the suggestion by #Mikel B, where he used a neural network method.

Not a solution in OpenCV as requested. But I leave it here in case a neural network-based solution could be of interest for anybody
First of all install the package by:
pip install craft-text-detector
Then copy paste the following
from craft_text_detector import Craft
# import craft functions
from craft_text_detector import (
read_image,
load_craftnet_model,
load_refinenet_model,
get_prediction,
export_detected_regions,
export_extra_results,
empty_cuda_cache
)
# set image path and export folder directory
image = '/path/to/JP6x0.png' # can be filepath, PIL image or numpy array
output_dir = 'outputs/'
# read image
image = read_image(image)
# load models
refine_net = load_refinenet_model(cuda=False)
craft_net = load_craftnet_model(cuda=False)
# perform prediction
prediction_result = get_prediction(
image=image,
craft_net=craft_net,
refine_net=refine_net,
text_threshold=0.5,
link_threshold=0.1,
low_text=0.1,
cuda=False,
long_size=1280
)
# export detected text regions
exported_file_paths = export_detected_regions(
image=image,
regions=prediction_result["boxes"],
output_dir=output_dir,
rectify=True
)
# export heatmap, detection points, box visualization
export_extra_results(
image=image,
regions=prediction_result["boxes"],
heatmaps=prediction_result["heatmaps"],
output_dir=output_dir
)
# unload models from gpu
empty_cuda_cache()
My results with these parameters:

Text line segmentation using OpenCV Python

I am trying to extract the text lines from the below figure and I tried the below code but I am getting only a single image without any data. While masking the images all the lines are masked perfectly. Below i attached the masked image, final output image and desired output image.
img = cv.imread('Handwritten_data/Ostu_images/H_1.jpg')
lower = (0, 0, 0)
upper = (0, 120, 150)
# threshold on border color
mask = cv.inRange(img, lower, upper)
# dilate threshold
kernel = cv.getStructuringElement(cv.MORPH_RECT, (250,10))
mask = cv.morphologyEx(mask, cv.MORPH_DILATE, kernel)
# recolor border to white
img[mask==255] = (255,255,255)
# Inverting the mask by
# performing bitwise-not operation
mask_black = cv.bitwise_not(mask)
Mask = cv.bitwise_and(img, img, mask = mask_black)
gray = cv.cvtColor(img, cv.COLOR_BGR2GRAY)
# otsu threshold
thresh = cv.threshold(gray, 0, 255, cv.THRESH_OTSU )[1]
# apply morphology open
kernel = cv.getStructuringElement(cv.MORPH_RECT, (250,10))
morph = cv.morphologyEx(thresh, cv.MORPH_OPEN, kernel)
# creating a folder
try:
# creating a folder named data
if not os.path.exists('Image_0'):
os.makedirs('Image_0')
# if not created then raise error
except OSError:
print ('Error: Creating directory of data')
# find contours and bounding boxes
bboxes = []
bboxes_img = img.copy()
contours = cv.findContours(morph, cv.RETR_EXTERNAL, cv.CHAIN_APPROX_SIMPLE)
contours = contours[0] if len(contours) == 2 else contours[1]
for cntr in contours:
x,y,w,h = cv.boundingRect(cntr)
cv.rectangle(bboxes_img, (x, y), (x+w, y+h), (0,0,255), 1)
bboxes.append((x,y,w,h))
for j in range(len(bboxes)):
(x,y,w,h) = bboxes[j]
crop = img[y-10:y+h+10, x-10:x+w+10]
cv.imwrite(f'Image_0/S_{j}.jpg', crop)
Any suggestions or help to solve this problem.
Below is the masking image
final image ouput
desired image output be like
Thanks in advance

The idea you presented for text line segmentation is correct. But the approach needs some tweaking.
Note:
Whenever you want to mask a portion of the image, make sure the masked region is in white. Because when finding contours the algorithm looks for white regions. Since you were looking for contours that were in black the algorithm missed it.
The following is modified with the above idea.
Solution:
img = cv2.imread(image_file)
img_gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
ret, thresh2 = cv2.threshold(img_gray, 150, 255, cv2.THRESH_BINARY_INV)
kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (150,2))
mask = cv2.morphologyEx(thresh2, cv2.MORPH_DILATE, kernel)
Below is the mask image:
bboxes = []
bboxes_img = img.copy()
contours = cv2.findContours(mask, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
contours = contours[0] if len(contours) == 2 else contours[1]
for cntr in contours:
x,y,w,h = cv2.boundingRect(cntr)
cv2.rectangle(bboxes_img, (x, y), (x+w, y+h), (0,0,255), 1)
bboxes.append((x,y,w,h))
The following is the final result with bounding boxes:
You can now add your portion of the code to save each region as a separate image file. You can try modifying the kernel parameters and exploring different morphological operations to get a better mask region.
Hope this puts you on the right track!

detect an initial/a sketch drawing on a text page

I would like to get the coordinates of the box around the initial ("H") on the following page (and similar ones with other initials, so opencv template matching is not an option):
Following this tutorial, I tried to solve the problem with opencv contours:
import cv2
import matplotlib.pyplot as plt
page = "image.jpg"
# read the image
image = cv2.imread(page)
# convert to RGB
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
# convert to grayscale
gray = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)
# create a binary thresholded image
_, binary = cv2.threshold(gray, 0,150,cv2.THRESH_BINARY+cv2.THRESH_OTSU)
# find the contours from the thresholded image
contours, hierarchy = cv2.findContours(binary, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
# draw all contours
image = cv2.drawContours(image, contours, 3, (0, 255, 0), 2)
plt.savefig("result.png")
The result is of course not exactly what I wanted:
Does anyone know of an viable algorithm (and possibly an implementation thereof) that could provide an easy solution to my task?

You can find the target area by filtering your contours. Now, there's at least two filtering criteria that you can use. One is filter by area - that is, discard too small and too large contours until you get the contour you are looking for. The other one is by computing the extent of every contour. The extent is the ratio of the contour's area to its bounding rectangle area. You are looking for a square-like contour, so its extent should be close to 1.0.
Let's see the code:
# imports:
import cv2
import numpy as np
# Reading an image in default mode:
inputImage = cv2.imread(path + fileName)
# Deep copy for results:
inputImageCopy = inputImage.copy()
# Convert RGB to grayscale:
grayscaleImage = cv2.cvtColor(inputImage, cv2.COLOR_BGR2GRAY)
# Get binary image via Otsu:
_, binaryImage = cv2.threshold(grayscaleImage, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)
The first portion of the code gets you a binary image that you can use as a mask to compute contours:
Now, let's filter contours. Let's use the area approach first. You need to define a range of minimum area and maximum area to filter everything that does not fall in this range. I've heuristically determined a range of areas from 30000 px to 150000 px:
# Find the contours on the binary image:
contours, hierarchy = cv2.findContours(binaryImage, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
# Look for the outer bounding boxes (no children):
for _, c in enumerate(contours):
# Get blob area:
currentArea = cv2.contourArea(c)
print("Contour Area: "+str(currentArea))
# Set an area range:
minArea = 30000
maxArea = 150000
if minArea < currentArea < maxArea:
# Get the contour's bounding rectangle:
boundRect = cv2.boundingRect(c)
# Get the dimensions of the bounding rect:
rectX = boundRect[0]
rectY = boundRect[1]
rectWidth = boundRect[2]
rectHeight = boundRect[3]
# Set bounding rect:
color = (0, 0, 255)
cv2.rectangle( inputImageCopy, (int(rectX), int(rectY)),
(int(rectX + rectWidth), int(rectY + rectHeight)), color, 2 )
cv2.imshow("Rectangles", inputImageCopy)
cv2.waitKey(0)
Once you successfully filter the area, you can then compute the bounding rectangle of the contour with cv2.boundingRect. You can retrieve the bounding rectangle's x, y (top left) coordinates as well as its width and height. After that just draw the rectangle on a deep copy of the original input.
Now, let's see the second option, using the contour's extent. The for loop gets modified as follows:
# Look for the outer bounding boxes (no children):
for _, c in enumerate(contours):
# Get blob area:
currentArea = cv2.contourArea(c)
# Get the contour's bounding rectangle:
boundRect = cv2.boundingRect(c)
# Get the dimensions of the bounding rect:
rectX = boundRect[0]
rectY = boundRect[1]
rectWidth = boundRect[2]
rectHeight = boundRect[3]
# Calculate extent:
extent = float(currentArea)/(rectWidth *rectHeight)
print("Extent: " + str(extent))
# Set the extent filter, look for an extent close to 1.0:
delta = abs(1.0 - extent)
epsilon = 0.1
if delta < epsilon:
# Set bounding rect:
color = (0, 0, 255)
cv2.rectangle( inputImageCopy, (int(rectX), int(rectY)),
(int(rectX + rectWidth), int(rectY + rectHeight)), color, 2 )
cv2.imshow("Rectangles", inputImageCopy)
cv2.waitKey(0)
Both approaches yield this result:

You almost have it. You just need to filter contours on area and aspect ratio. Here is my approach in Python/OpenCV.
Input:
import cv2
import numpy as np
# read image as grayscale
img = cv2.imread('syriados.jpg')
# convert to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# threshold to binary
#thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY)[1]
thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY+cv2.THRESH_OTSU)[1]
# invert threshold
thresh = 255 - thresh
# apply morphology to remove small white regions and to close the rectangle boundary
kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (3,3))
morph = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel)
kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (7,7))
morph = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel)
# find contours
result = img.copy()
cntrs = cv2.findContours(morph, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
cntrs = cntrs[0] if len(cntrs) == 2 else cntrs[1]
# filter on area and aspect ratio
for c in cntrs:
area = cv2.contourArea(c)
x,y,w,h = cv2.boundingRect(c)
if area > 10000 and abs(w-h) < 100:
cv2.drawContours(result, [c], 0, (0,0,255), 2)
# write results
cv2.imwrite("syriados_thresh.jpg", thresh)
cv2.imwrite("syriados_morph.jpg", morph)
cv2.imwrite("syriados_box.jpg", result)
# show results
cv2.imshow("thresh", thresh)
cv2.imshow("morph", morph)
cv2.imshow("result", result)
cv2.waitKey(0)
Threshold image:
Morphology image:
Resulting contour image:

To get a result like this:
You'll need to detect the contour in the image with the second to the greatest area, as the one possessing the greatest area would be the border of the image.
So with the list of contours, we can get the one with the second greatest area via the built-in sorted method, using the cv2.contourArea method as the custom key:
import cv2
import numpy as np
def process(img):
img_gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
img_blur = cv2.GaussianBlur(img_gray, (7, 7), 2)
img_canny = cv2.Canny(img_blur, 50, 50)
kernel = np.ones((6, 6))
img_dilate = cv2.dilate(img_canny, kernel, iterations=1)
img_erode = cv2.erode(img_dilate, kernel, iterations=2)
return img_erode
def get_contours(img):
contours, _ = cv2.findContours(process(img), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)
cnt = sorted(contours, key=cv2.contourArea)[-2]
peri = cv2.arcLength(cnt, True)
approx = cv2.approxPolyDP(cnt, 0.02 * peri, True)
cv2.drawContours(img, [approx], -1, (0, 255, 0), 2)
page = "image.jpg"
image = cv2.imread(page)
get_contours(image)
cv2.imshow("Image", image)
cv2.waitKey(0)
The above only puts the area of the contours into consideration; if you want more reliable results, you can make it so that it will only detect contours that are 4-sided.

How to rotate an image to align the text for extraction?

I am using pytessearct to extract the text from images. But it doesn't work on images which are inclined. Consider the image given below:
Here is the code to extract text, which is working fine on images which are not inclined.
img = cv2.imread(<path_to_image>)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
blur = cv2.GaussianBlur(gray, (5,5),0)
ret3, thresh = cv2.threshold(blur,0,255,cv2.THRESH_BINARY+cv2.THRESH_OTSU)
def findSignificantContours (img, edgeImg):
contours, heirarchy = cv2.findContours(edgeImg, cv2.RETR_TREE, cv2.CHAIN_APPROX_NONE)
# Find level 1 contours
level1 = []
for i, tupl in enumerate(heirarchy[0]):
# Each array is in format (Next, Prev, First child, Parent)
# Filter the ones without parent
if tupl[3] == -1:
tupl = np.insert(tupl, 0, [i])
level1.append(tupl)
significant = []
tooSmall = edgeImg.size * 5 / 100 # If contour isn't covering 5% of total area of image then it probably is too small
for tupl in level1:
contour = contours[tupl[0]];
area = cv2.contourArea(contour)
if area > tooSmall:
significant.append([contour, area])
# Draw the contour on the original image
cv2.drawContours(img, [contour], 0, (0,255,0),2, cv2.LINE_AA, maxLevel=1)
significant.sort(key=lambda x: x[1])
#print ([x[1] for x in significant]);
mx = (0,0,0,0) # biggest bounding box so far
mx_area = 0
for cont in contours:
x,y,w,h = cv2.boundingRect(cont)
area = w*h
if area > mx_area:
mx = x,y,w,h
mx_area = area
x,y,w,h = mx
# Output to files
roi = img[y:y+h,x:x+w]
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
blur = cv2.GaussianBlur(gray, (5,5),0)
ret3, thresh = cv2.threshold(blur,0,255,cv2.THRESH_BINARY+cv2.THRESH_OTSU)
cv2_imshow(thresh)
text = pytesseract.image_to_string(roi);
print(text); print("\n"); print(pytesseract.image_to_string(thresh));
print("\n")
return [x[0] for x in significant];
edgeImg_8u = np.asarray(thresh, np.uint8)
# Find contours
significant = findSignificantContours(img, edgeImg_8u)
mask = thresh.copy()
mask[mask > 0] = 0
cv2.fillPoly(mask, significant, 255)
# Invert mask
mask = np.logical_not(mask)
#Finally remove the background
img[mask] = 0;
Tesseract can't extract the text from this image. Is there a way I can rotate it to align the text perfectly and then feed it to pytesseract? Please let me know if my question require any more clarity.

Here's a simple approach:
Obtain binary image. Load image, convert to grayscale,
Gaussian blur, then Otsu's threshold.
Find contours and sort for largest contour. We find contours then filter using contour area with cv2.contourArea() to isolate the rectangular contour.
Perform perspective transform. Next we perform contour approximation with cv2.contourArea() to obtain the rectangular contour. Finally we utilize imutils.perspective.four_point_transform to actually obtain the bird's eye view of the image.
Binary image
Result
To actually extract the text, take a look at
Use pytesseract OCR to recognize text from an image
Cleaning image for OCR
Detect text area in an image using python and opencv
Code
from imutils.perspective import four_point_transform
import cv2
import numpy
# Load image, grayscale, Gaussian blur, Otsu's threshold
image = cv2.imread("1.jpg")
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
blur = cv2.GaussianBlur(gray, (7,7), 0)
thresh = cv2.threshold(blur, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)[1]
# Find contours and sort for largest contour
cnts = cv2.findContours(thresh, cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if len(cnts) == 2 else cnts[1]
cnts = sorted(cnts, key=cv2.contourArea, reverse=True)
displayCnt = None
for c in cnts:
# Perform contour approximation
peri = cv2.arcLength(c, True)
approx = cv2.approxPolyDP(c, 0.02 * peri, True)
if len(approx) == 4:
displayCnt = approx
break
# Obtain birds' eye view of image
warped = four_point_transform(image, displayCnt.reshape(4, 2))
cv2.imshow("thresh", thresh)
cv2.imshow("warped", warped)
cv2.waitKey()

To Solve this problem you can also use minAreaRect api in opencv which will give you a minimum area rotated rectangle with an angle of rotation. You can then get the rotation matrix and apply warpAffine for the image to straighten it. I have also attached a colab notebook which you can play around on.
Colab notebook : https://colab.research.google.com/drive/1SKxrWJBOHhGjEgbR2ALKxl-dD1sXIf4h?usp=sharing
import cv2
from google.colab.patches import cv2_imshow
import numpy as np
def rotate_image(image, angle):
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)
return result
img = cv2.imread("/content/sxJzw.jpg")
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
mask = np.zeros((img.shape[0], img.shape[1]))
blur = cv2.GaussianBlur(gray, (5,5),0)
ret, thresh = cv2.threshold(blur,0,255,cv2.THRESH_BINARY+cv2.THRESH_OTSU)
cv2_imshow(thresh)
contours, _ = cv2.findContours(thresh, cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE)
largest_countour = max(contours, key = cv2.contourArea)
binary_mask = cv2.drawContours(mask, [largest_countour], 0, 1, -1)
new_img = img * np.dstack((binary_mask, binary_mask, binary_mask))
minRect = cv2.minAreaRect(largest_countour)
rotate_angle = minRect[-1] if minRect[-1] < 0 else -minRect[-1]
new_img = rotate_image(new_img, rotate_angle)
cv2_imshow(new_img)

Enhancing corner detection of lamp with OpenCV

I'm using the following code to detect the brightly illuminated lamp. The illumination might vary. I'm using the following code to detect the same.
img = cv2.imread("input_img.jpg")
rgb = img.copy()
img_grey = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
while True:
th3 = cv2.adaptiveThreshold(img_grey, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, \
cv2.THRESH_BINARY, 11, 2)
cv2.imshow("th3",th3)
edged = cv2.Canny(th3, 50, 100)
edged = cv2.dilate(edged, None, iterations=1)
edged = cv2.erode(edged, None, iterations=1)
cv2.imshow("edge", edged)
cnts = cv2.findContours(edged.copy(), cv2.RETR_TREE,
cv2.CHAIN_APPROX_SIMPLE)
cnts = imutils.grab_contours(cnts)
areaArray = []
for i, c in enumerate(cnts):
area = cv2.contourArea(c)
areaArray.append(area)
sorteddata = sorted(zip(areaArray, cnts), key=lambda x: x[0], reverse=True)
thirdlargestcontour = sorteddata[2][1]
x, y, w, h = cv2.boundingRect(thirdlargestcontour)
cv2.drawContours(rgb, thirdlargestcontour, -1, (255, 0, 0), 2)
cv2.rectangle(rgb, (x, y), (x + w, y + h), (0, 255, 0), 2)
cv2.imshow("rgb", rgb)
if cv2.waitKey(1) == 27:
break
The above code works but,
It only gives the rectangle that encompasses the lamp. How do I get the four corner points of the lamp precisely?
How can I improve detection? at the moment I'm picking the third-largest contour which does not guarantee that it will always be the lamp as the environment poses challenge?
ApproxPolydp works when the contour is complete but if the contour is incomplete, ApproxPolydp is not returning the proper coordinate. for instance in the following image the approxpolydp returns a wrong coordinates.

Here is one way to do that in Python/OpenCV.
Read the input image and convert to grayscale
Use adaptive thresholding to get a thick outline of the lamp region
Find the contours
Filter the contours on area to remove extraneous regions and keep only the larger of the two (inner and outer contours of thresholded region)
Get the perimeter
Fit the perimeter to a polygon, which should be a quadrilateral with the right choice of arguments.
Draw the contour (red) and polygon (blue) over a copy of the input image as the result
Input:
import cv2
import numpy as np
# load image
img = cv2.imread("lamp.jpg")
# convert to gray
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# threshold image
thresh = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 11, 10)
thresh = 255 - thresh
# find contours
cntrs = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
cntrs = cntrs[0] if len(cntrs) == 2 else cntrs[1]
# Contour filtering -- remove small objects and those that are too large
# Keep the larger of the two contours (inner and outer contours from thresh)
area_thresh = 0
for c in cntrs:
area = cv2.contourArea(c)
if area > 200 and area > area_thresh:
big_contour = c
area_thresh = area
# draw big_contour on image in red and polygon in blue and print corners
results = img.copy()
cv2.drawContours(results,[big_contour],0,(0,0,255),1)
peri = cv2.arcLength(big_contour, True)
corners = cv2.approxPolyDP(big_contour, 0.04 * peri, True)
cv2.drawContours(results,[corners],0,(255,0,0),1)
print(len(corners))
print(corners)
# write result to disk
cv2.imwrite("lamp_thresh.jpg", thresh)
cv2.imwrite("lamp_corners.jpg", results)
cv2.imshow("THRESH", thresh)
cv2.imshow("RESULTS", results)
cv2.waitKey(0)
cv2.destroyAllWindows()
Thresholded Image:
Result Image:
Corner Coordinates:
[[[233 145]]
[[219 346]]
[[542 348]]
[[508 153]]]