Hello! Off and on lurker for years, first time poster. I am a novice programmer and hardware tinkerer hoping to get some input from those of you who are more experienced, knowledgeable, and skilled than I.
I have fiddled with mostly low-cost equipment for personal use for the past decade or so, and in recent months I've started to really dive into the industry. I'm sure that I've only touched the tip of the iceberg of what is available and possible, but I'm getting the impression that the state of the video surveillance industry is kind of a mess. I guess that is to be expected given the rapid pace of advancement, but I still don't feel satisfied with any of the various solutions that I know of.
What follows is my criteria for what I think would be the ideal surveillance system:
1. The camera hard/soft/firmware should be as open as possible. I'm thinking something like a RPi or similar SBC with camera module or usb cam. I know that we can already easily put together this combination to build a DIY IP camera which can be monitored by other software, but...
2. The camera should operate as a node in a cluster, no central server(s) should be necessary for recording or analysis (but still possible of course). Zoneminder is actually already really, really close to being this type of system. ZM will run on RPi, monitor its own local camera, record to its own local storage, and participate as a server in a multi-server configuration. Really the only missing piece is decentralized database, and as a "stretch goal" pooling and redundancy of storage. Also, development of ZM has been very active as of late, and they are making great strides in cpu and memory use. However...
3. The monitoring, configuration, and analysis software should be easy to use. ZM can be a pain to install, configure, use, etc. Don't get me wrong, I love it and it does some things really well, but it's not the most user-friendly piece of software. I know that writing software is hard, and that making it easy to use while retaining the power is even harder. Some of the expensive commercial solutions seem to achieve this, but...
4. Ideally, and IMO, anything related to security should be open source. If we can't see how it works and what it does, how can we know it is secure? And if it's not secure, how can we make it secure if commercial support ends?
So on to the request for input:
First, are there any products or projects out there which address any/all of the points above? If so, I would certainly appreciate some links and info.
Second, would anyone like to discuss or participate in a project which tries to build what I've described? As I mentioned, I am a novice when it comes to hardware and software, but I think the approach should be modular and follow the "UNIX philosophy". Each piece of the puzzle should be useful on its own, but it should also satisfy a defined set of interfaces so that it can be combined with other modules, or replaced with something else.
Maybe this is too ambitious, or maybe it's the wrong approach, or maybe it's an overly-complex solution to a minor problem. But I see systems like Verkada and Meraki which do a lot on-cam and which are managed via the cloud, and I think it would be great to have something similar except open source, locally or remotely managed, and of course much cheaper.
I have fiddled with mostly low-cost equipment for personal use for the past decade or so, and in recent months I've started to really dive into the industry. I'm sure that I've only touched the tip of the iceberg of what is available and possible, but I'm getting the impression that the state of the video surveillance industry is kind of a mess. I guess that is to be expected given the rapid pace of advancement, but I still don't feel satisfied with any of the various solutions that I know of.
What follows is my criteria for what I think would be the ideal surveillance system:
1. The camera hard/soft/firmware should be as open as possible. I'm thinking something like a RPi or similar SBC with camera module or usb cam. I know that we can already easily put together this combination to build a DIY IP camera which can be monitored by other software, but...
2. The camera should operate as a node in a cluster, no central server(s) should be necessary for recording or analysis (but still possible of course). Zoneminder is actually already really, really close to being this type of system. ZM will run on RPi, monitor its own local camera, record to its own local storage, and participate as a server in a multi-server configuration. Really the only missing piece is decentralized database, and as a "stretch goal" pooling and redundancy of storage. Also, development of ZM has been very active as of late, and they are making great strides in cpu and memory use. However...
3. The monitoring, configuration, and analysis software should be easy to use. ZM can be a pain to install, configure, use, etc. Don't get me wrong, I love it and it does some things really well, but it's not the most user-friendly piece of software. I know that writing software is hard, and that making it easy to use while retaining the power is even harder. Some of the expensive commercial solutions seem to achieve this, but...
4. Ideally, and IMO, anything related to security should be open source. If we can't see how it works and what it does, how can we know it is secure? And if it's not secure, how can we make it secure if commercial support ends?
So on to the request for input:
First, are there any products or projects out there which address any/all of the points above? If so, I would certainly appreciate some links and info.
Second, would anyone like to discuss or participate in a project which tries to build what I've described? As I mentioned, I am a novice when it comes to hardware and software, but I think the approach should be modular and follow the "UNIX philosophy". Each piece of the puzzle should be useful on its own, but it should also satisfy a defined set of interfaces so that it can be combined with other modules, or replaced with something else.
Maybe this is too ambitious, or maybe it's the wrong approach, or maybe it's an overly-complex solution to a minor problem. But I see systems like Verkada and Meraki which do a lot on-cam and which are managed via the cloud, and I think it would be great to have something similar except open source, locally or remotely managed, and of course much cheaper.