I am working on a video with many people where few of them are wearing red colored t-shirt. I have all the persons detected and tracked with person detection and tracking models. How can I distinguish the persons wearing red from the others.
I am reading the frames in OpenCV format. If I know the coordinates, suppose x,y is a coordinate of the body where the color is red. How can I get the color information from the coordinate in OpenCV format and check whether that comes under the red color range?
I only need to highlight the bounding box of the persons wearing red from others.
Can someone help me in figuring out a solution.
Thank you!
The better way to change the colour space into HSV and find the Hue value range for colour.
Take each frame of the video
Detect humans first then extract the human region (source)
Convert from BGR to HSV color-space
Threshold the HSV image for a range of red colour
Identifying red colour t-shirt guys in Video
We can identify the human region in images using the following code
import time
import cv2
import imutils
import numpy as np
from imutils.video import FPS
# import the necessary packages
from imutils.video import VideoStream
def get_centered_contours(mask):
# find contours
cntrs = cv2.findContours(mask, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cntrs = cntrs[0] if len(cntrs) == 2 else cntrs[1]
sorted_contours = sorted(cntrs, key=cv2.contourArea, reverse=True)
filterd_contours = []
if sorted_contours != []:
for k in range(len(sorted_contours)):
if cv2.contourArea(sorted_contours[k]) < 1000.0:
filterd_contours = sorted_contours[0:k]
return filterd_contours
return filterd_contours
def check_red_colour_person(roi):
hsv = cv2.cvtColor(roi, cv2.COLOR_BGR2HSV)
# define range of blue color in HSV
lower_red = np.array([0, 50, 50])
upper_red = np.array([10, 255, 255])
# Threshold the HSV image to get only blue colors
mask = cv2.inRange(hsv, lower_red, upper_red)
cnts = get_centered_contours(mask)
if cnts != []:
return True
else:
return False
# construct the argument parse and parse the arguments
prototxt = 'MobileNetSSD_deploy.prototxt.txt'
model = 'MobileNetSSD_deploy.caffemodel'
confidence_level = 0.8
# initialize the list of class labels MobileNet SSD was trained to
# detect, then generate a set of bounding box colors for each class
CLASSES = ["person"]
COLORS = np.random.uniform(0, 255, size=(len(CLASSES), 3))
# load our serialized model from disk
print("[INFO] loading model...")
net = cv2.dnn.readNetFromCaffe(prototxt, model)
# initialize the video stream, allow the cammera sensor to warmup,
# and initialize the FPS counter
print("[INFO] starting video stream...")
vs = VideoStream(src=0).start()
time.sleep(2.0)
fps = FPS().start()
# loop over the frames from the video stream
while True:
try:
# grab the frame from the threaded video stream and resize it
# to have a maximum width of 400 pixels
frame = vs.read()
frame = imutils.resize(frame, width=400)
# grab the frame dimensions and convert it to a blob
(h, w) = frame.shape[:2]
blob = cv2.dnn.blobFromImage(cv2.resize(frame, (300, 300)),
0.007843, (300, 300), 127.5)
# pass the blob through the network and obtain the detections and
# predictions
net.setInput(blob)
detections = net.forward()
# loop over the detections
for i in np.arange(0, detections.shape[2]):
# extract the confidence (i.e., probability) associated with
# the prediction
confidence = detections[0, 0, i, 2]
# filter out weak detections by ensuring the `confidence` is
# greater than the minimum confidence
if confidence > confidence_level:
# extract the index of the class label from the
# `detections`, then compute the (x, y)-coordinates of
# the bounding box for the object
idx = int(detections[0, 0, i, 1])
box = detections[0, 0, i, 3:7] * np.array([w, h, w, h])
(startX, startY, endX, endY) = box.astype("int")
roi = frame[startY:endY, startX:endX]
# cv2.imwrite('roi_{}_{}_{}_{}.png'.format(startX,startY,endX,endY),roi)
if check_red_colour_person(roi):
label = "{}: {:.2f}%".format(' Red T-shirt person',
confidence * 100)
cv2.imwrite(
'Red-T-shirt_guy_{}_{}_{}_{}.png'.format(startX, startY, endX,
endY), roi)
cv2.rectangle(frame, (startX, startY), (endX, endY),
(0, 0, 255), 2)
else:
cv2.rectangle(frame, (startX, startY), (endX, endY),
(255, 0, 0), 2)
y = startY - 15 if startY - 15 > 15 else startY + 15
cv2.putText(frame, label, (startX, y),
cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 0), 2)
cv2.imshow("Frame", frame)
key = cv2.waitKey(1) & 0xFF
# if the `q` key was pressed, break from the loop
if key == ord("q"):
break
# update the FPS counter
fps.update()
except Exception as e:
print("Exception is occured")
continue
# stop the timer and display FPS information
fps.stop()
print("[INFO] elapsed time: {:.2f}".format(fps.elapsed()))
print("[INFO] approx. FPS: {:.2f}".format(fps.fps()))
# do a bit of cleanup
cv2.destroyAllWindows()
vs.stop()
You can set the color boundaries
boundaries = [
([17, 15, 100], [50, 56, 200])]
So here tuple ([17, 15, 100], [50, 56, 200]) .
Here, we are saying that all pixels in our image that have a R >= 100, B >= 15, and G >= 17 along with R <= 200, B <= 56, and G <= 50 will be considered red.
You can implement like follow:
for (lower, upper) in boundaries:
lower = np.array(lower, dtype = "uint8")
upper = np.array(upper, dtype = "uint8")
# find the colors within the specified boundaries and apply
# the mask
mask = cv2.inRange(image, lower, upper)
output = cv2.bitwise_and(image, image, mask = mask)
# show the images
cv2.imshow("images", np.hstack([image, output]))
Related
I want to implement simple traffic light detection algorithm in Python with help of OpenCV. Of course if we want to get high accuracy, we should use some pre-trained deep learning models, but now I want just simplest and not comprehensive approach. Namely I know that traffic lights are (green, red , yellow), therefore I found following link which contains code about three color detection.
I know once again, that it is not an accurate method, just self learning. I have tested this code on one video. Here is a cropped frame from the corresponding video:
After running my code, I got following image:
As you can see, the lowest part of the image is ignored and upper part is considered.
How can I adapt or change my code such that it should check whole image and detect actual traffic lights as well?
Should I resize the image to a lower resolution?
Should I use some other approach?
import numpy as np
import cv2
import warnings
warnings.filterwarnings("ignore")
# Capturing video through webcam
live_video = cv2.VideoCapture("traffic_light.mp4")
# Start a while loop
while (1):
# Reading the video from the
# webcam in image frames
_, imageFrame = live_video .read()
# Convert the imageFrame in
# BGR(RGB color space) to
# HSV(hue-saturation-value)
# color space
hsvFrame = cv2.cvtColor(imageFrame, cv2.COLOR_BGR2HSV)
# Set range for red color and
# define mask
red_lower = np.array([136, 87, 111], np.uint8)
red_upper = np.array([180, 255, 255], np.uint8)
red_mask = cv2.inRange(hsvFrame, red_lower, red_upper)
# Set range for green color and
# define mask
green_lower = np.array([25, 52, 72], np.uint8)
green_upper = np.array([102, 255, 255], np.uint8)
green_mask = cv2.inRange(hsvFrame, green_lower, green_upper)
# Set range for blue color and
# define mask
blue_lower = np.array([94, 80, 2], np.uint8)
blue_upper = np.array([120, 255, 255], np.uint8)
blue_mask = cv2.inRange(hsvFrame, blue_lower, blue_upper)
# Morphological Transform, Dilation
# for each color and bitwise_and operator
# between imageFrame and mask determines
# to detect only that particular color
kernal = np.ones((5, 5), "uint8")
# For red color
red_mask = cv2.dilate(red_mask, kernal)
res_red = cv2.bitwise_and(imageFrame, imageFrame,
mask=red_mask)
# For green color
green_mask = cv2.dilate(green_mask, kernal)
res_green = cv2.bitwise_and(imageFrame, imageFrame,
mask=green_mask)
# For blue color
blue_mask = cv2.dilate(blue_mask, kernal)
res_blue = cv2.bitwise_and(imageFrame, imageFrame,
mask=blue_mask)
# Creating contour to track red color
contours, hierarchy = cv2.findContours(red_mask,
cv2.RETR_TREE,
cv2.CHAIN_APPROX_SIMPLE)
for pic, contour in enumerate(contours):
area = cv2.contourArea(contour)
if (area > 300):
x, y, w, h = cv2.boundingRect(contour)
imageFrame = cv2.rectangle(imageFrame, (x, y),
(x + w, y + h),
(0, 0, 255), 2)
cv2.putText(imageFrame, "Red Colour", (x, y),
cv2.FONT_HERSHEY_SIMPLEX, 1.0,
(0, 0, 255))
# Creating contour to track green color
contours, hierarchy = cv2.findContours(green_mask,
cv2.RETR_TREE,
cv2.CHAIN_APPROX_SIMPLE)
for pic, contour in enumerate(contours):
area = cv2.contourArea(contour)
if (area > 300):
x, y, w, h = cv2.boundingRect(contour)
imageFrame = cv2.rectangle(imageFrame, (x, y),
(x + w, y + h),
(0, 255, 0), 2)
cv2.putText(imageFrame, "Green Colour", (x, y),
cv2.FONT_HERSHEY_SIMPLEX,
1.0, (0, 255, 0))
# Creating contour to track blue color
contours, hierarchy = cv2.findContours(blue_mask,
cv2.RETR_TREE,
cv2.CHAIN_APPROX_SIMPLE)
for pic, contour in enumerate(contours):
area = cv2.contourArea(contour)
if (area > 300):
x, y, w, h = cv2.boundingRect(contour)
imageFrame = cv2.rectangle(imageFrame, (x, y),
(x + w, y + h),
(255, 0, 0), 2)
cv2.putText(imageFrame, "Blue Colour", (x, y),
cv2.FONT_HERSHEY_SIMPLEX,
1.0, (255, 0, 0))
# Program Termination
cv2.imshow("Multiple Color Detection in Real-TIme", imageFrame)
if cv2.waitKey(10) & 0xFF == ord('q'):
live_video .release()
cv2.destroyAllWindows()
break
I am thinking about resizing of frames, but first I would like to listen your opinions. The main tricky part, I think, is located in the following code, since it tries to locate cursor in the specific contours determined by coordinates. How can I change it?
I am trying to detect the filling levels of bottles moving on a conveyor belt with opencv. Since the bottles are colored, I give a white light from the back and determine the liquid contours and measure. I'm getting an error in a certain part of the code. Mistake;
(contours, areas) = zip(*sorted(zip(contours, areas), key = lambda a:a[1])) ValueError: not enouhg values to unpack (expected 2, got 0)
For ex photograph
When I increase the threshold value in the code, the program runs. But I want this value to remain constant. The reason for this is related to the quality of the contours. Thank you from now.
Source Code:
_, frame = cap.read()
# hsv_frame = cv2.cvtColor(frame, cv2.COLOR.BGR2HSV)
bottle_gray = cv2.split(frame)[0]
bottle_gray = cv2.GaussianBlur(bottle_gray, (7,7), 0)
(T, bottle_threshold) = cv2.threshold(bottle_gray, 49.5, 250, cv2.THRESH_BINARY_INV)
kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (5,5))
bottle_open = cv2.morphologyEx(bottle_threshold, cv2.MORPH_CLOSE, kernel)
contours = cv2.findContours(bottle_open.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
contours = imutils.grab_contours(contours)
bottle_clone = frame.copy()
cv2.drawContours(bottle_clone, contours, 0, (255,0,0), 2)
areas = [cv2.contourArea(contour) for contour in contours]
(contours, areas) = zip(*sorted(zip(contours,areas),key = lambda a:a[1]))```
import sys
import cv2
import numpy as np
# Fix HSV color range
def fixHSVRange(h, s, v):
return (180 * h / 360, 255 * s / 100, 255 * v / 100)
# Load image
dir = sys.path[0]
im = cv2.imread(dir+'/im.png')
# Convert image to HSV and find empty bottle range
hsv = cv2.cvtColor(im, cv2.COLOR_BGR2HSV)
msk = cv2.inRange(hsv, fixHSVRange(0, 50, 60), fixHSVRange(5, 100, 100))
# Smooth noise
msk = cv2.medianBlur(msk, 21)
# Invert colors
msk = ~msk
# Find empty space bounderies
cnts, _ = cv2.findContours(msk, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
x, y, w, h = cv2.boundingRect(cnts[1])
# Draw level line
msk = cv2.cvtColor(msk, cv2.COLOR_GRAY2BGR)
cv2.line(msk, (0, y+h), (im.shape[1], y+h), (0, 255, 0), 2, cv2.LINE_AA)
# Save output
cv2.imwrite(dir+'/im_msk.png', np.hstack((im, msk)))
Need your help. Now I'm writing python script to recognize text in a shape. This shape can be captured from RTSP (IP Camera) at any angle.
For the example see attached file. My code is here, but coords to crop rotated shape is sets manually
import cv2
import numpy as np
def main():
fn = cv2.VideoCapture("rtsp://admin:Admin123-#172.16.10.254")
flag, img = fn.read()
cnt = np.array([
[[64, 49]],
[[122, 11]],
[[391, 326]],
[[308, 373]]
])
print("shape of cnt: {}".format(cnt.shape))
rect = cv2.minAreaRect(cnt)
print("rect: {}".format(rect))
box = cv2.boxPoints(rect)
box = np.int0(box)
print("bounding box: {}".format(box))
cv2.drawContours(img, [box], 0, (0, 255, 0), 2)
img_crop, img_rot = crop_rect(img, rect)
print("size of original img: {}".format(img.shape))
print("size of rotated img: {}".format(img_rot.shape))
print("size of cropped img: {}".format(img_crop.shape))
new_size = (int(img_rot.shape[1]/2), int(img_rot.shape[0]/2))
img_rot_resized = cv2.resize(img_rot, new_size)
new_size = (int(img.shape[1]/2)), int(img.shape[0]/2)
img_resized = cv2.resize(img, new_size)
cv2.imshow("original contour", img_resized)
cv2.imshow("rotated image", img_rot_resized)
cv2.imshow("cropped_box", img_crop)
# cv2.imwrite("crop_img1.jpg", img_crop)
cv2.waitKey(0)
def crop_rect(img, rect):
# get the parameter of the small rectangle
center = rect[0]
size = rect[1]
angle = rect[2]
center, size = tuple(map(int, center)), tuple(map(int, size))
# get row and col num in img
height, width = img.shape[0], img.shape[1]
print("width: {}, height: {}".format(width, height))
M = cv2.getRotationMatrix2D(center, angle, 1)
img_rot = cv2.warpAffine(img, M, (width, height))
img_crop = cv2.getRectSubPix(img_rot, size, center)
return img_crop, img_rot
if __name__ == "__main__":
main()
example pic
You may start with the example in the following post.
The code sample detects the license plate, and it also detects your "shape" with text.
After detecting the "shape" with the text, you may use the following stages:
Apply threshold the cropped area.
Find contours, and find the contour with maximum area.
Build a mask, and mask area outside the contour (like in the license plate example).
Use minAreaRect (as fmw42 commented), and get the angle of the rectangle.
Rotate the cropped area (by angle+90 degrees).
Apply OCR using pytesseract.image_to_string.
Here is the complete code:
import cv2
import numpy as np
import imutils
import pytesseract
pytesseract.pytesseract.tesseract_cmd = r'C:\Program Files\Tesseract-OCR\tesseract.exe' # I am using Windows
# Read the input image
img = cv2.imread('Admin123.jpg')
# Reused code:
# https://stackoverflow.com/questions/60977964/pytesseract-not-recognizing-text-as-expected/60979089#60979089
################################################################################
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) #convert to grey scale
gray = cv2.bilateralFilter(gray, 11, 17, 17)
edged = cv2.Canny(gray, 30, 200) #Perform Edge detection
cnts = cv2.findContours(edged.copy(), cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
cnts = imutils.grab_contours(cnts)
cnts = sorted(cnts, key = cv2.contourArea, reverse = True)[:10]
screenCnt = None
# loop over our contours
for c in cnts:
# approximate the contour
peri = cv2.arcLength(c, True)
approx = cv2.approxPolyDP(c, 0.018 * peri, True)
# if our approximated contour has four points, then
# we can assume that we have found our screen
if len(approx) == 4:
screenCnt = approx
break
# Masking the part other than the "shape"
mask = np.zeros(gray.shape,np.uint8)
new_image = cv2.drawContours(mask,[screenCnt],0,255,-1,)
new_image = cv2.bitwise_and(img,img,mask=mask)
# Now crop
(x, y) = np.where(mask == 255)
(topx, topy) = (np.min(x), np.min(y))
(bottomx, bottomy) = (np.max(x), np.max(y))
cropped = gray[topx:bottomx+1, topy:bottomy+1]
################################################################################
# Apply threshold the cropped area
_, thresh = cv2.threshold(cropped, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)
# Find contours
cnts = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)
cnts = imutils.grab_contours(cnts)
# Get contour with maximum area
c = max(cnts, key=cv2.contourArea)
# Build a mask (same as the code above)
mask = np.zeros(cropped.shape, np.uint8)
new_cropped = cv2.drawContours(mask, [c], 0, 255, -1)
new_cropped = cv2.bitwise_and(cropped, cropped, mask=mask)
# Draw green rectangle for testing
test = cv2.cvtColor(new_cropped, cv2.COLOR_GRAY2BGR)
cv2.drawContours(test, [c], -1, (0, 255, 0), thickness=2)
# Use minAreaRect as fmw42 commented
rect = cv2.minAreaRect(c)
angle = rect[2] # Get angle of the rectangle
# Rotate the cropped rectangle.
rotated_cropped = imutils.rotate(new_cropped, angle + 90)
# Read the text in the "shape"
text = pytesseract.image_to_string(rotated_cropped, config='--psm 3')
print("Extracted text is:\n\n", text)
# Show images for testing:
cv2.imshow('cropped', cropped)
cv2.imshow('thresh', thresh)
cv2.imshow('test', test)
cv2.imshow('rotated_cropped', rotated_cropped)
cv2.waitKey(0)
cv2.destroyAllWindows()
OCR output result:
AB12345
DEPARTMENT OF
INFORMATION
COMMUNICATION
TECHNOLOGY
cropped:
thresh:
test:
rotated_cropped:
I have been working with OpenCV in order to detect an squared obstacle. So far this is the image I get after applying filters and canny.
The obstacle I am trying to identify is the horizontal one, the three vertical rectangles are guide lines on the floor.My goal is to keep only the horizontal rectangle, separating it from the others, but after applying find Contours I only get I single object that includes all the shapes.This is the code I have been using in order to fin only the biggest rectangle by their area:
# find the biggest countour (c) by the area
if contours != 0:
if not contours:
print("Empty")
else:
bigone = max(contours, key=cv2.contourArea) if max else None
area = cv2.contourArea(bigone)
if area > 10000:
x, y, w, h = cv2.boundingRect(bigone)
cv2.rectangle(img, (x, y), (x+w, y+h), (0, 0, 255), 2)
cv2.putText(img, "Obstacle", (x+w/2, y-20),
cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 255, 255), 2)
pts = np.array(
[[[x, y], [x+w, y], [x+w, y+h], [x, y+h]]], dtype=np.int32)
cv2.fillPoly(mask, pts, (255, 255, 255))
#values = img[np.where((mask == (255, 255, 255)).all(axis=2))]
res = cv2.bitwise_and(img, mask) # View only the obstacle
obs_area = w*h
print(obs_area)
if obs_area <= 168000:
command_publisher.publish("GO")
cv2.putText(
img, "GO", (380, 400), cv2.FONT_HERSHEY_SIMPLEX, 2, (0, 255, 255), 1)
else:
command_publisher.publish("STOP")
cv2.putText(img, "STOP", (380, 400),
cv2.FONT_HERSHEY_SIMPLEX, 2, (0, 255, 255), 1)
# show the output image
cv2.imshow("Image", img)
cv2.waitKey(1)
And this is the result I am getting:
Is there a way of separating my obstacle from the lines on the floor with some kind of filter or algorithm?
Here is an example image to work with:
Here is one way to do that using Python/OpenCV.
- Read the input
- Convert to HSV and extract only the saturation channel (black/white/gray have zero saturation)
- Threshold
- Apply morphology open and close to remove the extranous white regions
- Get the contour and approximate to simple polygon
- Draw the polygon on the input
- Save the results
Input:
import cv2
import numpy as np
# read image
img = cv2.imread('board.png')
# convert to HSV and extract saturation channel
sat = cv2.cvtColor(img, cv2.COLOR_RGB2HSV)[:,:,1]
# threshold
thresh = cv2.threshold(sat, 90, 255, 0)[1]
# apply morphology close to fill interior regions in mask
kernel = np.ones((7,7), np.uint8)
morph = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel)
kernel = np.ones((13,13), np.uint8)
morph = cv2.morphologyEx(morph, cv2.MORPH_CLOSE, kernel)
# get contours (presumably only 1) and fit to simple polygon (quadrilateral)
cntrs = cv2.findContours(morph, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cntrs = cntrs[0] if len(cntrs) == 2 else cntrs[1]
c = cntrs[0]
peri = cv2.arcLength(c, True)
approx = cv2.approxPolyDP(c, 0.05 * peri, True)
# draw polygon on input
result = img.copy()
cv2.polylines(result, [np.int32(approx)], True, (0,0,255), 1, cv2.LINE_AA)
# write result to disk
cv2.imwrite("board_saturation.png", sat)
cv2.imwrite("board_thresh.png", thresh)
cv2.imwrite("board_morph.png", morph)
cv2.imwrite("board_contour.png", result)
# display it
cv2.imshow("IMAGE", img)
cv2.imshow("SAT", sat)
cv2.imshow("THRESH", thresh)
cv2.imshow("MORPH", morph)
cv2.imshow("RESULT", result)
cv2.waitKey(0)
Saturation channel image:
Thresholded image:
Morphology cleaned image:
Contour on input:
In your image the problem seems white rectangles. My approach is checking each line and if line consist many pixels which are close to white(255,255,255) then make the line black.
Here is my code:
import cv2
import numpy as np
import random as rng
img=cv2.imread("/ur/image/directory/obstacle.png")
height, width, channels = img.shape
cv2.imshow('Source',img)
# Check each line and eliminate white rectangles(if line consist white pixels more than limit)
for x in range(0,height):
white_counter = 0
for y in range(0,width):
if img[x,y,0] >= 180 and img[x,y,1] >= 180 and img[x,y,2] >= 180:
white_counter = white_counter + 1
if white_counter>10:
for y in range(0,width):
img[x,y,0] = 0
img[x,y,1] = 0
img[x,y,2] = 0
cv2.imshow('Elimination White Rectangles', img)
# Find contours and draw rectangle for each
src_gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
threshold = 300
canny_output = cv2.Canny(src_gray, threshold, threshold * 2)
contours, _ = cv2.findContours(canny_output, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
contours_poly = [None]*len(contours)
boundRect = [None]*len(contours)
for i, c in enumerate(contours):
contours_poly[i] = cv2.approxPolyDP(c, 3, True)
boundRect[i] = cv2.boundingRect(contours_poly[i])
rng.seed(12345)
drawing = np.zeros((canny_output.shape[0], canny_output.shape[1], 3), dtype=np.uint8)
for i in range(len(contours)):
color = (rng.randint(0,256), rng.randint(0,256), rng.randint(0,256))
cv2.rectangle(drawing, (int(boundRect[i][0]), int(boundRect[i][1])), \
(int(boundRect[i][0]+boundRect[i][2]), int(boundRect[i][1]+boundRect[i][3])), color, 2)
cv2.imshow('Output', drawing)
cv2.waitKey(0)
cv2.destroyAllWindows()
Eliminate White Rectangles:
Result:
I am trying to detect the outer boundary of the circular object in the images below:
I tried OpenCV's Hough Circle, but the code is not working for every image. I also tried to adjust parameters such as minRadius and maxRadius in Hough Circle but its not working on every image.
The aim is to detect the object from the image and crop it.
Expected output:
Source code:
import imutils
import cv2
import numpy as np
from matplotlib import pyplot as plt
image = cv2.imread("path to the image i have provided")
r = 600.0 / image.shape[1]
dim = (600, int(image.shape[0] * r))
resized = cv2.resize(image, dim, interpolation = cv2.INTER_AREA)
cv2.imwrite("path to were we want to save downscaled image", resized)
image = cv2.imread('path of downscaled image')
image1 = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
image2 = cv2.GaussianBlur(image1, (5, 5), 0)
edged = cv2.Canny(image2, 30, 150)
img = cv2.medianBlur(image2,5)
cimg = cv2.cvtColor(img,cv2.COLOR_GRAY2BGR)
circles = cv2.HoughCircles(edged,cv2.HOUGH_GRADIENT,1,20,
param1=50,param2=30,minRadius=200,maxRadius=280)
circles = np.uint16(np.around(circles))
max_circle = max(circles[0,:], key=lambda x:x[2])
# print(max_circle)
# # Create mask
height,width = image1.shape
mask = np.zeros((height,width), np.uint8)
for i in [max_circle]:
cv2.circle(mask,(i[0],i[1]),i[2],(255,255,255),thickness=-1)
masked_data = cv2.bitwise_and(image, image, mask=mask)
_,thresh = cv2.threshold(mask,1,255,cv2.THRESH_BINARY)
# Find Contour
contours = cv2.findContours(thresh,cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE)[0]
x,y,w,h = cv2.boundingRect(contours[0])
# Crop masked_data
crop = masked_data[y:y+h,x:x+w]
#Code to close Window
cv2.imshow('OG',image)
cv2.imshow('Cropped ROI',crop)
cv2.imwrite("path to save roi image", crop)
cv2.waitKey(0)
cv2.destroyAllWindows()
Second Answer: an approach based on color segmentation.
While I was editing the question to improve it's readability and was inserting and resizing all the images from the link you shared to make it easier for everyone to visualize what you are trying to do, it occurred to me that this problem might be a better candidate for an approach based on segmentation by color:
This simpler (but clever) approach assumes that the reel appears pretty much in the same location and has more or less the same dimensions every time:
To discover the approximate color of the reel in the image, define a list of Regions of Interest (ROIs) to sample pixels from and determine the min and max color of that area in the HSV color space. The location and size of the ROI are values derived from the size of the image. In the images below, you can see the ROIs as draw as blue-ish rectangles:
Once the min and max HSV colors have been found, a threshold operation with cv2.inRange() can be executed to segment the reel:
Then, iterate though all the contours in the binary image and assume that the largest one represents the reel. Use this contour and draw it in a separate mask to be able to extract the pixels from original image:
At this stage, it is also possible to compute a bounding box for the contour and extract it's precise location to be able to perform a crop operation later and completely isolate the reel in the image:
This approach works for EVERY image shared on the question.
Source code:
import cv2
import numpy as np
import sys
# initialize global H, S, V values
min_global_h = 179
min_global_s = 255
min_global_v = 255
max_global_h = 0
max_global_s = 0
max_global_v = 0
# load input image from the cmd-line
filename = sys.argv[1]
img = cv2.imread(sys.argv[1])
if (img is None):
print('!!! Failed imread')
sys.exit(-1)
# create an auxiliary image for debugging purposes
dbg_img = img.copy()
# initiailize a list of Regions of Interest that need to be scanned to identify good HSV values to threhsold by color
w = img.shape[1]
h = img.shape[0]
roi_w = int(w * 0.10)
roi_h = int(h * 0.10)
roi_list = []
roi_list.append( (int(w*0.25), int(h*0.15), roi_w, roi_h) )
roi_list.append( (int(w*0.25), int(h*0.60), roi_w, roi_h) )
# convert image to HSV color space
hsv_img = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
# iterate through the ROIs to determine the min/max HSV color of the reel
for rect in roi_list:
x, y, w, h = rect
x2 = x + w
y2 = y + h
print('ROI rect=', rect)
cropped_hsv_img = hsv_img[y:y+h, x:x+w]
h, s, v = cv2.split(cropped_hsv_img)
min_h = np.min(h)
min_s = np.min(s)
min_v = np.min(v)
if (min_h < min_global_h):
min_global_h = min_h
if (min_s < min_global_s):
min_global_s = min_s
if (min_v < min_global_v):
min_global_v = min_v
max_h = np.max(h)
max_s = np.max(s)
max_v = np.max(v)
if (max_h > max_global_h):
max_global_h = max_h
if (max_s > max_global_s):
max_global_s = max_s
if (max_v > max_global_v):
max_global_v = max_v
# debug: draw ROI in original image
cv2.rectangle(dbg_img, (x, y), (x2, y2), (255,165,0), 4) # red
cv2.imshow('ROIs', cv2.resize(dbg_img, dsize=(0, 0), fx=0.5, fy=0.5))
#cv2.waitKey(0)
cv2.imwrite(filename[:-4] + '_rois.png', dbg_img)
# define min/max color for threshold
low_hsv = np.array([min_h, min_s, min_v])
max_hsv = np.array([max_h, max_s, max_v])
#print('low_hsv=', low_hsv)
#print('max_hsv=', max_hsv)
# threshold image by color
img_bin = cv2.inRange(hsv_img, low_hsv, max_hsv)
cv2.imshow('binary', cv2.resize(img_bin, dsize=(0, 0), fx=0.5, fy=0.5))
cv2.imwrite(filename[:-4] + '_binary.png', img_bin)
#cv2.imshow('img_bin', cv2.resize(img_bin, dsize=(0, 0), fx=0.5, fy=0.5))
#cv2.waitKey(0)
# create a mask to store the contour of the reel (hopefully)
mask = np.zeros((img_bin.shape[0], img_bin.shape[1]), np.uint8)
crop_x, crop_y, crop_w, crop_h = (0, 0, 0, 0)
# iterate throw all the contours in the binary image:
# assume that the first contour with an area larger than 100k belongs to the reel
contours, hierarchy = cv2.findContours(img_bin, cv2.RETR_LIST, cv2.CHAIN_APPROX_SIMPLE)
for contourIdx, cnt in enumerate(contours):
area = cv2.contourArea(contours[contourIdx])
print('contourIdx=', contourIdx, 'area=', area)
# draw potential reel blob on the mask (in white)
if (area > 100000):
crop_x, crop_y, crop_w, crop_h = cv2.boundingRect(cnt)
centers, radius = cv2.minEnclosingCircle(cnt)
cv2.circle(mask, (int(centers[0]), int(centers[1])), int(radius), (255), -1) # fill with white
break
cv2.imshow('mask', cv2.resize(mask, dsize=(0, 0), fx=0.5, fy=0.5))
cv2.imwrite(filename[:-4] + '_mask.png', mask)
# copy just the reel area into its own image
reel_img = cv2.bitwise_and(img, img, mask=mask)
cv2.imshow('reel_img', cv2.resize(reel_img, dsize=(0, 0), fx=0.5, fy=0.5))
cv2.imwrite(filename[:-4] + '_reel.png', reel_img)
# crop the reel to a smaller image
if (crop_w != 0 and crop_h != 0):
cropped_reel_img = reel_img[crop_y:crop_y+crop_h, crop_x:crop_x+crop_w]
cv2.imshow('cropped_reel_img', cv2.resize(cropped_reel_img, dsize=(0, 0), fx=0.5, fy=0.5))
output_filename = filename[:-4] + '_crop.png'
cv2.imwrite(output_filename, cropped_reel_img)
cv2.waitKey(0)
First answer: an approach based on pre-processing the image and executing an adaptiveThreshold operation.
There might be other ways of solving this problem that are not based on Hough Circles. Here is the result of an approach that is not:
Preprocess the image! Decreasing the size of the image and executing a blur helps with segmentation:
The segmentation method uses a cv2.adaptiveThreshold() to create a binary image that preserves the most important objects: the center of the reel and the external edge of the reel. This is an important step since we are only interested in what exists between these two objects. However, life is not perfect and neither is this segmentation. The shadow of reel on the table became part of the binary objects detected. Also, the outer edge is not fully connected as you can see on the resulting image on the right (look at the top left of the circumference):
To join broken segments, a morphological operation can be executed:
Finally, the entire reel area can be exposed by iterating through the contours of the image above and discarding those whose area is larger than what is expected for a reel. The resulting binary image (on the left) can then be used as a mask to identify the reel location on the original image:
Keep in mind that I'm not trying to find an universal solution for your problem. I'm merely showing that there might be other solutions that don't depend on Hough Circles.
Also, this code might need some adjustments to work on a larger number of cases.
Source code:
import cv2
import numpy as np
import sys
img = cv2.imread("test_images/reel.jpg")
if (img is None):
print('!!! Failed imread')
sys.exit(-1)
# create output image
output_img = img.copy()
# 1. Preprocess the image: downscale to speed up processing and execute a blur
SCALE_FACTOR = 0.5
smaller_img = cv2.resize(img, dsize=(0, 0), fx=SCALE_FACTOR, fy=SCALE_FACTOR)
blur_img = cv2.medianBlur(smaller_img, 9)
cv2.imwrite('reel1_blur_img.png', blur_img)
# 2. Segment the image to identify the 2 most important contours: the center of the reel and the outter edge
gray_img = cv2.cvtColor(blur_img, cv2.COLOR_BGR2GRAY)
img_bin = cv2.adaptiveThreshold(gray_img, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY_INV, 19, 4)
cv2.imwrite('reel2_img_bin.png', img_bin)
green_mask = np.zeros((img_bin.shape[0], img_bin.shape[1]), np.uint8)
#green_mask = cv2.cvtColor(img_bin, cv2.COLOR_GRAY2RGB) # debug
contours, hierarchy = cv2.findContours(img_bin, cv2.RETR_LIST, cv2.CHAIN_APPROX_SIMPLE)
for contourIdx, cnt in enumerate(contours):
x, y, w, h = cv2.boundingRect(cnt)
area = cv2.contourArea(contours[contourIdx])
#print('contourIdx=', contourIdx, 'w=', w, 'h=', h, 'area=', area)
# filter out tiny segments
if (area < 5000):
#cv2.fillPoly(green_mask, pts=[cnt], color=(0, 0, 255)) # red
continue
# draw green contour (filled)
#cv2.fillPoly(green_mask, pts=[cnt], color=(0, 255, 0)) # green
cv2.fillPoly(green_mask, pts=[cnt], color=(255)) # white
# debug:
#cv2.imshow('green_mask', green_mask)
#cv2.waitKey(0)
cv2.imshow('green_mask', green_mask)
cv2.imwrite('reel2_green_mask.png', green_mask)
# 3. Fix mask: join segments nearby
kernel = np.ones((3,3), np.uint8)
img_dilation = cv2.dilate(green_mask, kernel, iterations=1)
green_mask = cv2.erode(img_dilation, kernel, iterations=1)
cv2.imshow('fixed green_mask', green_mask)
cv2.imwrite('reel3_img.png', green_mask)
# 4. Extract the reel area from the green mask
reel_mask = np.zeros((green_mask.shape[0], green_mask.shape[1]), np.uint8)
#reel_mask = cv2.cvtColor(green_mask, cv2.COLOR_GRAY2RGB) # debug
contours, hierarchy = cv2.findContours(green_mask, cv2.RETR_LIST, cv2.CHAIN_APPROX_SIMPLE)
for contourIdx, cnt in enumerate(contours):
x, y, w, h = cv2.boundingRect(cnt)
area = cv2.contourArea(contours[contourIdx])
print('contourIdx=', contourIdx, 'w=', w, 'h=', h, 'area=', area)
# filter out smaller segments
if (area > 110000):
#cv2.fillPoly(reel_mask, pts=[cnt], color=(0, 0, 255)) # red
continue
# draw green contour (filled)
#cv2.fillPoly(reel_mask, pts=[cnt], color=(0, 255, 0)) # green
cv2.fillPoly(reel_mask, pts=[cnt], color=(255)) # white
# debug:
#cv2.imshow('reel_mask', reel_mask)
#cv2.waitKey(0)
cv2.imshow('reel_mask', reel_mask)
cv2.imwrite('reel4_reel_mask.png', reel_mask)
# 5. Draw the reel area on the original image
contours, hierarchy = cv2.findContours(reel_mask, cv2.RETR_LIST, cv2.CHAIN_APPROX_SIMPLE)
for contourIdx, cnt in enumerate(contours):
centers, radius = cv2.minEnclosingCircle(cnt)
# rescale these values back to the original image size
centers_orig = (centers[0] // SCALE_FACTOR, centers[1] // SCALE_FACTOR)
radius_orig = radius // SCALE_FACTOR
print('centers=', centers_orig, 'radius=', radius_orig)
cv2.circle(output_img, (int(centers_orig[0]), int(centers_orig[1])), int(radius_orig), (128,0,255), 5) # magenta
cv2.imshow('output_img', output_img)
cv2.imwrite('reel5_output.png', output_img)
# display just the pixels from the original image
larger_reel_mask = cv2.resize(reel_mask, (int(img.shape[1]), int(img.shape[0])))
output_reel_img = cv2.bitwise_and(img, img, mask=larger_reel_mask)
cv2.imshow('output_reel_img', output_reel_img)
cv2.imwrite('reel5_output_reel.png', output_reel_img)
cv2.waitKey(0)
At this point, its possible to use larger_reel_maskand compute a minimal enclosing circle, draw it over this mask to make it a little bit more round and allow us to retrieve the area of the reel more accurately:
But the 4 lines of code that achieve this improvement I leave as an exercise for the reader.