ฝัง
- เผยแพร่เมื่อ 4 เม.ย. 2018
- source code: pysource.com/2018/04/05/object...
➤ Full Videocourses:
Object Detection: pysource.com/object-detection...
➤ Follow me on:
Instagram: / pysource7
LinkedIn: / pysource
➤ For business inquiries:
pysource.com/contact
![มองหน้ากันไม่ติด (Awkward) - โอ๊ต ปราโมทย์ FEAT. MAIYARAP [Official MV]](http://i.ytimg.com/vi/4b02SY_r6Gw/mqdefault.jpg)
One of the best tutorial for me. With this way, we can detect and recognize objects quickly and without internet connection on the raspberry pi. Thanks a lot!!!
some of the best opencv videos I have watched. Thanks. :-l
Thank you, you saved my project!
excellent video and nice explanation
One of the best Tutorials that I have ever seen. Thank you so much for recording and sharing this video!
Nice video, thank you !
Thank you so much sir.. great informative video...
one of the best tutorials for open cv i have found on youtube thanks a lot for the videos
Thank you for the great tutorial and great explanations of the object detection process. Btw the link to your source code is down, so I could not put it into practice. I would really like to give it a try on this cool tutorial.
I want to track bubbles ascending in a tube and save the data automatically, can it do that for several bubbles? I want their position, velocity and acceleration. It'd be nice if it could measure changes in diameter.
interesting👍
Can algorithm detect when have no object in picture? It also draw any rectangle.
Cześć pomożesz mi z zastosowaniem OpenCV do zmiany TCP robota przy pobieraniu detalu ?
Subscribed (Y)
can someone explain me please why we are getting hue value greater than 360
Hi Sir, I want to count products moving on 1 conveyor by the camera. You can help me? and maybe the Guide to do this project?
Hi, this is something I teach on my course Object Detection (with Opencv and Deep Learning) and I actually have a built in project as well that does that.
you can find it on pysource.com
@@pysource-com thanks so much
is this algorithm color-based? If there are similar colors in the scene, does it fail?
Yes, it might fail if there is a similar color to the objects color
Thanks for the reply. Do you have any experience with KCF tracker? Is it a better option for scenes in which the object to track is similar to the background color?
yes it can fail
where is the moutwash video ?
Incredile
Can you please tell me why hsv only ?? And why not bgr
because hist give us hsv values of the image
how did you find the position of sticker 3:49
Hi, I found that manually bty tryal and error, designing a square on the first frame until I got the sticker right.
There are for sure better ways to do that, I just didn't want to take much time for that as I wanted the tutorial to be focused only on the meanshift.
You could use mouse events to select the area you want to track.
Take a look a this video th-cam.com/video/H_068uiMR9M/w-d-xo.html
the easiest way to do it would be to hover your mouse over the video since cv2 shows you the location of your mouse cursor in the left bottom corner
@@ChristianHaschek How to get the cursor location in cv?
@@neilsumanth8013 If you already see the image in cv you can just hover the mouse and in the bottom left corner you see the corrdinates of the mouse
@@ChristianHaschekThankx Haschek..But .. Iam using atom as my IDE and running code in cmd, at which left corner we get that axis ?
I am facing a bit of a problem here. I tried the mask, but instead of focussing on the object to be white, it is instead putting it in black and rest in white due to which my tracker was not working correctly. Would be glad if u could help me! Also, great video man!
You can use bitwise operators to make the inverse of the mask. Look for my video where I talk about bitwise operators and you'll get there the solution for your mask
@@pysource-com got it, just gonna have to make an attempt! Thanks 🤘🤘
hey, I just tried the bitwise_not function of open cv2, and apparently the conversion worked in a similar way as in your other video. But, it still did not work with respect to tracking. Any clue why?
actually, on your video it's working fine, but on a custom object detection it is not.
hello!
i get this problem when running the tracking code "Process finished with exit code -1
",
can anyone please explain?
can it be an issue of graphics card or some hardware? i am using thinkpad T420 laptop with intel HD graphics 3000
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
in ()
13 mask = cv2.inRange(hsv, lower_blue, upper_blue)
14
---> 15 _, contours, _ = cv2.findContours(mask, cv2.RETR_TREE, cv2.CHAIN_APPROX_NONE)
16
17 for contour in contours:
ValueError: need more than 2 values to unpack