Contour detection with OpenCV/Python fails if a pixel is in diagonal proximity
0
I want to detect contours in images and fill them with a arbitrary values.
Unfortunately, I have problems finding the contours if two patches a close to each orther.
E.g. in the following image I expect to find three contours, but just one diagonal pixel in proximity is enough to fail in detecting three separate contours. Is there a way to make contour detection more robust?.
This is the current code
# find contours
cv_img = cv2.imread(f_name+".png")
imgray = cv2.cvtColor(cv_img, cv2.COLOR_BGR2GRAY)
inverted = cv2.bitwise_not(imgray)
#cv2.imwrite(f_name+"_inverted.png", inverted)
im2, contours, hierarchy = cv2.findContours(inverted, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
print ("Found ", len(contours), " contours")
contours = sorted(contours, key=cv2.contourArea)
if len(contours) == 3:
blank_image = np.zeros((int(model_size[1]/4), int(model_size[0]/4), 1), np.uint8)
cv2.drawContours(blank_image, contours, 0, 255, thickness=cv2.FILLED)
cv2.drawContours(blank_image, contours, 1, 255, thickness=cv2.FILLED)
cv2.imwrite(f_name+"_contours.png", blank_image)
python opencv
asked Nov 15 '18 at 15:22
dgratdgrat
1,07621233
add a comment |
0
I want to detect contours in images and fill them with a arbitrary values.
Unfortunately, I have problems finding the contours if two patches a close to each orther.
E.g. in the following image I expect to find three contours, but just one diagonal pixel in proximity is enough to fail in detecting three separate contours. Is there a way to make contour detection more robust?.
This is the current code
# find contours
cv_img = cv2.imread(f_name+".png")
imgray = cv2.cvtColor(cv_img, cv2.COLOR_BGR2GRAY)
inverted = cv2.bitwise_not(imgray)
#cv2.imwrite(f_name+"_inverted.png", inverted)
im2, contours, hierarchy = cv2.findContours(inverted, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
print ("Found ", len(contours), " contours")
contours = sorted(contours, key=cv2.contourArea)
if len(contours) == 3:
blank_image = np.zeros((int(model_size[1]/4), int(model_size[0]/4), 1), np.uint8)
cv2.drawContours(blank_image, contours, 0, 255, thickness=cv2.FILLED)
cv2.drawContours(blank_image, contours, 1, 255, thickness=cv2.FILLED)
cv2.imwrite(f_name+"_contours.png", blank_image)
python opencv
asked Nov 15 '18 at 15:22
dgratdgrat
1,07621233
Have you tried using morphological operations on the image? Additionally, I have worked with lung-segmentation before, and there was this very intuitive method called explosion based thresholding mentioned in this research paper.
– Rick M.
Nov 15 '18 at 15:39
No I haven't, but I fear the current image set is rather binary like. No true gray scale gradient.
– dgrat
Nov 15 '18 at 15:42
Maybe there is a way to split the contour based on concavity
– dgrat
Nov 15 '18 at 15:50
add a comment |
0
0
0
I want to detect contours in images and fill them with a arbitrary values.
Unfortunately, I have problems finding the contours if two patches a close to each orther.
E.g. in the following image I expect to find three contours, but just one diagonal pixel in proximity is enough to fail in detecting three separate contours. Is there a way to make contour detection more robust?.
This is the current code
# find contours
cv_img = cv2.imread(f_name+".png")
imgray = cv2.cvtColor(cv_img, cv2.COLOR_BGR2GRAY)
inverted = cv2.bitwise_not(imgray)
#cv2.imwrite(f_name+"_inverted.png", inverted)
im2, contours, hierarchy = cv2.findContours(inverted, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
print ("Found ", len(contours), " contours")
contours = sorted(contours, key=cv2.contourArea)
if len(contours) == 3:
blank_image = np.zeros((int(model_size[1]/4), int(model_size[0]/4), 1), np.uint8)
cv2.drawContours(blank_image, contours, 0, 255, thickness=cv2.FILLED)
cv2.drawContours(blank_image, contours, 1, 255, thickness=cv2.FILLED)
cv2.imwrite(f_name+"_contours.png", blank_image)
python opencv
asked Nov 15 '18 at 15:22
dgratdgrat
1,07621233
I want to detect contours in images and fill them with a arbitrary values.
Unfortunately, I have problems finding the contours if two patches a close to each orther.
E.g. in the following image I expect to find three contours, but just one diagonal pixel in proximity is enough to fail in detecting three separate contours. Is there a way to make contour detection more robust?.
This is the current code
# find contours
cv_img = cv2.imread(f_name+".png")
imgray = cv2.cvtColor(cv_img, cv2.COLOR_BGR2GRAY)
inverted = cv2.bitwise_not(imgray)
#cv2.imwrite(f_name+"_inverted.png", inverted)
im2, contours, hierarchy = cv2.findContours(inverted, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
print ("Found ", len(contours), " contours")
contours = sorted(contours, key=cv2.contourArea)
if len(contours) == 3:
blank_image = np.zeros((int(model_size[1]/4), int(model_size[0]/4), 1), np.uint8)
cv2.drawContours(blank_image, contours, 0, 255, thickness=cv2.FILLED)
cv2.drawContours(blank_image, contours, 1, 255, thickness=cv2.FILLED)
cv2.imwrite(f_name+"_contours.png", blank_image)
python opencv
python opencv
asked Nov 15 '18 at 15:22
dgratdgrat
1,07621233
asked Nov 15 '18 at 15:22
dgratdgrat
1,07621233
edited Nov 15 '18 at 15:29
dgrat
asked Nov 15 '18 at 15:22
dgratdgrat
1,07621233
asked Nov 15 '18 at 15:22
dgratdgrat
1,07621233
asked Nov 15 '18 at 15:22
dgratdgrat
1,07621233
1,07621233
Have you tried using morphological operations on the image? Additionally, I have worked with lung-segmentation before, and there was this very intuitive method called explosion based thresholding mentioned in this research paper.
– Rick M.
Nov 15 '18 at 15:39
No I haven't, but I fear the current image set is rather binary like. No true gray scale gradient.
– dgrat
Nov 15 '18 at 15:42
Maybe there is a way to split the contour based on concavity
– dgrat
Nov 15 '18 at 15:50
add a comment |
Have you tried using morphological operations on the image? Additionally, I have worked with lung-segmentation before, and there was this very intuitive method called explosion based thresholding mentioned in this research paper.
– Rick M.
Nov 15 '18 at 15:39
No I haven't, but I fear the current image set is rather binary like. No true gray scale gradient.
– dgrat
Nov 15 '18 at 15:42
Maybe there is a way to split the contour based on concavity
– dgrat
Nov 15 '18 at 15:50
Have you tried using morphological operations on the image? Additionally, I have worked with lung-segmentation before, and there was this very intuitive method called explosion based thresholding mentioned in this research paper.
– Rick M.
Nov 15 '18 at 15:39
Have you tried using morphological operations on the image? Additionally, I have worked with lung-segmentation before, and there was this very intuitive method called explosion based thresholding mentioned in this research paper.
– Rick M.
Nov 15 '18 at 15:39
No I haven't, but I fear the current image set is rather binary like. No true gray scale gradient.
– dgrat
Nov 15 '18 at 15:42
No I haven't, but I fear the current image set is rather binary like. No true gray scale gradient.
– dgrat
Nov 15 '18 at 15:42
Maybe there is a way to split the contour based on concavity
– dgrat
Nov 15 '18 at 15:50
Maybe there is a way to split the contour based on concavity
– dgrat
Nov 15 '18 at 15:50
add a comment |
0
active
oldest
votes
Your Answer
StackExchange.ifUsing("editor", function () {
StackExchange.using("externalEditor", function () {
StackExchange.using("snippets", function () {
StackExchange.snippets.init();
});
});
}, "code-snippets");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "1"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
draft saved
draft discarded
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53322618%2fcontour-detection-with-opencv-python-fails-if-a-pixel-is-in-diagonal-proximity%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
0
active
oldest
votes
0
active
oldest
votes
active
oldest
votes
active
oldest
votes
draft saved
draft discarded
Thanks for contributing an answer to Stack Overflow!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.
draft saved
draft discarded
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53322618%2fcontour-detection-with-opencv-python-fails-if-a-pixel-is-in-diagonal-proximity%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Have you tried using morphological operations on the image? Additionally, I have worked with lung-segmentation before, and there was this very intuitive method called explosion based thresholding mentioned in this research paper.
– Rick M.
Nov 15 '18 at 15:39
No I haven't, but I fear the current image set is rather binary like. No true gray scale gradient.
– dgrat
Nov 15 '18 at 15:42
Maybe there is a way to split the contour based on concavity
– dgrat
Nov 15 '18 at 15:50