Cross detection and tracking in a video

classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Cross detection and tracking in a video

opencv-users mailing list
Hi,

I am trying to detect and track a black cross (+) in white background in a grayscale video. The lines of the cross are about 1 meters, but the distance of the camera from the cross changes in the video, so I don't know the size beforehand. Unfortunately the view angle is also changing, therefore it is seen as X in the video and the angle between the lines is not seen as perpendicular. It is the only black cross in the video, however the environment is very cluttered and also due to illumination changes and light reflection in the video one can not suppose the cross to be seen exactly as black in a white background. Would you please me some idea of how I can detect and track this cross?

Best regards,
Azarakhsh Keipour
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Cross detection and tracking in a video

opencv-users mailing list
Try to use the key points detector and descriptor (like SIFT, SURF, BRISK ...) to detect your cross. because this descriptor are local and invariant to the luminosity and affine transform
 

    Le Mardi 26 avril 2016 18h10, "[hidden email] [OpenCV]" <[hidden email]> a écrit :
 

     Hi,

I am trying to detect and track a black cross (+) in white background in a grayscale video. The lines of the cross are about 1 meters, but the distance of the camera from the cross changes in the video, so I don't know the size beforehand. Unfortunately the view angle is also changing, therefore it is seen as X in the video and the angle between the lines is not seen as perpendicular. It is the only black cross in the video, however the environment is very cluttered and also due to illumination changes and light reflection in the video one can not suppose the cross to be seen exactly as black in a white background. Would you please me some idea of how I can detect and track this cross?

Best regards,
Azarakhsh Keipour  #yiv1267104573 #yiv1267104573 -- #yiv1267104573ygrp-mkp {border:1px solid #d8d8d8;font-family:Arial;margin:10px 0;padding:0 10px;}#yiv1267104573 #yiv1267104573ygrp-mkp hr {border:1px solid #d8d8d8;}#yiv1267104573 #yiv1267104573ygrp-mkp #yiv1267104573hd {color:#628c2a;font-size:85%;font-weight:700;line-height:122%;margin:10px 0;}#yiv1267104573 #yiv1267104573ygrp-mkp #yiv1267104573ads {margin-bottom:10px;}#yiv1267104573 #yiv1267104573ygrp-mkp .yiv1267104573ad {padding:0 0;}#yiv1267104573 #yiv1267104573ygrp-mkp .yiv1267104573ad p {margin:0;}#yiv1267104573 #yiv1267104573ygrp-mkp .yiv1267104573ad a {color:#0000ff;text-decoration:none;}#yiv1267104573 #yiv1267104573ygrp-sponsor #yiv1267104573ygrp-lc {font-family:Arial;}#yiv1267104573 #yiv1267104573ygrp-sponsor #yiv1267104573ygrp-lc #yiv1267104573hd {margin:10px 0px;font-weight:700;font-size:78%;line-height:122%;}#yiv1267104573 #yiv1267104573ygrp-sponsor #yiv1267104573ygrp-lc .yiv1267104573ad {margin-bottom:10px;padding:0 0;}#yiv1267104573 #yiv1267104573actions {font-family:Verdana;font-size:11px;padding:10px 0;}#yiv1267104573 #yiv1267104573activity {background-color:#e0ecee;float:left;font-family:Verdana;font-size:10px;padding:10px;}#yiv1267104573 #yiv1267104573activity span {font-weight:700;}#yiv1267104573 #yiv1267104573activity span:first-child {text-transform:uppercase;}#yiv1267104573 #yiv1267104573activity span a {color:#5085b6;text-decoration:none;}#yiv1267104573 #yiv1267104573activity span span {color:#ff7900;}#yiv1267104573 #yiv1267104573activity span .yiv1267104573underline {text-decoration:underline;}#yiv1267104573 .yiv1267104573attach {clear:both;display:table;font-family:Arial;font-size:12px;padding:10px 0;width:400px;}#yiv1267104573 .yiv1267104573attach div a {text-decoration:none;}#yiv1267104573 .yiv1267104573attach img {border:none;padding-right:5px;}#yiv1267104573 .yiv1267104573attach label {display:block;margin-bottom:5px;}#yiv1267104573 .yiv1267104573attach label a {text-decoration:none;}#yiv1267104573 blockquote {margin:0 0 0 4px;}#yiv1267104573 .yiv1267104573bold {font-family:Arial;font-size:13px;font-weight:700;}#yiv1267104573 .yiv1267104573bold a {text-decoration:none;}#yiv1267104573 dd.yiv1267104573last p a {font-family:Verdana;font-weight:700;}#yiv1267104573 dd.yiv1267104573last p span {margin-right:10px;font-family:Verdana;font-weight:700;}#yiv1267104573 dd.yiv1267104573last p span.yiv1267104573yshortcuts {margin-right:0;}#yiv1267104573 div.yiv1267104573attach-table div div a {text-decoration:none;}#yiv1267104573 div.yiv1267104573attach-table {width:400px;}#yiv1267104573 div.yiv1267104573file-title a, #yiv1267104573 div.yiv1267104573file-title a:active, #yiv1267104573 div.yiv1267104573file-title a:hover, #yiv1267104573 div.yiv1267104573file-title a:visited {text-decoration:none;}#yiv1267104573 div.yiv1267104573photo-title a, #yiv1267104573 div.yiv1267104573photo-title a:active, #yiv1267104573 div.yiv1267104573photo-title a:hover, #yiv1267104573 div.yiv1267104573photo-title a:visited {text-decoration:none;}#yiv1267104573 div#yiv1267104573ygrp-mlmsg #yiv1267104573ygrp-msg p a span.yiv1267104573yshortcuts {font-family:Verdana;font-size:10px;font-weight:normal;}#yiv1267104573 .yiv1267104573green {color:#628c2a;}#yiv1267104573 .yiv1267104573MsoNormal {margin:0 0 0 0;}#yiv1267104573 o {font-size:0;}#yiv1267104573 #yiv1267104573photos div {float:left;width:72px;}#yiv1267104573 #yiv1267104573photos div div {border:1px solid #666666;height:62px;overflow:hidden;width:62px;}#yiv1267104573 #yiv1267104573photos div label {color:#666666;font-size:10px;overflow:hidden;text-align:center;white-space:nowrap;width:64px;}#yiv1267104573 #yiv1267104573reco-category {font-size:77%;}#yiv1267104573 #yiv1267104573reco-desc {font-size:77%;}#yiv1267104573 .yiv1267104573replbq {margin:4px;}#yiv1267104573 #yiv1267104573ygrp-actbar div a:first-child {margin-right:2px;padding-right:5px;}#yiv1267104573 #yiv1267104573ygrp-mlmsg {font-size:13px;font-family:Arial, helvetica, clean, sans-serif;}#yiv1267104573 #yiv1267104573ygrp-mlmsg table {font-size:inherit;font:100%;}#yiv1267104573 #yiv1267104573ygrp-mlmsg select, #yiv1267104573 input, #yiv1267104573 textarea {font:99% Arial, Helvetica, clean, sans-serif;}#yiv1267104573 #yiv1267104573ygrp-mlmsg pre, #yiv1267104573 code {font:115% monospace;}#yiv1267104573 #yiv1267104573ygrp-mlmsg * {line-height:1.22em;}#yiv1267104573 #yiv1267104573ygrp-mlmsg #yiv1267104573logo {padding-bottom:10px;}#yiv1267104573 #yiv1267104573ygrp-msg p a {font-family:Verdana;}#yiv1267104573 #yiv1267104573ygrp-msg p#yiv1267104573attach-count span {color:#1E66AE;font-weight:700;}#yiv1267104573 #yiv1267104573ygrp-reco #yiv1267104573reco-head {color:#ff7900;font-weight:700;}#yiv1267104573 #yiv1267104573ygrp-reco {margin-bottom:20px;padding:0px;}#yiv1267104573 #yiv1267104573ygrp-sponsor #yiv1267104573ov li a {font-size:130%;text-decoration:none;}#yiv1267104573 #yiv1267104573ygrp-sponsor #yiv1267104573ov li {font-size:77%;list-style-type:square;padding:6px 0;}#yiv1267104573 #yiv1267104573ygrp-sponsor #yiv1267104573ov ul {margin:0;padding:0 0 0 8px;}#yiv1267104573 #yiv1267104573ygrp-text {font-family:Georgia;}#yiv1267104573 #yiv1267104573ygrp-text p {margin:0 0 1em 0;}#yiv1267104573 #yiv1267104573ygrp-text tt {font-size:120%;}#yiv1267104573 #yiv1267104573ygrp-vital ul li:last-child {border-right:none !important;}#yiv1267104573

 
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Cross detection and tracking in a video

opencv-users mailing list
Thank you very much for your reply. Actually I forgot to tell that I need it to work in real-time (30HZ ideally and 5HZ in worst case), so I think the feature detector/descriptors like SIFT, SURF, etc. are off the table. Is there a detector/descriptor that is fast enough? Also are they capable of finding good features for such a simple shape without a rich background?
Thank you,Azarakhsh







    On Wednesday, April 27, 2016 9:24 AM, "yahia Bammoune [hidden email] [OpenCV]" <[hidden email]> wrote:



     Try to use the key points detector and descriptor (like SIFT, SURF, BRISK ...) to detect your cross. because this descriptor are local and invariant to the luminosity and affine transform



    Le Mardi 26 avril 2016 18h10, "[hidden email] [OpenCV]" <[hidden email]> a écrit :



     Hi,


I am trying to detect and track a black cross (+) in white background in a grayscale video. The lines of the cross are about 1 meters, but the distance of the camera from the cross changes in the video, so I don't know the size beforehand. Unfortunately the view angle is also changing, therefore it is seen as X in the video and the angle between the lines is not seen as perpendicular. It is the only black cross in the video, however the environment is very cluttered and also due to illumination changes and light reflection in the video one can not suppose the cross to be seen exactly as black in a white background. Would you please me some idea of how I can detect and track this cross?


Best regards,
Azarakhsh Keipour


     #yiv6217283304 #yiv6217283304 -- #yiv6217283304ygrp-mkp {border:1px solid #d8d8d8;font-family:Arial;margin:10px 0;padding:0 10px;}#yiv6217283304 #yiv6217283304ygrp-mkp hr {border:1px solid #d8d8d8;}#yiv6217283304 #yiv6217283304ygrp-mkp #yiv6217283304hd {color:#628c2a;font-size:85%;font-weight:700;line-height:122%;margin:10px 0;}#yiv6217283304 #yiv6217283304ygrp-mkp #yiv6217283304ads {margin-bottom:10px;}#yiv6217283304 #yiv6217283304ygrp-mkp .yiv6217283304ad {padding:0 0;}#yiv6217283304 #yiv6217283304ygrp-mkp .yiv6217283304ad p {margin:0;}#yiv6217283304 #yiv6217283304ygrp-mkp .yiv6217283304ad a {color:#0000ff;text-decoration:none;}#yiv6217283304 #yiv6217283304ygrp-sponsor #yiv6217283304ygrp-lc {font-family:Arial;}#yiv6217283304 #yiv6217283304ygrp-sponsor #yiv6217283304ygrp-lc #yiv6217283304hd {margin:10px 0px;font-weight:700;font-size:78%;line-height:122%;}#yiv6217283304 #yiv6217283304ygrp-sponsor #yiv6217283304ygrp-lc .yiv6217283304ad {margin-bottom:10px;padding:0 0;}#yiv6217283304 #yiv6217283304actions {font-family:Verdana;font-size:11px;padding:10px 0;}#yiv6217283304 #yiv6217283304activity {background-color:#e0ecee;float:left;font-family:Verdana;font-size:10px;padding:10px;}#yiv6217283304 #yiv6217283304activity span {font-weight:700;}#yiv6217283304 #yiv6217283304activity span:first-child {text-transform:uppercase;}#yiv6217283304 #yiv6217283304activity span a {color:#5085b6;text-decoration:none;}#yiv6217283304 #yiv6217283304activity span span {color:#ff7900;}#yiv6217283304 #yiv6217283304activity span .yiv6217283304underline {text-decoration:underline;}#yiv6217283304 .yiv6217283304attach {clear:both;display:table;font-family:Arial;font-size:12px;padding:10px 0;width:400px;}#yiv6217283304 .yiv6217283304attach div a {text-decoration:none;}#yiv6217283304 .yiv6217283304attach img {border:none;padding-right:5px;}#yiv6217283304 .yiv6217283304attach label {display:block;margin-bottom:5px;}#yiv6217283304 .yiv6217283304attach label a {text-decoration:none;}#yiv6217283304 blockquote {margin:0 0 0 4px;}#yiv6217283304 .yiv6217283304bold {font-family:Arial;font-size:13px;font-weight:700;}#yiv6217283304 .yiv6217283304bold a {text-decoration:none;}#yiv6217283304 dd.yiv6217283304last p a {font-family:Verdana;font-weight:700;}#yiv6217283304 dd.yiv6217283304last p span {margin-right:10px;font-family:Verdana;font-weight:700;}#yiv6217283304 dd.yiv6217283304last p span.yiv6217283304yshortcuts {margin-right:0;}#yiv6217283304 div.yiv6217283304attach-table div div a {text-decoration:none;}#yiv6217283304 div.yiv6217283304attach-table {width:400px;}#yiv6217283304 div.yiv6217283304file-title a, #yiv6217283304 div.yiv6217283304file-title a:active, #yiv6217283304 div.yiv6217283304file-title a:hover, #yiv6217283304 div.yiv6217283304file-title a:visited {text-decoration:none;}#yiv6217283304 div.yiv6217283304photo-title a, #yiv6217283304 div.yiv6217283304photo-title a:active, #yiv6217283304 div.yiv6217283304photo-title a:hover, #yiv6217283304 div.yiv6217283304photo-title a:visited {text-decoration:none;}#yiv6217283304 div#yiv6217283304ygrp-mlmsg #yiv6217283304ygrp-msg p a span.yiv6217283304yshortcuts {font-family:Verdana;font-size:10px;font-weight:normal;}#yiv6217283304 .yiv6217283304green {color:#628c2a;}#yiv6217283304 .yiv6217283304MsoNormal {margin:0 0 0 0;}#yiv6217283304 o {font-size:0;}#yiv6217283304 #yiv6217283304photos div {float:left;width:72px;}#yiv6217283304 #yiv6217283304photos div div {border:1px solid #666666;height:62px;overflow:hidden;width:62px;}#yiv6217283304 #yiv6217283304photos div label {color:#666666;font-size:10px;overflow:hidden;text-align:center;white-space:nowrap;width:64px;}#yiv6217283304 #yiv6217283304reco-category {font-size:77%;}#yiv6217283304 #yiv6217283304reco-desc {font-size:77%;}#yiv6217283304 .yiv6217283304replbq {margin:4px;}#yiv6217283304 #yiv6217283304ygrp-actbar div a:first-child {margin-right:2px;padding-right:5px;}#yiv6217283304 #yiv6217283304ygrp-mlmsg {font-size:13px;font-family:Arial, helvetica, clean, sans-serif;}#yiv6217283304 #yiv6217283304ygrp-mlmsg table {font-size:inherit;font:100%;}#yiv6217283304 #yiv6217283304ygrp-mlmsg select, #yiv6217283304 input, #yiv6217283304 textarea {font:99% Arial, Helvetica, clean, sans-serif;}#yiv6217283304 #yiv6217283304ygrp-mlmsg pre, #yiv6217283304 code {font:115% monospace;}#yiv6217283304 #yiv6217283304ygrp-mlmsg * {line-height:1.22em;}#yiv6217283304 #yiv6217283304ygrp-mlmsg #yiv6217283304logo {padding-bottom:10px;}#yiv6217283304 #yiv6217283304ygrp-msg p a {font-family:Verdana;}#yiv6217283304 #yiv6217283304ygrp-msg p#yiv6217283304attach-count span {color:#1E66AE;font-weight:700;}#yiv6217283304 #yiv6217283304ygrp-reco #yiv6217283304reco-head {color:#ff7900;font-weight:700;}#yiv6217283304 #yiv6217283304ygrp-reco {margin-bottom:20px;padding:0px;}#yiv6217283304 #yiv6217283304ygrp-sponsor #yiv6217283304ov li a {font-size:130%;text-decoration:none;}#yiv6217283304 #yiv6217283304ygrp-sponsor #yiv6217283304ov li {font-size:77%;list-style-type:square;padding:6px 0;}#yiv6217283304 #yiv6217283304ygrp-sponsor #yiv6217283304ov ul {margin:0;padding:0 0 0 8px;}#yiv6217283304 #yiv6217283304ygrp-text {font-family:Georgia;}#yiv6217283304 #yiv6217283304ygrp-text p {margin:0 0 1em 0;}#yiv6217283304 #yiv6217283304ygrp-text tt {font-size:120%;}#yiv6217283304 #yiv6217283304ygrp-vital ul li:last-child {border-right:none !important;}#yiv6217283304


Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Cross detection and tracking in a video

opencv-users mailing list
The biggest issue you may have with SIFT/SURF is that they are patented, so
there are legal issues involved.

With regards to speed, they are multi-scale operations, so there is scope
for significant parallelism, which might allow them to be speeded up
sufficiently for your purposes (especially depending on the hardware you
use).  However, my guess is that the legal issues are the bigger obstacle
for the time being.

Frank

On 28 April 2016 at 03:54, Azarakhsh Keipour [hidden email] [OpenCV] <
[hidden email]> wrote:

>
>
> Thank you very much for your reply. Actually I forgot to tell that I need
> it to work in real-time (30HZ ideally and 5HZ in worst case), so I think
> the feature detector/descriptors like SIFT, SURF, etc. are off the table.
> Is there a detector/descriptor that is fast enough? Also are they capable
> of finding good features for such a simple shape without a rich background?
>
> Thank you,
> Azarakhsh
>
>
>
>
> On Wednesday, April 27, 2016 9:24 AM, "yahia Bammoune
> [hidden email] [OpenCV]" <[hidden email]> wrote:
>
>
>
> Try to use the key points detector and descriptor (like SIFT, SURF, BRISK
> ...) to detect your cross. because this descriptor are local and invariant
> to the luminosity and affine transform
>
>
> Le Mardi 26 avril 2016 18h10, "[hidden email] [OpenCV]" <
> [hidden email]> a écrit :
>
>
>
> Hi,
>
> I am trying to detect and track a black cross (+) in white background in a
> grayscale video. The lines of the cross are about 1 meters, but the
> distance of the camera from the cross changes in the video, so I don't know
> the size beforehand. Unfortunately the view angle is also changing,
> therefore it is seen as X in the video and the angle between the lines is
> not seen as perpendicular. It is the only black cross in the video, however
> the environment is very cluttered and also due to illumination changes and
> light reflection in the video one can not suppose the cross to be seen
> exactly as black in a white background. Would you please me some idea of
> how I can detect and track this cross?
>
> Best regards,
> Azarakhsh Keipour
>
>
>
>
>
>
>
>
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Cross detection and tracking in a video

opencv-users mailing list
In reply to this post by opencv-users mailing list
Hi,
I give you two options as follows:
1. Use cascade classifier with Adaboost. The tutorial and explanation is
available in the link below.
http://docs.opencv.org/2.4/doc/tutorials/objdetect/cascade_classifier/cascade_classifier.html
2. Use conventional methods, for example, histogram distribution, line
detection, etc., to locate ROI, then refine the location of cross mark by
mean shift or cam shift.
This will work well in real-time manner. However, some aspects such as
robustness or accuracy need to be more carefully examined.
Hope you find the best solutions.
---------------------------------------------------
Prasarn Kiddee




*12th Floor,Intelligent BuildingState Key Laboratory of Management and
Control for Complex SystemsInstitute of Automation, Chinese Academy of
Sciences95 Zhongguancun East Road,*
*Haidian District, Beijing, China, 100190*
-----------------------------------------------------------
桑 凯迪




*12 层 复杂系统与智能科学实验室自动化研究所,中国科学院大学北京市海淀区 中关村东路 95号 邮编:100190*
[hidden email]
--------------------------------------------------
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Cross detection and tracking in a video

opencv-users mailing list
In reply to this post by opencv-users mailing list
Additive comment to my last post!
If you are a newbie in OpenCv, I would strongly recommend that you should
try option 2. firstly.
It is worth practicing such the fundamental methods.
Good luck!
---------------------------------------------------
Prasarn Kiddee




*12th Floor,Intelligent BuildingState Key Laboratory of Management and
Control for Complex SystemsInstitute of Automation, Chinese Academy of
Sciences95 Zhongguancun East Road,*
*Haidian District, Beijing, China, 100190*
-----------------------------------------------------------
桑 凯迪




*12 层 复杂系统与智能科学实验室自动化研究所,中国科学院大学北京市海淀区 中关村东路 95号 邮编:100190*
[hidden email]
--------------------------------------------------
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Cross detection and tracking in a video

opencv-users mailing list
Thank you very much Prasarn, Frank and Yahia. Your suggestions helped me a lot. I will try the cascade classifier with Adaboost and conventional methods to see which is more robust.


About the SIFT/SURF I couldn't make the nonfree module working on the target platform (an on-board tegra pc) and it was very slow on my more powerful desktop pc (about 1 HZ); I will try it again if the Prasarn's suggestions don't work. 
Thank you all again for your helpful ideas.



    On Thursday, April 28, 2016 1:58 AM, "Prasarn Kiddee [hidden email] [OpenCV]" <[hidden email]> wrote:



     Additive comment to my last post!If you are a newbie in OpenCv, I would strongly recommend that you should try option 2. firstly.It is worth practicing such the fundamental methods.Good luck! 
---------------------------------------------------
Prasarn Kiddee
12th Floor,Intelligent Building
State Key Laboratory of Management and Control for Complex Systems
Institute of Automation, Chinese Academy of Sciences
95 Zhongguancun East Road,Haidian District, Beijing, China, 100190
-----------------------------------------------------------
桑 凯迪
12 层 复杂系统与智能科学实验室
自动化研究所,中国科学院大学
北京市海淀区 中关村东路 95号
 邮编:100190
[hidden email]
--------------------------------------------------
  #yiv2171346812 #yiv2171346812 -- #yiv2171346812ygrp-mkp {border:1px solid #d8d8d8;font-family:Arial;margin:10px 0;padding:0 10px;}#yiv2171346812 #yiv2171346812ygrp-mkp hr {border:1px solid #d8d8d8;}#yiv2171346812 #yiv2171346812ygrp-mkp #yiv2171346812hd {color:#628c2a;font-size:85%;font-weight:700;line-height:122%;margin:10px 0;}#yiv2171346812 #yiv2171346812ygrp-mkp #yiv2171346812ads {margin-bottom:10px;}#yiv2171346812 #yiv2171346812ygrp-mkp .yiv2171346812ad {padding:0 0;}#yiv2171346812 #yiv2171346812ygrp-mkp .yiv2171346812ad p {margin:0;}#yiv2171346812 #yiv2171346812ygrp-mkp .yiv2171346812ad a {color:#0000ff;text-decoration:none;}#yiv2171346812 #yiv2171346812ygrp-sponsor #yiv2171346812ygrp-lc {font-family:Arial;}#yiv2171346812 #yiv2171346812ygrp-sponsor #yiv2171346812ygrp-lc #yiv2171346812hd {margin:10px 0px;font-weight:700;font-size:78%;line-height:122%;}#yiv2171346812 #yiv2171346812ygrp-sponsor #yiv2171346812ygrp-lc .yiv2171346812ad {margin-bottom:10px;padding:0 0;}#yiv2171346812 #yiv2171346812actions {font-family:Verdana;font-size:11px;padding:10px 0;}#yiv2171346812 #yiv2171346812activity {background-color:#e0ecee;float:left;font-family:Verdana;font-size:10px;padding:10px;}#yiv2171346812 #yiv2171346812activity span {font-weight:700;}#yiv2171346812 #yiv2171346812activity span:first-child {text-transform:uppercase;}#yiv2171346812 #yiv2171346812activity span a {color:#5085b6;text-decoration:none;}#yiv2171346812 #yiv2171346812activity span span {color:#ff7900;}#yiv2171346812 #yiv2171346812activity span .yiv2171346812underline {text-decoration:underline;}#yiv2171346812 .yiv2171346812attach {clear:both;display:table;font-family:Arial;font-size:12px;padding:10px 0;width:400px;}#yiv2171346812 .yiv2171346812attach div a {text-decoration:none;}#yiv2171346812 .yiv2171346812attach img {border:none;padding-right:5px;}#yiv2171346812 .yiv2171346812attach label {display:block;margin-bottom:5px;}#yiv2171346812 .yiv2171346812attach label a {text-decoration:none;}#yiv2171346812 blockquote {margin:0 0 0 4px;}#yiv2171346812 .yiv2171346812bold {font-family:Arial;font-size:13px;font-weight:700;}#yiv2171346812 .yiv2171346812bold a {text-decoration:none;}#yiv2171346812 dd.yiv2171346812last p a {font-family:Verdana;font-weight:700;}#yiv2171346812 dd.yiv2171346812last p span {margin-right:10px;font-family:Verdana;font-weight:700;}#yiv2171346812 dd.yiv2171346812last p span.yiv2171346812yshortcuts {margin-right:0;}#yiv2171346812 div.yiv2171346812attach-table div div a {text-decoration:none;}#yiv2171346812 div.yiv2171346812attach-table {width:400px;}#yiv2171346812 div.yiv2171346812file-title a, #yiv2171346812 div.yiv2171346812file-title a:active, #yiv2171346812 div.yiv2171346812file-title a:hover, #yiv2171346812 div.yiv2171346812file-title a:visited {text-decoration:none;}#yiv2171346812 div.yiv2171346812photo-title a, #yiv2171346812 div.yiv2171346812photo-title a:active, #yiv2171346812 div.yiv2171346812photo-title a:hover, #yiv2171346812 div.yiv2171346812photo-title a:visited {text-decoration:none;}#yiv2171346812 div#yiv2171346812ygrp-mlmsg #yiv2171346812ygrp-msg p a span.yiv2171346812yshortcuts {font-family:Verdana;font-size:10px;font-weight:normal;}#yiv2171346812 .yiv2171346812green {color:#628c2a;}#yiv2171346812 .yiv2171346812MsoNormal {margin:0 0 0 0;}#yiv2171346812 o {font-size:0;}#yiv2171346812 #yiv2171346812photos div {float:left;width:72px;}#yiv2171346812 #yiv2171346812photos div div {border:1px solid #666666;height:62px;overflow:hidden;width:62px;}#yiv2171346812 #yiv2171346812photos div label {color:#666666;font-size:10px;overflow:hidden;text-align:center;white-space:nowrap;width:64px;}#yiv2171346812 #yiv2171346812reco-category {font-size:77%;}#yiv2171346812 #yiv2171346812reco-desc {font-size:77%;}#yiv2171346812 .yiv2171346812replbq {margin:4px;}#yiv2171346812 #yiv2171346812ygrp-actbar div a:first-child {margin-right:2px;padding-right:5px;}#yiv2171346812 #yiv2171346812ygrp-mlmsg {font-size:13px;font-family:Arial, helvetica, clean, sans-serif;}#yiv2171346812 #yiv2171346812ygrp-mlmsg table {font-size:inherit;font:100%;}#yiv2171346812 #yiv2171346812ygrp-mlmsg select, #yiv2171346812 input, #yiv2171346812 textarea {font:99% Arial, Helvetica, clean, sans-serif;}#yiv2171346812 #yiv2171346812ygrp-mlmsg pre, #yiv2171346812 code {font:115% monospace;}#yiv2171346812 #yiv2171346812ygrp-mlmsg * {line-height:1.22em;}#yiv2171346812 #yiv2171346812ygrp-mlmsg #yiv2171346812logo {padding-bottom:10px;}#yiv2171346812 #yiv2171346812ygrp-msg p a {font-family:Verdana;}#yiv2171346812 #yiv2171346812ygrp-msg p#yiv2171346812attach-count span {color:#1E66AE;font-weight:700;}#yiv2171346812 #yiv2171346812ygrp-reco #yiv2171346812reco-head {color:#ff7900;font-weight:700;}#yiv2171346812 #yiv2171346812ygrp-reco {margin-bottom:20px;padding:0px;}#yiv2171346812 #yiv2171346812ygrp-sponsor #yiv2171346812ov li a {font-size:130%;text-decoration:none;}#yiv2171346812 #yiv2171346812ygrp-sponsor #yiv2171346812ov li {font-size:77%;list-style-type:square;padding:6px 0;}#yiv2171346812 #yiv2171346812ygrp-sponsor #yiv2171346812ov ul {margin:0;padding:0 0 0 8px;}#yiv2171346812 #yiv2171346812ygrp-text {font-family:Georgia;}#yiv2171346812 #yiv2171346812ygrp-text p {margin:0 0 1em 0;}#yiv2171346812 #yiv2171346812ygrp-text tt {font-size:120%;}#yiv2171346812 #yiv2171346812ygrp-vital ul li:last-child {border-right:none !important;}#yiv2171346812


Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Cross detection and tracking in a video

opencv-users mailing list
take a look at (Best) way to track a circle with a cross inside - OpenCV Q&A Forum http://answers.opencv.org/question/93686/best-way-to-track-a-circle-with-a-cross-inside/ 
 
 http://answers.opencv.org/question/93686/best-way-to-track-a-circle-with-a-cross-inside/ 
 
 (Best) way to track a circle with a cross inside - OpenC... http://answers.opencv.org/question/93686/best-way-to-track-a-circle-with-a-cross-inside/ Hi, I am currently trying to find a solution on how to track a specific target which looks like this: It is just a circle with a cross inside. I am using opencv 2.4...
 
 
 
 View on answers.opencv.org http://answers.opencv.org/question/93686/best-way-to-track-a-circle-with-a-cross-inside/ 
 Preview by Yahoo
 
 
 
Loading...