Overthinker

Shock, I subtracted your IR off image from your IR full image to make a better comparison. Additional light sources can really change your perception of just what is being illuminated by what. This is why I took the eyeball cam to a quite dark place with no additional direct lighting. Overall your PTZ does a really good job. Now I want one. You can still see the center IR hot spot with the IR falling off at the edges of the field. Nothing though I couldn't live with.

Shock, Yeah I hear you on the spiders. I finally took a pad painter and used it to paint a insecticide circle around the camera. Now that slowed them down. I still get a occasional moth though. I have seen some of those PTZs with the IR emitters They are using 1 or 3 watt surface mount IR LEDs They were smart though in that they used emitters with different dispersion angles and turn on the emitters that are the most beneficial to the current zoom ratio of the PTZ. Clever! I have used a PTZ in the front of my garage for about 14 years now. Although nothing like todays models I still find it useful. Mine is only a 21x zoom. But not bad for 1998, What zoom ratio is your PTZ and who makes it if I may ask?

Since many people wanted to see the before and after of the modified eyeball camera I have made this new post. For anyone who is not familiar with the modified eyeball camera please refer to the post "Pimp my eyeball cam" Although I really do prefer low light video to IR illumination there just are some environments that are too dark for even the most sensitive cameras. Since I wanted to provide a fair comparison I took the modified eyeball camera to a location that has no direct artificial lighting. This way there are no streetlights of unknown type, unknown brightness, and at unknown distances to influence the images. And since I want other people to be able to see for themselves just how dark it was where I captured the images I also took an image with a DSLR. I used a DSLR because security cameras have no fixed standard of brightness output. Yes, some manufacturers may claim some finite lux ratings but the optimism of product performance varies between manufacturers. DSLR manufacturers calibrate the outputs of their DSLR's by matching sensor output to ISO settings. Any DSLR at a explicit shutter speed and F/stop will produce very closely the same image brightness. White balance and noise levels may vary but the image intensity is very close. This way anyone anywhere in the world can compare their night scene brightness to mine or to anyone else. To compare your night time site brightness to the one used here for testing set your DSLR to ISO 6400,F/3.5,and 30 second shutter speed. DSLR sensors are very linear devices. If your exposure is half as long to make as bright of an image as shown here then your site is twice as bright. The eyeball camera used for imaging is an exact replica to the camera in the previous post. The only inconsequential changes are that I tapped one of the mounting holes to 1/4-20 and I wire tied the cables for strain relief. I mounted the eyeball camera on a tripod extended to 7 and one half feet to better replicate mounting it on a building. Since I modified this camera to provide better imaging of my backyard I set up the eyeball camera in a large field with a warming house in the distance. The warming house is 55 feet from the camera. Trees run parallel on the right to the warming house and then across the horizontal axis of the image. The eyeball camera itself was used at its default settings. The images were captured with an 8 channel Dahua DVR at max quality and 14 FPS. I set the camera zoom to 2.8 mm to create the worst case scenario for IR scene coverage. This location had over 4 inches of rain since the beginning of May so the ground vegetation was very green and a good reflector of IR.

Hi Vector18, I am glad you like it. There are a lot of people on here so if just one person in ten would share one of their ideas... that could be a lot of good ideas. The camera with the emitters came to about $80. The camera was $67 and the emitters were $4 a piece on Ebay. I have seen emitters come and go on Ebay so I finally took the bait. The rest of the hardware was leftovers from other projects so it is hard to place a cost on them but I would suspect that the $80 total would cover them. I am not sure whom it was that manufactured the emitters. They had no labeling of any kind when they arrived. From looking into the front of the TO encapsulation it looks like they are a 45 degree dispersion angle. They didn't have a really harsh hot spot so I only set them back 15 degrees. Anymore and some of the emitter IR would fall outside of my camera field of view. Many people seem to want to see the before and after results. Since I replaced the old camera at the same time I modded the new one I currently have no valid comparison stills. Just playing with the new camera and emitters in the dark with a monitor convinced me it was worth doing. As I mentioned I am making another camera with emitters (exact to the previous) so before I mount it to the soffit I will take some images with just the camera IR and with the supplemental emitters. It will be a lot easier to play around with the unmounted camera for comparisons than one that is already mounted 14 feet off the ground.

Have u ever use or try camera with so called "smart IR" ? Yes I have used cameras with smart IR. However the smart IR works even better when you can aim the IR in multiple dimensions to even out the intensity differences.

First many thanks to all for the kudos. Okay some answers to questions in reverse order ak357: LOL yeah that looks like a real eyeball burner. However they all seem to suffer the same limitation. And that is even on the most flexible designs you can only adjust the IR emitters in one plane. Although on mine rotation is fixed after being optimized for an application I still have vertical adjustment... to say better illuminate an upper deck with one and the entrance to the deck access stairs with the other. thewireguys: I typically do not show video from my system for security sake. The internet is forever. However I am making another camera assembly and I will take some with and without stills with metrics on a generic testing grounds when finished. shockwave199: The camera and IR emmiters are powered by a single input. A camera plug is connected to the IR emitter power cable. So the line power feed first goes to the IR power cord. Inside one of the IR housings a second male 5.5 x 2.1 comes back out of the IR emitter housing, through the 5/16ths stud ,and back into the camera bell. This line powers the camera. All of the power cabling is internal so it is either up inside the soffit or running inside the 5/16ths studs when installed. This way no cabling is is exposed to vandals. The power source is a adjustable voltage switching supply located about 60 feet away. I read 12.12 volts at the power connection with a total power consumption of 7.86 watts.

Most of us are familiar with the eyeball camera or as some call it the turret camera. I prefer to save turret for the cameras that actually protrude beyond the socket mount but that is just me. The form factor of the eyeball camera is by far my favorite. With a metal enclosure, glass windows, compact size, and no over hung loads it does deserve the designation of vandal resistant. In fact mine have taken some blows even with a baseball bat and still functioned showing the perpetrator. Certainly one of the short comings of most eyeball cameras is the rather weak and narrow IR output. And yes I know that most people, myself included, prefer seperate camera and IR sources as long as the dichotomy doesn't get too bad. However sometimes that is just not practical or even possible. As a real world example take my home. In the area most advantageous to place and wire an outside IR illuminator, I would need to go squiding in a vaulted ceiling attic filled up to nearly the roof with shredded fiberglass insulation. Not an inviting prospect. So for a few dollars I bought some additional IR illuminators on Ebay and mounted them to the bell housing of the eyeball camera. Since I wanted to maintain at least some of the vandal resistance I drilled a hole in the sided of the illuminator casings and mounted the IR illuminators to the bell housing with 5/16ths cap screws. I drilled out the center of the cap screws so the power wires could run internally.The old hole for the IR power cord came out the back of the units so I filled them with roofing caulk. I have seen some bullet cameras that use additional IR illuminators but they always mount them rotationally parallel to the camera optical axis. On this camera I have it set for about a 60 degree field of view so I angled my illuminators back 15 degrees from the camera optical axis to spread out the IR hot spots instead of them all falling in the same area. With the unit assembled it is even easier to mount to the soffit than before since the retaining ring for the eyeball bell housing is now held captive by the IR mounting studs. I also cut a foam spacer to keep the vacant area inside of the bell housing from becoming a boxelder bug nirvana. Do I think this modified camera could withstand a hit from a baseball bat? No. But this one is mounted 14 feet off the ground and even a big wet snowball will not affect it. Now with the camera field more evenly illuminated motion detection is no longer influenced nearly as much by moths, june bugs, and cottonwood fuzz. It was a fun little project. Far more fun than squiding through shredded fiberglass. And it makes a compact solution to my particular security imaging problem.

I would certainly stay way from IR for babies. I have read the Osram eye hazard PDF and although it is a good attempt it fails to include important factors such as a babies iris size and the effects of heat production by chemical reduction. A 7mm iris size is average for a dark adapted adult eye. A baby and young children can have iris diameters of over 9 mm when dark adapted. This change in iris diameter nearly doubles the energy over what an adult would receive. Your retina is very sensitive to heat. We have all felt the cringe of pain when our eyes that are used to the dark are suddenly exposed to light. What we are actually feeling is the heat resulting from the reaction of accumulated rhodopsin being chemically reduced from the light exposure. And although rhodopsin peaks in sensitivity to light in the blue/green part of the spectrum it is still photo reactive all the way out to the near infrared. Why doesn't the sidewalk on a hot summer day burn our retinas? Because the hot sidewalk radiates at a wavelength greater than 8 microns. More than ten times longer in wavelength than the IR used in "night vision" cameras. At 8 microns and above your corneas are about as transparent as concrete. So when you consider that young children produce much more rhodopsin than adults, have larger iris sizes, tend to stare at bright or shiny objects for long periods, and are unable to articulate what is happening to them, directly visible night vision IR sources are inappropriate for children.

Does anyone have any experience with the Samsung 2080 and PIR controlled flood lights? More explicitly does the camera exit ICR BW mode when the floods come on? If so how long does it take for the exposure to normalize? I am using a Samsung box camera without ICR now and it will adjust exposure from frame integration to normal video in a fractional second after the the floods come on. Many thanks to all who reply.

Hi Mike. The 4mm was just an example. Very short focal length lenses at high f/ratios suffer from multiple diffraction effects. Actually the F/ratio of the lens is the deciding factor. Of course faster lenses are both harder to produce and design with limited lens aberrations. A typical 1/3 Sony Super HAD II CCD has pixels about 4.6 microns in size. The diameter of the diffraction limited spot size of a perfect optical system is given by 2.44 X wavelength of light used X the lens F/stop. So lets say you have your 8-80 set at 8mm or 80mm it doesn't matter, if the iris stops down more than about f/3.5 the diffraction spot size of the lens will be bigger than the pixels on your CCD sensor. That is if your lens was optically perfect and you were viewing a daylight scene. ( 2.44 x .55 X 3.5 = 4.697microns. Say you used the lens at night with typical IR illumination on... then your perfect lens would start to lose sharpness due to diffraction at about f/2.2. Again this is for a perfect lens. The lens aberrations of most lenses hide this phenomenon to some extent. Now think of my previous f/32 example. Using the diffraction formula the diffraction limited spot size would be 43 microns. Almost 10 times your pixel diameter! So even with an IR corrected lens the resolution of your camera can never equal day light performance since you are using a longer wavelength of light. The ND star helps this problem by keeping the F/ratio from getting too high. Like when someone is close to your camera and IR source and the iris needs to close to maintain the correct exposure. Since the star is maybe 10% of the lens area you lose a little less than that since the star is not completely opaque. I know the 10% seems like a lot however there are many other light losses in the camera system. For a single example you lose 4% at each surface of the glass window protecting your camera. And that does nothing to improve the performance of your camera system. I tend to personally like IR corrected aspheric lenses. Since they really do provide sharper images at wide f/ratios. Also since they are hyperapochromatic they work better in visible light too. Fujinon's IR corrected lenses are easy to spot. The Fujinon name is in red lettering.

Mike Va, Some lenses, particularly short focal length ones, may reach their diffraction limit before they could stop down enough to provide brightness control. If you think of a lens like a 4mm and say it needed to stop down to an f/32 aperture to provide an image that was not overexposed. The actual lens diameter used would only be an eighth of a millimeter. At such a small lens diameter the smallest features that could be reproduced on the CCD would be many times larger than the pixels on the same CCD. The image would not look sharp. With advent of 700 TVL cameras some of the more expensive lens manufacturers use a neutral density filter that moves into place when the camera would have needed to close the lens iris more than diffraction limited resolution would allow. This is done to maintain sharpness. The neutral density filter is not used when the light levels are low. Much like a the ICR of many modern cameras. Low cost lenses use a small neutral density "star" pattern in the center of the lens. This type does reduce the light throughput but the amount is small since the "star pattern is only a small percentage of the total lens area.