Thread Tools
Jul 31, 2017, 06:20 PM
Team USA F3F Member
ttraver's Avatar
So we tried to get things working at White Point yesterday for another contest.

Set it all up and it was ok, but then the clouds started to cause some serious problems as they were in the frame with the Sun directly behind and moving quite a lot. I fiddled with it quite a bit and we tried to start the contest, but there was just too much going on to be able to get it working in those conditions.

I'll try and look at video tonight, and share it with you Alvaro.

We couldn't spend valuable practice time in great conditions to mess with it, so we went manual buzzer guys for the contest...

Sign up now
to remove ads between posts
Aug 01, 2017, 02:49 AM
Team USA F3F Member
ttraver's Avatar
Alvaro,

Ok, I have my F3B Speed practice video, and I have annotated just about every flight and shown notes on every detection in the included excel file for the video that we had set up at base B of our speed practice day.

It was a hot day, and bugs were definitely an issue. In this case, the course pointed away from the sun for the most part, although it came into frame later. I tried to manually change the exposure to deal with that...

Here is the full video of the two cameras :

F3B Practice 7-29-2017 full Motcam Video (0 min 0 sec)


I had to convert the file to a PDF to load it up here, but the included file has all of the detections with my notes about them...

Funny, at the end of the video, and big fly lands on the lens and sits there for a few minutes cleaning himself...

Tim
Aug 01, 2017, 03:19 AM
Registered User
Alvaro Silgado's Avatar
Thread OP
Quote:
Originally Posted by ttraver
Alvaro,

Ok, I have my F3B Speed practice video, and I have annotated just about every flight and shown notes on every detection in the included excel file for the video that we had set up at base B of our speed practice day.

It was a hot day, and bugs were definitely an issue. In this case, the course pointed away from the sun for the most part, although it came into frame later. I tried to manually change the exposure to deal with that...

Here is the full video of the two cameras :

https://youtu.be/qBVBTUSOti0

I had to convert the file to a PDF to load it up here, but the included file has all of the detections with my notes about them...

Funny, at the end of the video, and big fly lands on the lens and sits there for a few minutes cleaning himself...

Tim
Wow! That's a great work Tim. Thanks a lot for taking your time to prepare it!

I'll see the video later, now it's still being processed. Thank you very much for sharing it.


Alvaro.
Aug 01, 2017, 03:20 AM
Registered User
Alvaro Silgado's Avatar
Thread OP
Quote:
Originally Posted by ttraver
So we tried to get things working at White Point yesterday for another contest.

Set it all up and it was ok, but then the clouds started to cause some serious problems as they were in the frame with the Sun directly behind and moving quite a lot. I fiddled with it quite a bit and we tried to start the contest, but there was just too much going on to be able to get it working in those conditions.

I'll try and look at video tonight, and share it with you Alvaro.

We couldn't spend valuable practice time in great conditions to mess with it, so we went manual buzzer guys for the contest...

I'm sorry to hear that. Yes please, send me the video so I can try to find a good setting or a fix for this situation.

Thanks a lot,

Alvaro.
Aug 01, 2017, 11:27 PM
Team USA F3F Member
ttraver's Avatar
Alvaro,

I sent you links to the raw videos in your email.

I mostly had issues with the Sun being in the frames shining through the clouds. Difficult situation for the settings...

Clouds and water are not our friends in this situation...

Tim
Aug 04, 2017, 11:02 AM
Registered User
Alvaro Silgado's Avatar
Thread OP
Quote:
Originally Posted by ttraver
Alvaro,

I sent you links to the raw videos in your email.

I mostly had issues with the Sun being in the frames shining through the clouds. Difficult situation for the settings...

Clouds and water are not our friends in this situation...

Tim
Hi Tim,

Wi-Fi where I am now is really poor. I will see them on Sunday. On the meantime, I have released a new version with special features to avoid clouds and waves false detections: a new processing algorithm more optimized (removing "ghost" objects) and a new "Force trajectories to start from border" function allowing to start from upper and lower borders.

I suggest you to use this function in case of waves or fast moving clouds.

Thank you very much for all the information provided!

Alvaro
Aug 13, 2017, 07:36 AM
Registered User
concreteman's Avatar
Awesome session with the gear yesterday. Hours of flying with flawless accuracy. Thanks to all you guys involved. A real game changer
Aug 13, 2017, 12:15 PM
Registered User
Alvaro Silgado's Avatar
Thread OP
Quote:
Originally Posted by concreteman
Awesome session with the gear yesterday. Hours of flying with flawless accuracy. Thanks to all you guys involved. A real game changer
Thanks a lot for you feedback, Warren!

Best regards,

Alvaro.
Aug 13, 2017, 09:36 PM
Team USA F3F Member
ttraver's Avatar
Hey Alvaro!

We had another good practice session today! I tried reducing the size to 1%, and that helped a little bit.

I think for this application we are probably the most difficult situation. Sun in our face, waves, clouds, birds...

I think the trick is getting the Gain and Exposure settings right, and that actually gets rid of a lot of the waves and small stuff...

I would say today we maybe missed about 5-10 beeps, out of maybe 500, which is pretty darn good. We totally would have been able to run a contest with that today...

I'm having a little issue with the wireless unit talking back to your software to tell it which detection direction, so I had to run base A with both directions turned on again. Not sure why that is happening. I'll talk with you and Simon off line about it.

Also, today was the second time when we went to go on the first run, and base B wasn't initially detecting. I would then go over there, and the screen would say that it was at 0 frames/sec processing. I restart the program, and then off it goes from there without any issues. Could it have something to do with the signals its getting from Simon's system?

After that, it was mostly flawless all day!!!

Tim
Aug 14, 2017, 01:40 PM
Registered User
concreteman's Avatar
yep, after a kick in sack it obeyed!
Sep 10, 2017, 10:24 PM
Registered User
concreteman's Avatar
Pleased to announce we ran a full event with the motcam, with close to 0 interruptions. I will let tech Tim elaborate, but it was totally awesome. Thanks guys!
Sep 10, 2017, 11:19 PM
Team USA F3F Member
ttraver's Avatar
Yes!!!

We had a great contest today with 9 pilots, and we used the MotCam system all day.

I would say that we probably only had 10 or so missed beeps out of roughly 1000 beeps, which puts it at 99% effective.

Some of the misses were because pilots flew under the camera, and I think there were a couple of misses due to the sun coming into the frame as we progressed. A quick break to adjust the exposure, and we were going again.

I used the grayscale setting, because we really don't need color, although it might make it a little more difficult if we are trying to identify planes in the videos, as they all have the same stripe patterns. I figured having the higher frame processing rate was worth it.

I was able to get the detection direction settings to work with Simon's wireless system this time, and that worked out all good.

When I get some time, I will look at the videos mostly to verify that a few times pilots actually cut when they think they didn't. They are quick to believe that the system missed them rather than believe that they never got to the base...

Working great Alvaro! (and Simon!)

Tim
Sep 11, 2017, 04:30 AM
Registered User
Alvaro Silgado's Avatar
Thread OP
Quote:
Originally Posted by ttraver
Hey Alvaro!

We had another good practice session today! I tried reducing the size to 1%, and that helped a little bit.

I think for this application we are probably the most difficult situation. Sun in our face, waves, clouds, birds...

I think the trick is getting the Gain and Exposure settings right, and that actually gets rid of a lot of the waves and small stuff...

I would say today we maybe missed about 5-10 beeps, out of maybe 500, which is pretty darn good. We totally would have been able to run a contest with that today...

I'm having a little issue with the wireless unit talking back to your software to tell it which detection direction, so I had to run base A with both directions turned on again. Not sure why that is happening. I'll talk with you and Simon off line about it.

Also, today was the second time when we went to go on the first run, and base B wasn't initially detecting. I would then go over there, and the screen would say that it was at 0 frames/sec processing. I restart the program, and then off it goes from there without any issues. Could it have something to do with the signals its getting from Simon's system?

After that, it was mostly flawless all day!!!

Tim
Ooops! I missed this post!

For getting the serial port in to manage the detection direction, you must actívate the option "Manage direction through serial port", on the "Serial" tab.

If the screen was showing 0 FPS you probably had a disconnection from the usb. Sometimes it happens just moving the wire a bit, when the connector is not very good. Restarting the program will fix it.

I'm glad you had a good time with it, thanks for sharing!

Cheers,


Alvaro.
Sep 11, 2017, 04:31 AM
Registered User
Alvaro Silgado's Avatar
Thread OP
Quote:
Originally Posted by concreteman
Pleased to announce we ran a full event with the motcam, with close to 0 interruptions. I will let tech Tim elaborate, but it was totally awesome. Thanks guys!
Wooow! Thanks for sharing, Warren! I'm pleased you enjoyed it!

Best regards,


Alvaro.
Sep 11, 2017, 04:33 AM
Registered User
Alvaro Silgado's Avatar
Thread OP
Quote:
Originally Posted by ttraver
Yes!!!

We had a great contest today with 9 pilots, and we used the MotCam system all day.

I would say that we probably only had 10 or so missed beeps out of roughly 1000 beeps, which puts it at 99% effective.

Some of the misses were because pilots flew under the camera, and I think there were a couple of misses due to the sun coming into the frame as we progressed. A quick break to adjust the exposure, and we were going again.

I used the grayscale setting, because we really don't need color, although it might make it a little more difficult if we are trying to identify planes in the videos, as they all have the same stripe patterns. I figured having the higher frame processing rate was worth it.

I was able to get the detection direction settings to work with Simon's wireless system this time, and that worked out all good.

When I get some time, I will look at the videos mostly to verify that a few times pilots actually cut when they think they didn't. They are quick to believe that the system missed them rather than believe that they never got to the base...

Working great Alvaro! (and Simon!)

Tim
That's great to know, Tim!!! Thank you VERY much for sharing it!

Cheers,


Alvaro.
Last edited by Alvaro Silgado; Sep 11, 2017 at 04:40 AM.


Quick Reply
Message:

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion F3X Wireless Timing System by Simon Watts - F3F and F3B David Olson Multi Task F3X 196 Jun 16, 2022 12:58 AM
New Product GW008 Mini Skull nano with automatic paralell system SeByDocKy Micro Multirotor Drones 193 Oct 28, 2016 12:06 AM
Discussion Speed measurement system Lussarn High Performance 20 Apr 16, 2016 03:07 PM
Discussion GW008 Mini Skull 2.4G 4CH 6Axis Automatic Parallel System 3D Rolling RC Quadcopter GeekBuying Mini Multirotor Drones 1 Oct 06, 2015 01:05 PM
Post Ratio Automatic System MattLarson29 Life, The Universe, and Politics 21 Jul 19, 2004 01:13 PM