I am trying to make a python program that can read digits from a seven segmented red led display. I will put some examples of what I am trying to process.
import cv2
import numpy as np
def color_filter(img):
"""Filters out only red-to-white color"""
hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
lower_red = np.array([0, 0, 200])
upper_red = np.array([60, 255, 255])
mask_1 = cv2.inRange(hsv, lower_red, upper_red)
lower_red = np.array([150, 0, 200])
upper_red = np.array([255, 255, 255])
mask_2 = cv2.inRange(hsv, lower_red, upper_red)
result = 255*(mask_1 + mask_2)
result = result.clip(0, 255).astype("uint8")
return cv2.bitwise_and(img, img, mask=result)
The program needs to read the digits displayed in the LCD display like from the images above. Till now I have wrote a color filtering code. My question is now that I have filtered the text out, How do I read it. I have tried using tesseract OCR, but it simply doesn't work.
Thanks In Advance.
Related
I am trying to count seeds in an image using cv2 thresholding. The test image is below:
When I run the below code to create a mask:
import cv2
import numpy as np
import matplotlib.pyplot as plt
img = cv2.imread('S__14278933.jpg')
#img = cv2.fastNlMeansDenoisingColored(img,None,10,10,7,21)
mask = cv2.threshold(img[:, :, 0], 255, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1]
plt.imshow(mask)
I get the following mask:
But ideally it should give a small yellow dot at the centre. I have tried this with other images and it works just fine.
Can someone help?
The lighting in your image seems not uniform. Try using Adaptive Thresholding:
import cv2
import numpy as np
# image path
path = "D://opencvImages//"
fileName = "c6pBO.jpg"
# Reading an image in default mode:
inputImage = cv2.imread(path + fileName)
# Convert the image to grayscale:
grayImage = cv2.cvtColor(inputImage, cv2.COLOR_BGR2GRAY)
# Get binary image via Adaptive Thresholding :
windowSize = 31
windowConstant = 40
binaryImage = cv2.adaptiveThreshold( grayImage, 255, cv2.ADAPTIVE_THRESH_MEAN_C,
cv2.THRESH_BINARY_INV, windowSize, windowConstant )
cv2.imshow("binaryImage", binaryImage)
cv2.waitKey(0)
You might want to apply an Area Filter after this, though, as dark portions on the image will yield noise.
Try it
img=cv2.imread('foto.jpg',0)
mask = cv2.threshold(img, 127, 255, cv2.THRESH_BINARY_INV )[1]
I have various type of images like those:
As you see, they are all kinda similar, however I do not manage to properly extract the number on them.
So far my code consists in the following:
lower = np.array([250,200,90], dtype="uint8")
upper = np.array([255,204,99], dtype="uint8")
mask = cv2.inRange(img, lower, upper)
res = cv2.bitwise_and(img, img, mask=mask)
data = image_to_string(res, lang="eng", config='--psm 13 --oem 3 -c tessedit_char_whitelist=0123456789')
numbers = int(''.join(re.findall(r'\d+', data)))
I tried twearking the psm parameter 6,8 and 13 they all work for some of those examples, but none on all, and I have no idea how I could circumvent my problem.
Another solution proposed is:
gry = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
(h, w) = gry.shape[:2]
gry = cv2.resize(gry, (w*2, h*2))
erd = cv2.erode(gry, None, iterations=1)
thr = cv2.threshold(erd, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)[1]
bnt = cv2.bitwise_not(thr)
However, on the first picture, bnt gives:
And then pytesseract sees 460..
Any idea please?
My approach:
Upsample
Erosion
Simple-thresholding
Bitwise-not
Upsampling is required for accurate recognition. Resizing two-times will make the image readable.
Erosion operation is a morphological operation helps to remove the boundary of the pixels. Erosion remove the strokes on the digit, make it easier to detect.
Thresholding (Binary and Inverse Binary) helps to reveal the features.
Bitwise-not is an arithmetic operation highly useful for extracting part of the image.
You can learn more methods simple reading from Improving the quality of the output
Erosion
Threshold
Bitwise-not
Update
The first image is easy to read, since it is not requiring any pre-processing technique. Please read How to Improve Quality of Tesseract
Result:
1460
720
3250
3146
2681
1470
Code:
import cv2
import pytesseract
img_lst = ["oqWjd.png", "YZDt1.png", "MUShJ.png", "kbK4m.png", "POIK2.png", "4W3R4.png"]
for i, img_nm in enumerate(img_lst):
img = cv2.imread(img_nm)
gry = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
(h, w) = gry.shape[:2]
if i == 0:
thr = gry
else:
gry = cv2.resize(gry, (w * 2, h * 2))
erd = cv2.erode(gry, None, iterations=1)
if i == len(img_lst)-1:
thr = cv2.threshold(erd, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1]
else:
thr = cv2.threshold(erd, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)[1]
bnt = cv2.bitwise_not(thr)
txt = pytesseract.image_to_string(bnt, config="--psm 6 digits")
print("".join([t for t in txt if t.isalnum()]))
cv2.imshow("bnt", bnt)
cv2.waitKey(0)
If you want to display comma in the result, change print("".join([t for t in txt if t.isalnum()])) line to print(txt).
Not that on the fourth image the threshold method changed from binary to inverse-binary. Binary thresholding is not working accurately on all images. Therefore you need to change.
I have the following image
lower = np.array([175, 125, 45], dtype="uint8")
upper = np.array([255, 255, 255], dtype="uint8")
mask = cv2.inRange(image, lower, upper)
img = cv2.bitwise_and(image, image, mask=mask)
plt.figure()
plt.imshow(img)
plt.axis('off')
plt.show()
now if I try to transform into grayscale like this:
gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
I get that:
And I would like to extract the number on it.
The suggestion:
gray = 255 - gray
emp = np.full_like(gray, 255)
emp -= gray
emp[emp==0] = 255
emp[emp<100] = 0
gauss = cv2.GaussianBlur(emp, (3,3), 1)
gauss[gauss<220] = 0
plt.imshow(gauss)
gives the image:
Then using pytesseract on any of the images:
data = pytesseract.image_to_string(img, config='outputbase digits')
gives:
'\x0c'
Another suggested solution is:
gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
thr = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV)[1]
txt = pytesseract.image_to_string(thr)
plt.imshow(thr)
And this gives
'\x0c'
Not very satisfying... Anyone has a better solution please?
Thanks!
I have a two step solution
Apply thresholding
Set psm mode to 7.
When you apply thresholding to the image:
Thresholding is a simplest method of displaying the features of the image.
Now from the output image, when we read:
txt = image_to_string(thr, config="--psm 7")
print(txt)
Result will be:
| 1,625 |
Now why do we set page-segmentation-mode (psm) mode to the 7?
Well, treating image as a single text line will give the accurate result.
But we have to modify the result. Since the current result is | 1,625 |
We should remove the |
print("".join([t for t in txt if t != '|']))
Result:
1,625
Code:
import cv2
from pytesseract import image_to_string
img = cv2.imread("LZ3vi.png")
gry = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
thr = cv2.threshold(gry, 0, 255,
cv2.THRESH_BINARY_INV)[1]
txt = image_to_string(thr, config="--psm 7")
print("".join([t for t in txt if t != '|']).strip())
Update
how do you get this clean black and white image from my original image?
Using 3-steps
Reading the image using opencv's imread function
img = cv2.imread("LZ3vi.png")
Now we read the image in BGR fashion. (Not RGB)
Convert the image to the graysclae
gry = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
Result will be:
Apply threshold
thr = cv2.threshold(gry, 0, 255, cv2.THRESH_BINARY_INV)[1]
Result will be:
Now if you are wondering about thresholding. Read the simple-threhsolding
All my filters, grayscale... get weird colored images
The reason is, when you are displaying the image using pyplot, you need to set color-map (cmap) to gray
plt.imshow(img, cmap='gray')
You can read the other types here
Two issues blocked the pytessract from detecting your number:
The white rectangle around the number(Inverting and filling is the solution).
The Noise in the numbers shape(Gaussian Smoothing dealt with that)
The solution that AlexAlex has proposed will work perfectly if it was followed by a Gaussian filter:
output: 1,625
import numpy as np
import pytesseract
import cv2
BGR = cv2.imread('11.png')
RGB = cv2.cvtColor(BGR, cv2.COLOR_BGR2RGB)
lower = np.array([175, 125, 45], dtype="uint8")
upper = np.array([255, 255, 255], dtype="uint8")
mask = cv2.inRange(RGB, lower, upper)
img = cv2.bitwise_and(RGB, RGB, mask=mask)
gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
gray = 255 - gray
emp = np.full_like(gray, 255)
emp -= gray
emp[emp==0] = 255
emp[emp<100] = 0
gauss = cv2.GaussianBlur(emp, (3,3), 1)
gauss[gauss<220] = 0
text = pytesseract.image_to_string(gauss, config='outputbase digits')
print(text)
I need to read the highest temperature on thermographic images, as shown below:
IR_1544_INFRA.jpg
IR_1546_INFRA.jpg
IR_1560_INFRA.jpg
IR_1564_INFRA.jpg
I used the following code, this was the best result.
I also tried several other ways, such as: blur, gray scale, binarization, and others but they all failed.
import cv2
import pytesseract
pytesseract.pytesseract.tesseract_cmd = r"C:\Users\User\AppData\Local\Tesseract-OCR\tesseract.exe"
# Load image, grayscale, Otsu's threshold
entrada = cv2.imread('IR_1546_INFRA.jpg')
image = entrada[40:65, 277:319]
#image = cv2.imread('IR_1546_INFRA.jpg')
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
thresh = 255 - cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[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('thresh', thresh)
cv2.waitKey()
In the first image, I found
this
In the second image, I found this.
The imagem layout is always the same, that is, the temperature is always in the same place, so I cropped the image to isolate only the number. I would like (97.7 here, and 85.2 here).
My code needs to find from these images to always detect this temperature and generate a list indicating from highest to lowest.
What do you indicate for me to improve the assertiveness of pytesseract in the case of these images?
Note 1: When I annalyze the entire image (without cropping), it returns data that is not even present.
Note 2: In some images even with the binary number, pytesseract (image_to_string) does not return any data.
Thank you all and sorry for the typos, writing in english is still a challenge for me.
Because you have same images, you can crop the area you want and then do processing there. The processing is also simple. Change to gray, get threshold, invert, resize, and then do the OCR. You can see it in my code below. It works on all your attached images.
import cv2
import pytesseract
import os
image_path = "temperature"
for nama_file in sorted(os.listdir(image_path)):
print(nama_file)
img = cv2.imread(os.path.join(image_path, nama_file))
crop = img[43:62, 278:319]
gray = cv2.cvtColor(crop, cv2.COLOR_BGR2GRAY)
thresh = cv2.threshold(gray, 200, 255, cv2.THRESH_BINARY)[1]
thresh = cv2.bitwise_not(thresh)
double = cv2.resize(thresh, None, fx=2, fy=2)
custom_config = r'-l eng --oem 3 --psm 7 -c tessedit_char_whitelist="1234567890." '
text = pytesseract.image_to_string(double, config=custom_config)
print("detected: " + text)
cv2.imshow("img", img)
cv2.imshow("double", double)
cv2.waitKey(0)
cv2.destroyAllWindows()
I am new to image processing and python. You might've seen my amateur codes on this site in the last couple of days.
I am trying to count the number of trees using aerial images. This is my code:
from PIL import Image
import cv2
import numpy as np
from skimage import io, filters, measure
from scipy import ndimage
img = Image.open("D:\\Texture analysis\\K-2.jpg")
row, col = img.size
hsvimg = img.convert('HSV')
hsvimg.mode = 'RGB'
hsvimg.save('newImage2.jpg')
npHSI = np.asarray(hsvimg) #Convert HSI Image to np image
blur = cv2.GaussianBlur(npHSI, (45, 45), 5)
assert isinstance(blur, np.ndarray) ##############################
assert len(blur.shape) == 3 #Convert np Image to HSI Image
assert blur.shape[2] == 3 ##############################
hsiBlur = Image.fromarray(blur, 'RGB')
hsiBlur.save('hsiBlur.jpg') #Save the blurred image
## Read
img = cv2.imread("D:\\Texture analysis\\hsiBlur.jpg")
## convert to hsv
hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
#Threshold the image and segment the trees
mask = cv2.inRange(hsv, (36, 25, 25), (70, 255,255))
imask = mask>0
green = np.zeros_like(img, np.uint8)
green[imask] = img[imask]
## save
cv2.imwrite("green.png", green)
#Count the number of trees
im = io.imread('green.png', as_grey=True)
val = filters.threshold_otsu(im)
drops = ndimage.binary_fill_holes(im < val)
labels = measure.label(drops)
print(labels.max())
Original image:
HSI image with gaussian filter:
Segmented image:
The last part of the code returns 7, which is a wrong output. The value should be above 50. How can I properly count the number of green segments in the final segmented image?
EDIT
I converted green.png to binary and applied erosion with a 3x3 filter and iterated it 7 times to remove the noise.
This is what I did at the end. I followed this stackoverflow link
##save
cv2.imwrite("green.png", green)
#Convert to grayscale
gray = np.dot(green[...,:3], [0.299, 0.587, 0.114])
cv2.imwrite("grayScale.jpg", gray)
#Binarize the grayscale image
ret,bin_img = cv2.threshold(gray,127,255,cv2.THRESH_BINARY)
cv2.imwrite("bin_img.jpg", bin_img)
#Erosion to remove the noise
kernel = np.ones((3, 3),np.uint8)
erosion = cv2.erode(gray, kernel, iterations = 7)
cv2.imwrite("erosion.jpg", erosion)
#Count the number of trees
finalImage = cv2.imread('erosion.jpg')
finalImage = cv2.cvtColor(finalImage, cv2.COLOR_BGR2GRAY)
ret, thresh = cv2.threshold(finalImage, 127, 255, 1)
im2, contours, hierarchy = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
for cnt in contours:
cv2.drawContours(finalImage,[cnt],0,(0,0,255),1)
Saurav mentioned in his answer ... size of "contours" will give you the count. This print(contour.size())gives an error and print(contour) just prints a long 2D array. How can i get the size of contour?
PS. I didn't upload the grayscale, binary and eroded image because i felt that the images were already taking too much space, I can still upload them if anyone wants to.
I've found 52 trees with that script:
from PIL import Image, ImageDraw, ImageFont
image = Image.open('04uX3.jpg')
pixels = image.load()
size = image.size
draw = ImageDraw.Draw(image)
font = ImageFont.truetype('arial', 60)
i = 1
for x in range(0, size[0], 100):
for y in range(0, size[1], 100):
if pixels[x, y][1] > 200:
draw.text((x, y), str(i), (255, 0, 0), font=font)
i += 1
image.save('result.png')
You can see that some trees weren't detected and some non-trees were detected. So this is very rough calculation: