jokes

Nice. Only $53,000 each. Are they serious?

This is what I use: It's a first generation tube that does near IR, not far IR (heat) like the FLIR unit. The 19KV is also lots of fun to work with. Image distortion is adjustable from bad to worse. There's also a dead zone in the middle of the image. I have some dark glass covers for my headlights. The big lens offers about 1ft of eye relief at 3ft, making it usable for driving a car or piloting a vessel in the dark. I also have two Russian made 2nd generation night vision monoculars, which are more sensitive, but lack the cool optics of the AN/SAR-4B.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

On a sunny day (Fri, 22 May 2015 18:11:46 -0700) it happened Jeff Liebermann wrote in :

The world is changing:

I am contemplating getting this for 299$:

206 x 156 resolution, like the first webcams or net videos in a stamp. But enough to detect a ship or object.

See the video:

But.. more is coming, with these chips becoming as cheap and as high resolution as normal CCD / CMOS sensors, arrays of thousands of micro bolometers:

This morning I was wondering if I can get thermal.com to release their API so I can write my own soft, Adroid and Java sucks. May even sign a NDA for that.

I do remember a poster here long time ago, Hubs or something like that, making a big IR bolometer with discrete components. He would like this stuff. ;-)

I solved the green smear in the Raspi transparent on screen text software I wrote, After trying PAL (that was worse, these little monitors do not have the real PAL delay line I think, just depend on averaging line by line by the eye), I found the problem is in the 32 bit video mode of the Raspi versus 32 bit on my PC (video card).. As I write directly to X11, I wrote something like this for every pixel: *plotptr++ = red; *plotptr++ = green; *plotptr++ = blue; if(color_depth == 32) *plotptr++ = 0; So I wrote a zero byte, the cards are usually arranged in address jumps of four (something to do with 2 fingers math), buy there are only 3 basic colors. So I expected that fourth address not to be used, and wrote a zero there just to have a defined.. well bit too enthusiastic,

So now it reads: /* plot a dot in fg color, BGR sequence */ *plotptr++ = fr; *plotptr++ = fg; *plotptr++ = fb; if(color_depth == 32) plotptr++;

Seems on Raspi the addressing of the graphics is organized a blue, green, red, red. Whatdoyou say. And by writing a zero to red as the last action anything pure blue or pure green is not affected, but anything that has red in it becomes less red, green if it was yellow as in this case. hehe

So, the question now becomes 'Are mermaids cold blooded', and if not

After a day of coding got AIS display working:

The orange squares are ships on the canal from Harlingen to Leeuwarden.

AIS is an identification system transmitted on VHF. I listen to aisdecoder on a UDP port, and use rtl_sdr to get the data packets, simple vertical antenna. Many ships carry a transmitter, all big ones MUST carry one. So you know where, and heading and speed of all vessels.

Decoding the data was fun, just added the received data to a linked list that is dynamically updated and contains all info.

Since we also know our own position (if GPS works mm), next is proximity alert (easy), and collision alert, this is an interesting math problem, to calculate if 2 or more things will be at the same point at the same time for the given known speed and course.

Good is that we are only in a 2 D plane...

The idea of writing my own automatic airtraffic control alarm software flashed by, but no. I am good at 2 D math, always was, I know why.

I am still at < 50% processor load on the Raspi.

-rwxr-xr-x 1 root root 108497 May 25 09:38 rgpspc

Just past the 100 kB code size limit on the PC :-)

No exotic libraries, just -lm -lpthread -lXaw