With this article I aim to compare the cameras between the existing standard kit of the UKMON network – the KDM-6101G – and the newer digital kit offered by the Global Meteor Network which I’ve been testing – a Sony IMX291 based IP camera.
At this point, I’ve not been comparing the quality of the data analysis for positions/speed/orbits etc, but rather the differences in sensitivity between the two cameras.
To this end, I have had one of each camera pointing at the same area of sky for a period of time and have chosen an almost 3 week period between 27/01/19 and 15/02/19 to compare.
The first thing to compare is that the older camera has a roughly 60 degree field of view. Any more than that, I believe that UFOAnalyzer starts to have trouble due to optical warping.
This new camera has closer to a 90 degree field of view and the GMN software can deal with up to an all sky camera for accurate detections without any issues.
This suggests that for a 360 degree coverage we may only need 4 cameras rather than the current 6 that some in our network have.
I’m still evaluating this and will post more later when I have built my own new camera from scratch, but here’s an indication so far.
The savings in costs are mainly in the computer required, which runs on a Raspberry Pi (so only around £30) and using Linux (free). The camera is also a lot cheaper at only around £30. There is also no UFOCapture license required.
As the new system uses a Raspberry Pi, this has implications on power consumption too. I haven’t calculated it, but this could be significant savings for a multi camera setup.
The takeaway from this is that we could feasibly get 360 degree coverage with the new kit at the same price as one of the old cameras.
Using Stellarium to look at which stars I can see in the stills from both cameras shows that with the old camera, I can see stars only down to about magnitude 3 or just below 4 at a maximum.
The new camera is definitely showing magnitude 5.8 and perhaps up to 6, which would put the new camera somewhere between 6 and 15 times more sensitive! (the specs seem to suggest it’s actually 10 times more sensitive).
Here’s an example below of the old vs the new camera clearly showing Orion in both. Also note that the new camera is a higher resolution too, which is not obvious from these blog images.
Over the period of analysis, this extra sensitivity resulted in almost exactly twice as many meteor detections with the new camera, although on a couple of nights 3 or 4 times as many detections occurred (in total 61 meteors on the old camera vs 121 with the new).
The new camera only missed 6 meteors that the old camera caught. These were all faint meteors, but I would have expected them to have shown up. However I did miss some I think possibly due to user error (I think the detection software terminated at some point on one day).
Everything seen so far suggests this should be an ideal replacement for the existing camera, with the potential to capture way more meteors with its combination of a wider field of view and sensitivity.
Couple this with the promise of also being fully automated and not requiring manual analysis, then it could take things to the next level. More testing on this part to come.
The only downside that I have noticed so far is that if you want to see a video of the meteor event, then the old system still wins. Due to the way that the new system captures frames and compresses and saves just the detected motion, then animation does not look anywhere near as good.
I’ve picked 10 of the side by side detections to display here for your own comparisons, with the old camera view followed by new in each case: