This commit is contained in:
Cédric Verstraeten
2017-07-04 16:27:02 +02:00
parent 9e7af3e698
commit e56a907f01
3 changed files with 34 additions and 10 deletions

View File

@@ -8,19 +8,19 @@ One of the strategies Kerberos.io follows is **the concept of isolation**. Read
## IP cameras
This is the most common scenario. Nowadays if you want to buy a camera, there is a 90% chance that it will be an IP camera.
This is the most common scenario. Nowadays if you want to buy a camera, there is a 90% chance that it will be an IP camera.
IP cameras have outnumbered the video camera industry, they come with many different flavours and technologies (e.g. ONVIF). Typically a user will **buy one or more IP cameras**, **position them** at various places at home or work, and **install a video surveillance system** to manage them; e.g. Kerberos.io.
### Desire
What we often see is that users ask for, or desire to have, one single application where they can access and manage all their different IP cameras. This desire may look obvious at first sight, but has **limitations on the long run**.
What we often see is that users ask for, or desire to have, one single application where they can access and manage all their different IP cameras. This desire may look obvious at first sight, but has **limitations on the long run**.
Many video surveillance solutions, other than Kerberos.io, uses the approach previously described. However this approach is against the Kerberos.io strategy/philosophy, read more about it on the [**strategy page**](/strategy) and the [**one vs multi camera support**](/strategy#one-vs-multi) discussion.
### Best practice
When you have an enterprise environment, or just have a lot of IP cameras (>2), you [**should consider to use Docker**](/installation/Multi-camera/Docker). When using **Kerberos.io with Docker**, you create an individual virtual processing units, **a Docker container**, for each camera.
When you have an enterprise environment, or just have a lot of IP cameras (>2), you [**should consider to use Docker**](/installation/Multi-camera/Docker). When using **Kerberos.io with Docker**, you create an individual virtual processing units, **a Docker container**, for each camera.
By using this approach the management of your cameras is transparant and you have the ability to pin point them by an IP address (or port). Next to that you will have **a highly scalable solution**, when new cameras are installed you create new containers for each camera. When processing a lot of cameras you can eventually use **Docker swarm** to load balance containers over multiple systems.
@@ -46,7 +46,7 @@ Just the like the IP camera scenario, users want to have a single application wh
If you consider to connect multiple local cameras to a processing unit (server, Raspberry Pi or other board), you should take into account following points:
* Did you positioned your cameras properly?
* Did you positioned your cameras properly?
* Has the second (or other) camera a lot of value? Maybe can you give it a better position somewhere else.
In most of the cases a second or third camera can be of a lot of value e.g. a 360° view. However in most of the cases this is totally overkill and you should better use it at a completely other position.
In some cases a second or third camera can be of a lot of value e.g. a 360° view, but most of the time this is totally overkill and you should better use it elsewhere.

View File

@@ -1,5 +1,6 @@
# Release notes
* [2.4.0](#2-4-0)
* [2.3.1](#2-3-1)
* [2.3.0](#2-3-0)
* [2.2.1](#2-2-1)
@@ -12,6 +13,29 @@
* [1.0.1](#1-0-1)
* [1.0.0](#1-0-0)
<a name="2-3-1"></a>
## 2.4.0
#### Ecosystem
* Enable/Disable force network mode.
* Enable/Disable auto removal media.
#### Web
* Upgrade from Laravel 4.2 to Laravel 5.4.
* Fix bug with heatmap radius slider.
* Design tweaks toggle button (remove outline).
#### Docker
* Fix SSL support.
#### Machinery
* OpenMAX IL hardware encoding for Raspberry Pi camera (benchmark 30FPS on Raspberry Pi Zero W).
* Basic authentication for live stream.
<a name="2-3-1"></a>
## 2.3.1

View File

@@ -10,30 +10,30 @@ When deploying on the Raspberry Pi there are two possible solutions. You can use
###KiOS
[KiOS](https://github.com/kerberos-io/kios) is a custom linux OS (created by buildroot) which runs Kerberos.io out-of-the-box. This is the **most simple** and **basic** installation procedure. Note that with KiOS you don't have multiple camera support.
[KiOS](https://github.com/kerberos-io/kios) is a custom linux OS (created by Buildroot) which runs Kerberos.io out-of-the-box. This is the **most simple** and **basic** installation procedure. Note that with KiOS you don't have multiple camera support.
<div style="float:right;margin-top:-15px;"><a href="installation/KiOS" style="color:#fff;background-color:#943633;border-radius:7px;padding:8px;">Install KiOS</a></div>
###Raspbian
If you already have a Raspberry Pi running with Raspbian, you probably **don't want to reflash your SD card**. Therefore you can install the different parts of Kerberos.io (machinery and web) manual.
If you already have a Raspberry Pi running with Raspbian, you probably **don't want to reflash your SD card**. Therefore you can install and download the different parts of Kerberos.io (the machinery and the web) without the need for complex and time consuming compiling.
<div style="float:right;margin-top:-15px;"><a href="installation/Raspbian" style="color:#fff;background-color:#943633;border-radius:7px;padding:8px;">Install on Raspbian</a></div>
## Armbian supported boards
If you have <a href="https://www.armbian.com/download/" target="_blank">**an Armbian supported board**</a>, you can follow the Armbian installation to transform it into a surveillance system.
If you have <a target="_blank" href="https://www.armbian.com/download/">*an Armbian supported board**</a>, you can follow the Armbian installation to transform it into a surveillance system.
<div style="float:right;margin-top:-15px;"><a href="installation/Armbian" style="color:#fff;background-color:#943633;border-radius:7px;padding:8px;">Install on Armbian</a></div>
##Docker
A Docker image is available on the Docker Hub for **x86 machines**. By using docker-compose you can run Kerberos.io with a one-liner.
A Docker image is available on the Docker Hub for **x86 machines**. By using docker-compose you can run Kerberos.io with a one-liner.
<div style="float:right;margin-top:-15px;"><a href="installation/Docker" style="color:#fff;background-color:#943633;border-radius:7px;padding:8px;">Run on Docker</a></div>
##Generic
You can install Kerberos.io on any Linux based OS like Ubuntu, Debian, OSX, etc. The only thing you need to do is to compile the machinery and install the web interface.
You can install Kerberos.io on any Linux based OS like Ubuntu, Debian, Raspbian, OSX, etc. The only thing you need to do is to compile the machinery and install the web interface. Kerberos.io can't be compiled on a native Windows machine, if you want to use Kerberos.io on windows you should Docker.
<div style="float:right;margin-top:-15px;"><a href="installation/generic" style="color:#fff;background-color:#943633;border-radius:7px;padding:8px;">Compile yourself</a></div>