Expecting to much

[mat200]

@Dsmtweaker

Hey, I do not see the Honda fit in the lot from the camera at all ... not certain at how you can compare the images when they are not taken from the same angle to the license plate...

 

[Dsmtweaker]

HI @Dsmtweaker

OK - let's look at your specs on that Nikon 3100:

1. 
   
   Effective Pixels (Megapixels)
   
   14.2 million
2. 
   
   Sensor Size
   
   23.1 mm
   x 15.4 mm

( that's a diagonal of about 27.7mm )

SO 14MP, and a big sensor.. and a big lens... ( as J Sigmo and Fenderman notes... )

ok, the closest I could find off the Dahua USA site:

12MP True Day/Night IP Box
Model Number: NK8BA4
Series: 12MP, Box, Built-in Microphone, DWDR, Network Cameras, SD Card Slot, Ultra Series, Ultra Series : Network
The NK8BA4 network camera features Smart H.265+ video compression, reducing bandwidth and storage requirements without sacrificing video quality. The multi-pronged approach works by making reductions of useless or unimportant data wherever possible. As a result, high quality video is maintained without straining the network.

These cameras are embedded with an IR cut filter making them suitable for applications with varying lighting conditions and onboard intelligent video system analytics for a more efficient surveillance system. The built-in IVS features people counting and heat mapping functionality, advanced features that add business intelligence to your security system. Intelligent Analysis tracks and stores daily, weekly, monthly, and yearly people flow through a defined scene and produces a configurable report. The system uses this data to produce a Heat Map, a two-dimensional representation of data that provides an immediate visual summary of information. People Counting and Heat Mapping can identify the most active areas of a retail space providing valuable data analysis for commercial applications.

Description Additional information Documents
1/1.7-in. 12 MP Progressive Scan STARVISTM CMOS Sensor
Triple-stream Encoding
H.265 and H.264 Dual Codecs
12MP at 20 fps (4000 x 3000) and 4K at 30 fps (3840 x 2160)
Digital WDR, Day/Night (ICR), 3DNR, AWB, AGC, BLC
Multiple Network Monitoring: Web Viewer, CMS (DSS/PSS) and DMSS
Auto Back Focus (ABF)
Micro SD Memory, PoE
Intelligent Video System with Intelligent Business Analysis
*Lens sold separate

This is what you want.

We're in luck, B&H carries the body.. has it for only $740+ for just the body.. remember to pair it with a lens and a box to protect it from the elements.

Dahua Technology NK8BA4 Ultra Series 12MP DWDR Network Box Camera

View attachment 45140


View attachment 45141

it is a really nice camera... definitely far superior to that cheap-o one you're attempting to use right now... ( hmm, it still have a significantly smaller sensor.. so, you'll actually want to look for a model which has a sensor closer to the one in the DLSR.. )

Alright, but still confused. I am not trying to induce a fight or troll. At this point just learning for myself.

So the camera you linked is closer, but still under the dslr. Also way more costly. A d3100 which is 14.3mp along with a 18mm lens will set you back $400-500. Granted it doesn’t have IR, or weather rating. But putting that in a box and connecting it via hdmi instead of IP would yield a substantially better picture and a fraction of the cost. So why is there such a divide going to security camera?

Network bottleneck shouldn’t matter, as a standard gigabit network could easily handle lossless video at 4K or even 8k.

So why are prices higher for lower equipment(image quality) in this realm? If they used dslr type sensors/lenses, Added in the IR and outdoor functions, it would seem like a camera that would handle wide angle and ranged detail would cost in the $1000/range and out preform the above mentioned avigilon 29mp

 

[mat200]

Alright, but still confused. I am not trying to induce a fight or troll. At this point just learning for myself.

So the camera you linked is closer, but still under the dslr. Also way more costly. A d3100 which is 14.3mp along with a 18mm lens will set you back $400-500. Granted it doesn’t have IR, or weather rating. But putting that in a box and connecting it via hdmi instead of IP would yield a substantially better picture and a fraction of the cost. So why is there such a divide going to security camera?

Network bottleneck shouldn’t matter, as a standard gigabit network could easily handle lossless video at 4K or even 8k.

So why are prices higher for lower equipment(image quality) in this realm? If they used dslr type sensors/lenses, Added in the IR and outdoor functions, it would seem like a camera that would handle wide angle and ranged detail would cost in the $1000/range and out preform the above mentioned avigilon 29mp

You certainly can setup a DSLR on a tripod or similar and set it up to take video.... I suggest buying one and setting it up as a security camera if you think that will be fit your requirements.

 

[Dsmtweaker]

@Dsmtweaker View attachment 45145

Hey, I do not see the Honda fit in the lot from the camera at all ... not certain at how you can compare the images when they are not taken from the same angle to the license plate...

See the yellow circle. This is all the same single shot, not a different angle, not different pictures. I only took one photo, then digitally zoomed in on it from my computer after I pulled it off the camera.

 

[mat200]

See the yellow circle. This is all the same single shot, not a different angle, not different pictures. I only took one photo, then digitally zoomed in on it from my computer after I pulled it off the camera.

So that is a straight on shot of the license plate unlike the others you have pointed to.

All you need to do is buy a Nikon DSLR 3100 and deploy it as a security camera.

 

[mat200]

btw - this is the image you posted.. so you must have been inspecting an image which had a lot more data in it for your still shot that you digitally zoomed in because the license plate data is clearly not visible here in the main picture from your DSLR

 

[Dsmtweaker]

So that is a straight on shot of the license plate unlike the others you have pointed to.

All you need to do is buy a Nikon DSLR 3100 and deploy it as a security camera.

Ok, but take out the license plates and angles. Let’s say it’s a person. If a person was standing at the end of the lot, I could id them with the dslr, and give a very good description. But with the 4K camera, and just under the same resolution. I could just say it’s a person, but probably not even tell if it was a male or female.

That’s where I am confused I guess. For only have a small deference in megapixels and resolution why does it have a hide difference in quality? I mean, the 4K camera is say 10% less res/MB, but 70% less image quality... just doesn’t make sense.

Unfortunately a dslr won’t work, as I need to keep it IP

 

[fenderman]

Ok, but take out the license plates and angles. Let’s say it’s a person. If a person was standing at the end of the lot, I could id them with the dslr, and give a very good description. But with the 4K camera, and just under the same resolution. I could just say it’s a person, but probably not even tell if it was a male or female.

That’s where I am confused I guess. For only have a small deference in megapixels and resolution why does it have a hide difference in quality? I mean, the 4K camera is say 10% less res/MB, but 70% less image quality... just doesn’t make sense.

Unfortunately a dslr won’t work, as I need to keep it IP

did you not read any of the responses explaining compression and lenses? Tell your boss he cant afford cameras that do what he wants.

 

[Dsmtweaker]

btw - this is the image you posted.. so you must have been inspecting an image which had a lot more data in it for your still shot that you digitally zoomed in because the license plate data is clearly not visible here in the main picture from your DSLR

View attachment 45147

I’m not sure I follow, zoomed out(or technically normal) it’s smaller and harder to read?

 

[Walrus]

btw - this is the image you posted.. so you must have been inspecting an image which had a lot more data in it for your still shot that you digitally zoomed in because the license plate data is clearly not visible here in the main picture from your DSLR

The main unzoomed pic from the DSLR is probably uploaded low res.

 

[mat200]

Ok, but take out the license plates and angles. Let’s say it’s a person. If a person was standing at the end of the lot, I could id them with the dslr, and give a very good description. But with the 4K camera, and just under the same resolution. I could just say it’s a person, but probably not even tell if it was a male or female.

That’s where I am confused I guess. For only have a small deference in megapixels and resolution why does it have a hide difference in quality? I mean, the 4K camera is say 10% less res/MB, but 70% less image quality... just doesn’t make sense.

Unfortunately a dslr won’t work, as I need to keep it IP

HI @Dsmtweaker

Are you certain you work in IT???????

I’m not sure I follow, zoomed out(or technically normal) it’s smaller and harder to read?

@Dsmtweaker that is from the image of the scene you posted. I simply cut out the section from the main view of the image you provided. Pixels are as in the jpg without any edits by me.

How large of a image is the principle image you are digitally zooming into? What is the dimensional size? What is the size of the image file?

 

[Dsmtweaker]

did you not read any of the responses explaining compression and lenses? Tell your boss he cant afford cameras that do what he wants.

Yes, as I stated at this point I am trying to learn for myself.

The compression doesn’t seem like it should factor as it’s not needed. Any standard modern network could handle fully raw(lossless) uncompressed 4K video(or even 8k and higher), even at high frame rates(UHD/blue ray ) quality without even noticing. So why compress the quality if there is no need.

 

[Dsmtweaker]

HI @Dsmtweaker

Are you certain you work in IT???????



@Dsmtweaker that is from the image of the scene you posted. I simply cut out the section from the main view of the image you provided. Pixels are as in the jpg without any edits by me.

How large of a image is the principle image you are digitally zooming into? What is the dimensional size? What is the size of the image file?

I will have to check tomorrow, as the image is on my computer at work, and I am currently on my phone. But i just used the “upload a file” button below, so I am not sure if the site shrinks it.

What does the IT dis have to do with anything? I said from the beginning I don’t know security cameras, or even video cameras. I dabble with a dslr for fun. I’m honestly asking questions that I don’t understand from people with more knowledge, so I can better educate myself on how these work and what to expect

 

[mat200]

I will have to check tomorrow, as the image is on my computer at work, and I am currently on my phone. But i just used the “upload a file” button below, so I am not sure if the site shrinks it.

What does the IT dis have to do with anything? I said from the beginning I don’t know security cameras, or even video cameras. I dabble with a dslr for fun. I’m honestly asking questions that I don’t understand from people with more knowledge, so I can better educate myself on how these work and what to expect

@Dsmtweaker

Because I have expectations that IT professionals would be well versed on topics like pixels, image sizes, image formats, compression, dots per inch, basics of image editing program ( which users always like to ask about ), jpg/jpeg vs other image formats, bandwidth, network storage, ... etc...

oh, and as well as the ability to RTFM ( i.e. in the case of this forum - the cliff notes )

 

[fenderman]

Yes, as I stated at this point I am trying to learn for myself.

The compression doesn’t seem like it should factor as it’s not needed. Any standard modern network could handle fully raw(lossless) uncompressed 4K video(or even 8k and higher), even at high frame rates(UHD/blue ray ) quality without even noticing. So why compress the quality if there is no need.

lol how to you intend to store that raw 4k? how would you stream it, even with the best networks with a few cams.

 

[Dsmtweaker]

lol how to you intend to store that raw 4k? how would you stream it, even with the best networks with a few cams.

Why would we stream it? It’s all stored on a NAS, just as our current 6 camera 1080p setup. We have 10 drives in the NAS currently which holds 180days of h264 footage. Moving to 4K would not impact that greatly, and storage is a simple/cheap upgrade


It’s only viewed on site, so internet bandwidth wouldn’t come into play. Uncompressed 4K is around 125mbps, average business network is built on 1000gbps, now many are moving to 1000GBps, so even on a 10 year old gig network(what 99% are). You could stream about 8-10 lossless 60fps 4K camera’s before bandwidth would be full.

 

[Dsmtweaker]

@Dsmtweaker

Because I have expectations that IT professionals would be well versed on topics like pixels, image sizes, image formats, compression, dots per inch, basics of image editing program ( which users always like to ask about ), jpg/jpeg vs other image formats, bandwidth, network storage, ... etc...

oh, and as well as the ability to RTFM ( i.e. in the case of this forum - the cliff notes )

Sorry, “IT” is broad, I work on the networking end. So I don’t deal with help desk type items or deal with formats, pixels, etc. I have an understanding of them, but that’s where my questions come from. Which was why is there a huge loss from a small pixel difference

 

[fenderman]

Why would we stream it? It’s all stored on a NAS, just as our current 6 camera 1080p setup. We have 10 drives in the NAS currently which holds 180days of h264 footage. Moving to 4K would not impact that greatly, and storage is a simple/cheap upgrade


It’s only viewed on site, so internet bandwidth wouldn’t come into play. Uncompressed 4K is around 125mbps, average business network is built on 1000gbps, now many are moving to 1000GBps, so even on a 10 year old gig network(what 99% are). You could stream about 8-10 lossless 60fps 4K camera’s before bandwidth would be full.

You understand that it needs to be streamed from the camera to the nas right?
you are storing compressed h.264 footage. Do you understand how much data flows in uncompressed raw 4k?
No one said anything about the internet.
Its obvious that you are on to something a major breakthrough in ip cameras, something these dumbass engineers developing the technology missed. Time to call the manufactures and sell them this idea.

 

[J Sigmo]

Don't feel bad for not being well-versed in the security camera field. I suspect most of us came into this expecting it to be easier and better.

I'll post a bit more when I get to a real keyboard.

Meanwhile, also don't get discouraged by the responses you get here. This does often take a lot of reading and studying.

Since a lot of the questions people ask initially here have been answered numerous times, folks are often a bit "short" with new people asking these questions, especially if the answers exist in the "cliff notes" and WIKI, etc.

 

[Dsmtweaker]

You understand that it needs to be streamed from the camera to the nas right?
you are storing compressed h.264 footage. Do you understand how much data flows in uncompressed raw 4k?
No one said anything about the internet.
Its obvious that you are on to something a major breakthrough in ip cameras, something these dumbass engineers developing the technology missed. Time to call the manufactures and sell them this idea.

Sorry, streaming to a network admin means online.

Streaming from a camera to a nas is simple, as I stated, 99% of current business networks could easily stream around 8 lossless uncompressed 4K feeds(at once) before using bandwidth. Each camera would require about 125mbps, most networks are gig 10/100/1000, or 1000mbps. So no issue there. Storage also isn’t hard, as the drives can read/write at more than 20 times that speed.

I never said we wouldn’t look into adding storage, but we currently have 10, 20TB optical drives.