Society of Robots - Robot Forum
General Misc => Robot Videos => Topic started by: superchiku on March 31, 2009, 02:47:19 AM
-
hey guys thanks to u ppl i was able to develop this simple blob detection algorithm which is fairly accurate and can serve any general purpose....enjoy the video and the song
[youtube]yXsVUuh_Frs[/youtube]
-
cheerio...i jst improvedthe algo..the accuracy is better now...
-
nice, very very nice. I have a few questions about it though. If there is only 2 blobs, will it only show blue and green? Does it have to be that exact color red?
-
yes it will show blue and green only..... actually i designed it for red blobs..but the code only needs minor modifications for blobs of any other colour...like only changing an array index value...
-
Is it hooked up to a webcam?
-
ya video processing...
-
Can I have the code so I can try it out?
-
its in opencv sure u can have t ..if u have yahoo messenger then give me ur id ..ill addd and send u
-
hi can i have to code as well..
i manage to try out the sample in opencv which detect reb image. i wada to learn to add blob and track it.
please sent to [email protected]
thanks.. :)
-
i detected sum bugs with the code...i need to add sum extra filtering to it...if u still need it then ill give u
-
yup.. thanks.. :)
-
ok ill send the code to ur email...correct the bugs.. .
-
hello..
i have yet to receive to code.. can you help me send again..
[email protected] or [email protected]
thanks.. :)
-
could you send me the code? greetings from Colombia
-
very nice :)
-
hi guys, i receive some error while trying to compile the above sample code, has it got to do with the initial declaration??
int height1=hw->height;
int width1=hw->width;
warning C4244: 'argument' : conversion from 'float' to 'int', possible loss of data
//finding centre of gravity...
for(float i=0;i<height1;i++)
{
for(float j=0;j<width1;j++)
{
s=cvGet2D(hw,i,j); ******** error is here******************
if(s.val[2]>120)
{
xcounter++;
ycounter++;
xpixel+=j;
ypixel+=i;
}
}
-
Hello may i have the code also?
[email protected]
Thank you very much