mroek

the Ambarella SOC is A5S on this model ?you sure ?

what sensor??

I just bought 4 of these from CCTV China and all of them now have the day of the week in English without any fix at all.

 

The cameras came in the OEM browm box too.

Most likely the supplier has patched them manually, but they are still Chinese region cameras.

I don't know if this will help or not...

From the release notes .PDF.

2. Add video analytic features: Face Detection, Traversing Virtual Plane Detection, Intrusion

Detection and Scene Change Detection. But “Face Detection” , “ Traversing Virtual Plane

Detection” and “Intrusion Detection” are mutually exclusive.

You can only use one of these at a time?

Ok, that means it doesn't work in the newer versions of SS either. That's a pity, because it looks like line traversal and intrusion detection might be better than general motion detection. At least in theory.

 

Maybe I'm not configuring it correctly. You have to leave motion detection still enabled still - correct? Or do you turn motion detection off completely and enable one of the new methods instead?

Ok, that means it doesn't work in the newer versions of SS either. That's a pity, because it looks like line traversal and intrusion detection might be better than general motion detection. At least in theory.

Ok, I've done some more testing, and I've concluded that there is no way to have anything other than regular (normal or expert) motion detection to get Surveillance Station to trigger. Line traversal, face detection and intrusion detection will not trigger under any circumstances.

 

That's been my experience too. Nothing I do gets Synology Surveillance Station capturing Traversal, Face Detection or Intrusion Detection.

I think the newer versions (6.3-xxxx) of Surveillance Station has better native support for these cameras, have you tried that? I'm using the PSIA mode in Surveillance Station for mine (with ONVIF I can't have motion detection by the cameras, and the native mode works no better than PSIA).

Two additional observations:

If I set up a shared folder on the NAS, and then connect to this folder using SMB/CIFS (not NFS!) in the camera, the camera will actually respect the quota for the user I set up. In other words, I can set up one unique user per camera, and set a quota for each of those users.

This means the camera will only "see" the quota size for any folder, and it will only "format" this amount of space in the folder.

And another interesting observation is that when using the camera playback function, you can have it overlay the detection features (like line traversal and intrusion detection) and show which one was tripped. This could be helpful when trying to fine-tune the settings. To play back the video without these overlays, just disable "Rules" under the camera local configuration tab.

Ok, I've done some more testing, and I've concluded that there is no way to have anything other than regular (normal or expert) motion detection to get Surveillance Station to trigger. Line traversal, face detection and intrusion detection will not trigger under any circumstances.

I then experiented a bit with setting up NFS on the NAS, and the cameras can connect just fine, but after initializing/"formatting" the NFS drive, it goes back to uninitialized, and it doesn't work. I've since discovered that this is a known bug, and that you have to patch the davinci file to fix it (or have a completely empty volume for the camera). I haven't done so, at least not yet.

Then I did as you suggested, and shared a USB stick on a Windows computer, and set that up for storage on one of the cameras. With this setup I can get both line traversal and intrusion detection to work, and what is more, now it is possible to search the log. My assumption that the log needed external storage to work was therefore correct.

I'm a bit annoyed by this, because it just seems easier to have all this centralized in Surveillance Station, instead of having to scrub each camera for events. I guess the real solution is to move everything to a dedicated computer or NVR, running some other software (like Milestone or something).

On mine the log is empty, even if motion events have been triggered. I suspect this is because I don't use NFS/FTP, it just alerts Surveillance Station to make a recording.

I'm on the latest/new 5.1.6 build, and I have tried resetting the config (except user and ip info).

To use intrusion detection and line traversal, should the normal/regular motion detection be enabled or disabled? Does the "Dynamic analysis for motion" work, so that you can see when the above mentioned features are triggered?

It could be that those features does not want to send the events to Surveillance Station (even though regular motion detection does), and that I'd have to do a NFS setup to get it working, but that would be a bit annoying.

I've updated my cameras to the new 5.1.6, and I tried getting some of the new features to work, but to no avail.

I am using Surveillance Station on a NAS as the recording device, and I can set up motion detection (using expert mode) in the cameras. When the cameras detect motion, they'll signal SS to start recording, which does work.

However, it looks to me like the new intrusion detection and line traversal options (but according to the release notes, these are mutually exclusive) could be a better fit for my needs, but I was unable to get any of those to work.

Are there any special tricks needed to get this working? I think I've tried most combinations of options, but no dice.

Thanks... now where's that CBX critter, I'll have to get one of my cams Mericanized so I can play.

 

I saw him around here somewhere. He looks a bit tired from lack of sleep though

 

This 5.1.6 is different than the old 5.1.6. I'm not sure how though. Some release notes /change logs would be nice

On the US FTP there is now a pdf document containing the release notes for the "new" 5.1.6.

ftp://ftp.hikvisionusa.com/

User: hikfirmware

Pass: Hikvision123

I have also updated my three Chinese cameras (region-changed by CBX) from the web-interface, no problems. I haven't tried any of the new features though, so all I can vouch for is that the actual update was flawless.

I have Chinese versions of the DS-2CD2732F-I(S), DS-2CD2032-I, and DS-2CD2532F-ISW. All have English firmware. I've been playing around with them for a few days trying to get either the facial recognition or the virtual plane traversal to work, and all to no avail. Not too sure if it's the camera not sending it, or the fact that Surveillance Station on the Synology doesn't recognize/support it.

 

I checked the Synology support page, and it says that Alarm support is available for both of those cameras.

 

Any ideas on how I can troubleshoot to determine if it's the camera or the NAS? I would love to be able to cut down on the amount of false recordings I get on motion detection and the idea of virtual plane traversal makes a lot of sense to me.

 

Appreciate any suggestions!

 

Regards,

 

Matt

I haven't tried 5.1.6, but I am using my cameras (older firmware) with Surveillance Station, and motion detection by the cameras, which is working (but it is difficult to avoid a lot of false detections if you also want to catch real action).

I think you need to verify if regular motion detection works first, and if it does, the more advanced detection features *should* also work, since then you know that the cameras are able to signal to Surveillance Station that a recording should be made.

That looks like the output from some third-party patching software, and it is definitely not something Hikvision has put in there. I'd wager a guess that it has been put on there by some Chinese supplier to convert/patch the camera to appear English.

Whether uploading the patched davinci file to this camera will work or not, I can't say.

Sorry if this is a dumb questions... but how do i know if I have the English version of the camera OR the Chinese version (which is also English language).

 

I have 5.1.6 on 1 out of 4 of my cameras... I'd like to upload it to the rest but I'm afraid to upload the wrong one and brick it.

You'll not brick your camera if you try to update it from the web interface. If you use the English 5.1.6 version, and the camera is Chinese, you'll get a message saying something like "language mismatch".

You could also try to turn on the day of week in the OSD settings. If the camera is Chinese, you'll get the name of the day in Chinese letters in the text overlay.

The higher average load might be the reason, but at the same time it would be expected due to the added functionality. In other words, if that's why they pulled it, I'd say they didn't do their homework properly in the first place.

However, if it has no other negative impact, then it should be OK to use this version. What is the forum consensus on this?

Hasn't this version been pulled? It doesn't seem to be available on any of the official Hikvision sites?

Very nice writeup, buellwinkle!

While not being too difficult, it would have been even better if we could change the actual embedded camera region from Chinese to English.

I'm fairly certain that CBX, given his extensive digging into these things, could figure out how to do that, but there is of course a risk of permanent bricking. Perhaps we could do a small crowdfunding to get him a camera to experiment with? I'd be willing to throw in some money for this, provided that he'd be willing to try, of course.

I wouldn't say that the writeup was wasted, but I guess that actually performing all these operations is too involved and difficult for most people. Personally, I haven't bothered, at least not yet. One of the reasons is that the procedure will have to be repeated for each new version of the firmware, and it might (probably) also be different for a new version. That means someone will have to figure out how to patch each new version, and you haven't really documented what you did to find out how to do the patch. In other words, if people apply the current patch, they will depend on you (or someone else with enough time and knowledge) to publish a new patch in order to keep the functionality.

That's why it would be vastly better if it was possible to patch the relevant locations in the flash instead, since you'd (at least in theory) have to apply the patch only once.

I wonder what that open header on the board is used for. Any ideas?

Now posted.

 

Also, just to let you know the language flag also seems to be set in mtd6 also.

 

Specifically,

 

VAR=`dd if=/dev/mtd6ro skip=16 bs=1 count=1 2>/dev/null` ; if [ $VAR == $'\002' ]; then echo Chinese ; elif [ $VAR == $'\001' ]; then echo English ; else echo Unknown; fi

 

As to how to write this to flash, I think I'll keep quiet on that. If you don't know how, you shouldn't be doing it

As far as I know, the mtd layer should handle the necessary operations (erase before write etc) if the flash is accessed through either the block or the character device for the partition.

If this is true, then using dd to modify data within the partition should work. Are you saying that this is not the case here? If so, the only other option I can see would be nanddump and nandwrite.

Am I correct in thinking this is a 'cosmetic' change and will not overcome the issue of upgrading the Chinese camera with EU/US firmware via the browser?

 

Haven't tested that as it wasn'y my goal - doubt it. But it's worth noting that the EU, US and Chinese firmware are all identical so there's no compelling reason to flash with a different version.

That's why it would be even better if the language indicator location(s) in the flash could be changed permanently. I'm quite sure this is fully possible, but experimenting with it is obviously risky.

Yes, it would be better to change the language flag in the flash, and the commands for doing so are already on the camera. Then no need to change davinci.

 

But I won't be doing that as I don't know what the ramifications are. To anyone that wants to try I would advise you nanddump all mtds to a server first and have access to the bootloader via UART just in case.

 

As you can see from the command I posted the language flag looks to be at 0x0654 in /dev/mtd5. Overwriting it with 0x01 may work, not work or brick your camera.

 

I'll stick with my davinci change as that won't kill that camera

Great info, good job!

I can fully understand that you don't want to try modifying the flash partition, I'm also reluctant. My cameras are all installed outside my house, and taking one down to experiment with this is not something I have the time to at the moment. However, if there is some brave soul that would try it, it would certainly be interesting. If there is in fact a simple way to permanently turn a Chinese camera into a US/European one, that would be good news.

I gather it should be possible to change this one byte in the mtd by using the dd-command to write to the flash instead of reading. That's what you meant, right?

When will you publish the required changes to the davinci binary?