LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Tracking / Classification / Extracting Objects Trajectory Data Points


Hello,
I Have Data points of Multiple Objects Trajectory which I had Recorded at a Specific time Interval as Shown in the Figure Below.

Capture2.PNG

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I want to Extract/ Track Multiple Objects Trajectory so for that I have Implemented GNN (Global Nearest Neighbor) but it wont work in case of Missing Data Points and when Object Trajectory Cross each other.

 

 

Capture1.PNG

 

 

I have Attached the Data log File Kindly Help me Out in this Regard.

0 Kudos
Message 1 of 14
(458 Views)

Perhaps you should post the GNN algorithm.

0 Kudos
Message 2 of 14
(421 Views)

Kindly Check This File I have Implemented GNN Algorithm you can set gate Size Manually by Changing the constants in the Loop this code has a bug as it didn't give you the Track Trajectory Correctly

0 Kudos
Message 3 of 14
(401 Views)

@mrjunaidkahn wrote:

Kindly Check This File I have Implemented GNN Algorithm you can set gate Size Manually by Changing the constants in the Loop this code has a bug as it didn't give you the Track Trajectory Correctly


There's no file.

0 Kudos
Message 4 of 14
(397 Views)

Kindly Check this

0 Kudos
Message 5 of 14
(391 Views)

@mrjunaidkahn wrote:

Kindly Check this


Lots of things don't make sense to me.

 

Lots of duplicate code...

 

Why convert from integer strings to integer, to double, to string, to integer, and then compare which strings is > or <?

 

I'd start with parsing the input file, something like this (although I can't say for sure what you're trying to do):

Hop To Data.png

0 Kudos
Message 6 of 14
(379 Views)

 

do you also need to label each trajectory, or

do you want to solely remove noise from a signal e.g.

 

alexderjuengere_0-1637769606737.png

 

you define a starting point via double-click in the input graph, and then calculate distances regarding to this initial value over and over again.

shouldn't you update the initial position after a while?

 

 

 

 

0 Kudos
Message 7 of 14
(363 Views)

I would just read the file as follows for further processing:

 

altenbach_0-1637862424684.png

 

 

You seem to do a lot of 2D trigonometry, so I recommend to use a complex datatype!

 

0 Kudos
Message 8 of 14
(340 Views)

Just to give you some ideas, here's a tiny codelet to replace screenfuls of your Rube Goldberg code to create a cleaned up version of your first xy graph.

 

 

altenbach_0-1637867174857.png

0 Kudos
Message 9 of 14
(337 Views)

Altenbach used the technique that once a wise scholar said  

 

"Complex" just isn't "Real", it is also "Imaginary" (at the same time ...).

- Bob Schor

-Santhosh
Semiconductor Validation & Production Test
Soliton Technologies
NI CLD, CTD
LabVIEW + TestStand + TestStand Semiconductor Module (2013 - 2020)
NI STS for Mixed signal and RF

New to the forum? Please read community guidelines and how to ask smart questions
0 Kudos
Message 10 of 14
(315 Views)