Subscriber Discussion

Advice On My Surveillance Startup Idea

Greetings. Just became a member. Am hoping to receive some feedback on the following issue.

A few days ago, I read an interesting article by The Taneja Group about how DVRs are known to fail.

Most people (especially housing complexes and smaller establishments) seem to believe that the story ends at the installation of CCTVs. However, the article I am referring to, clearly brings out the fact that DVRs cannot be relied upon completely. 'More' needs to be done; and this is exactly the point that is ignored. I would like to present 4 simple points to achieve this:

1. Take Safety Seriously, and as a continuous process.
2. Develop Processes.
3. Train Staff.
4. Make Best Use of Technology. Keep it simple though.

Special reference needs to be made about the second point. Processes can be as simple as:

1. Review earlier day's footage. It may be time consuming, and needs manual intervention, but, it's better to be safe than sorry.
2. Capture images of relevant footage and save with comments/tags.
3. Share the above with concerned authorities internally.
4. Save above to the cloud for longer retention where needed.

We are aware that most DVRs/Services will overwrite data after some days/weeks/months. This is where crucial data could be lost.

There is a simple method to automate the above processes, and avoid the painful task of copy/paste video footage. I would be happy to share more information about this method.

The article by Taneja Group is here.

Thank you for reading this.

Gautam D Goradia

Gautam, welcome!

The Tenaja group 'article' is 5 year old promotional play by the now out of business Intransa Corporation.

No electronics can be 'relied upon completely.' They will die at some point, just like companies go out of business.

Are you personally having problems with DVRs failing or? I am trying to understand how we can best provide you feedback.

Thanks for the welcome John, and for throwing light on the promotional article. No, I do not have any personal experiences with DVRs failing, but, this bit about 'failing' has been observed, even by people who supply and install CCTV systems.

Having said that, I'd like your opinion on whether there is no other way to capture images from a video footage, other than Print Screen, and paste them to, for example a Word document. Now, I am aware that some systems do offer some kind of automated print screen, but, these are the more expensive ones, I guess. I've never used these, so, I do not know how they work, and whether they can add meta data to the images.

I am working on something that helps capture images from video footage very efficiently (no more copy paste), aggregate them, tag them, auto-create office docs, or html files, and upload them to the cloud. Since I do not want to sound promoting my work, I would be happy to send you more information about this separately.

Doing print screens and pasting them to Word documents or embedding them into some other format is indeed a convuluted procoess, but at the same time usually an unnecessary one as every system I've seen has some way of exporting an image straight to a JPG, PNG or BMP file. And every DVR from the past 5 years or so has some way to export video as a common AVI or MOV file. Is there a particular brand you're coming across that does not do this? If so it's a rare one and not that common, or maybe a very "old" model.

As for your process for housing establishments and smaller businesses, it's a common one that they rarely ever check their systems after they are installed unless an incident occurs. So if a problem with the system occurs, like a camera going out, they often don't notice it until they go looking for video. No matte rhow many times you advise them to check their systems intermittently, the majority don't. Cloud based recording is a developing process, but limited by available Internet bandwidths and customer unrealistic expectations for performance.

Thank you Luis. My interest emanates from the fact that we're working on enabling anyone who installs a cctv, to effortlessly capture relevant images from a video (one click), aggregate them, tag them, edit them (borders, captions, arrows and so on...), and with the click of a button, automatically create a PowerPoint/Word/OneNote/PDF document/Web Page (no copy/paste, or even adding a text box in a PPT slide - the software does it all). I believe that when there is an incident, the affected party can also be of great help to law enforcement. So, instead of just handing over the video footage to law enforcement, if the affected party could also take part in the process, the results could be better and faster.

I stand corrected; but, DVRs that export images to JPG and so on, do not do anything more than that (they are not supposed to) - as in, nothing of what I have mentioned as above. And, I agree with your comments about housing establishments and small businesses not paying adequate attention. But, I am hoping that with simple to use tools, sufficient free storage available on sites like Dropbox, Skydrive, Google Drive and so on (the software will send the images to user-selected sites), coupled with the fact that there's never been a need for cctvs as it is today, people will be more pro-active and capture and tag relevant footage from time to time.

I am more than happy to share my work with you in the next few weeks, if you like.


Gautam, DVRs do fail but there's no 'nightmare' nor 'crisis'. This is a theme pushed forward by companies like Intransa, selling centralized storage or VSaaS companies, trying to motivate people to store in the cloud. The problem is the economics of cloud storage (thereby eliminating the DVR) is weak (see our VSaaS ROI analysis).

As for savings images, Luis has already clearly stated what the modern status is for DVRs.

Feel free to explain what you are doing here as others can therefore learn from and provide feedback.

Thanks John. Clearly, I understand that there is no crisis or nightmare. I have just replied to Luis, so, you should have the details of what we are trying to do. Happy to share the work with you, and anyone else; in the next few weeks.

Guatam, I am intrigued by what you are attempting to devise.... It seems like a tall order to pull off all that you mention.

To clarify, you are interested in doing all of those things for 'images', correct? ...and I mean images (JPEG, GIF, etc) vs video clips (AVI, MPEG, etc)?

Hi Marty.

Yes. You are correct - it's, for images that have been grabbed from any video.

To explain further:

(1) You get video footage - from a DVR, or YouTube, for example.

(2) Play it.

(3) While it's playing, you continue to make clicks of frames that are relevant (no pause needed).

(4) Each time you make the click, an image is created and aggregated at the same time in a form - so, let's say, you made 20 clicks - then, you have 20 images waiting for you in the 'aggregator' - no need to go save images in a folder manually - 20 times!

(5) Once you open the form, you have the ability to add batch remarks, or batch edit - all 20 of the aggregated images at one go, or individually.

(6) Now, you want to create, for example, Word, PPT, OneNote, WebPage - go ahead. With one click, all 20 images are sent to either of the formats. Example, if it's PPT you are wanting to create, you can choose how many images you want on one slide, and where the text is to appear (remember you added some remarks to one or more of those images? - that's the text which will appear in the slide) - there is no need to open PPT, drag the images in, or to go through the pain of creating a text box, and's all automated! - One more interesting output is a 'Contact Sheet' - where several images are put together as thumbnails, with the text appearing below them. This makes it very easy to share lots of images without running into tems of pages. Contact Sheets are created in Word.

(7) All the images that have been tagged (remarks=tags), become searchable for future use. So, if you've tagged some image with the name of a person, or a courier company, you can search with those keywords, and it'll bring up images on your system that match the tags.

(8) You can also upload the images to Dropbox, Skydrive, Google Drive...and more... If you share the images with someone else, who is using the same system, he/she would also receive the tags along with the images.

Hope this helps...

"While it's playing, you continue to make clicks of frames that are relevant (no pause needed) ... Each time you make the click, an image is created and aggregated at the same time in a form"

Who in surveillance has a need for this? What would the use case be?

Let's say this is a greatly needed feature that surveillance users would really want. It's something that could be copied fairly easily (at least the selection part, not the cloud management one).

My gut feel is that this is more of a consumer / entertainment / photography application, which might be a winner (or not) but is a poor compliment for most security/surveillance users.

Use Case (1) - Post incident - example theft at a store. Video footage is reviewed, images captured, tagged, quickly converted to html or Office docs and shared across law enforcement. Perhaps, a quicker arrest? If the store does this bit of review itself, it has 2 implications (a) the incident happened that their place - so, they are the best people to tag the images with apt information - they can recognize people faster/better? (b) lesser work for law enforcement - at least they have a starting point

Use Case (2) - As a good house-keeping habit, if images are taken (retained until needed), and tagged, then search becomes easier

More and more housing complexes in India are installing cctvs. I guess that it would be a great tool to play the citizen's part to help law enforcement.

Gautam, are you looking to build your own DVR/recorder?

Btw, see our VMS exporting comparison for how different VMSes handle this.

No John. We are not building any hardware. It's a piece of software (Windows for starters) that sits on your PC. Give it a video file. As long as you, as the user of the software can view the video, you can do everything that I've mentioned in my earlier messages.

I did look at the comparisons. Great information. Thanks.

If the software is separate, does the user need to export the video first from the recorder?

Yes. The user would give relevant footage to the software for next steps.

I think that's a deal breaker. It really needs to be built in. The whole point of this feature is to make it easy to share video. However, the plan here requires first exporting video, which is a real barrier.

Happy to work with DVR makers if there is interest.

I am sure you would be happy but I doubt it would be mutual. These sort of third party integrations rarely happens. Consider Briefcam who has spent years and millions of dollars trying to add on specialized synopsis / search with very few integrations.

I don't think this is that revolutionary. Verint did this 5 years ago when I worked for them, and I don't think the feature was ever really used...

In my honest opinion, unless it was super cheap, say under $50 a license, I don't think there'd be that much interest. Even then, I haven't found a situation yet where I wished there was something like that.

If your working on making some revolutionary software, how about some middleware piece that facilitates integration between VMS and access control systems that don't have direct integration yet. As ridiculously slow as this industry is coming up with inter-system communication standards, that's something I've been really wishing there was a solution for.

Thanks Luis. The product will be sharply priced. I agree with you about situations not haven arisen yet. If people had a good tool, well priced, did things that would help them quickly, post an incident, I think that there is a need for it. Happy to send you a beta to play around with, in a few weeks.


Thanks Hal, for the feedback. If you'd like to give it a try, I am more than happy to send you a beta to play with.

I am too lazy to read through all of the posts, but I think i get what you are doing. I can kind of see the screen shot aggregating would be somewhat useful to surveillance, but I think it would be more useful to your average guy who likes to take a lot of screenshots on his PC. I cant imagine there isn't another software out there like this though, but I could be wrong.

We use Jing a lot, it takes a screenshot and automatically uploads it to a server, and then it automatically saves the link of the screenshot to your clipboard. Its incredibly efficient and is a great tool for tech support for us.

If a guy wants to save multiple screenshots without having to copy paste and save everytime, then I think you have a good product.

Thanks Sean. I agree Jing is great. But, I think we are far different. I am going to share a link to my presentation on my story and philosophy.

Folks...sharing a story...reported a few days ago... I believe that had the banks had the immediate ability to share MORE than just the video footage with law enforcement, they could have cracked the case faster. Public needs to be proactive, and supply more than just the footage to the bank. The incident occurred at their premises. They are the best party to share the most apt information with the police.

No breakthrough in ATM theft probe

SURAT: No headway has been made in the investigation of withdrawal of Rs one lakh from the bank account of a resident with the help of a cloned debit card. The thefts were made in the city between July 23 and 27 and police were informed on Friday last. But they are yet to make a breakthrough and still awaiting the CCTV footage of the ATMs from where the thefts took place.

Withdrawal of cash from the accounts with cloned debit cards are being reported on a regular basis now. One such incident on an average is reported every 15 days in the city since the first episode in April 2013. The police investigation is dependent on the details and help from the banks concerned. "The unidentified miscreant had withdrawn cash from the account of the resident from five ATMs of different banks. We will have to check the footage of these banks' ATMs. The miscreants had used these ATMs 14 times to withdraw Rs one lakh cash," said M B Vasava, police sub-inspector, Varachha police station. Police said they cannot move ahead in the case without the CCTV footage. Banks have informed police that the CCTV footage will be provided to them from their offices in Mumbai.

Complete Your Surveillance Story... My strong belief is that those who have installed CCTVs should take one small extra step...

Gautam, the primary issue is not the value of the feature, it's the logistical difficulty of getting it embedded into 3rd party existing systems. This is going to be a show stopper. You should talk to the people at Briefcam to understand the challenges here (different technology but same business barriers).

John. I seem to be missing a point, or conveying my idea incorrectly. As I've mentioned earlier, I am not speaking of getting my software embedded into 3rd party systems. I am simply saying "get a video footage from your DVR, and let our software help in taking quick screen shots, adding tags, sending to MS Office, PDF etc". From what I've learnt through companies in this space in India, after an incident, a CCTV footage is handed over to the police. I suppose that this method is used all over the world. My point is, that if the affected party itself takes the screen shots with great ease, adds the needed information to each screen shot, and hands over this information in an MS Office format, along with the video footage, things would be easier for law enforcement. I continue to maintain that the best party to take screen shots, and add relevant information, is the affected party. They would do themselves as well as law enforcement agencies, a great deal of good by being proactive, and taking the screen shots themselves. Where we come into the picture is our software that makes all this very easy. Thanks

Your product is going to be applied to video that has already been exported? Well then your addressable market is going to be very very small. How many people are going to buy such a product? How will they even find out about it?

What you have is an incremental feature that could be a nice enhancement to an incumbent DVR/VMR provider but one that will be very very hard to sell alone.

We have 2 versions - one free, and one at a small yearly fee. Let's take the case of India, and any small city in it. There are thousands of cameras being installed every day. I believe that our free version can very easily build a name for itself as a must-have to (a) help law enforcement, when there is an incident (b) capture images from time to time - this ensures that people are taking care, so as to ensure that things are working. I believe that housing complexes, small and large enterprises, where such cameras are installed, would love to have a piece of software that makes things so easy for them. We will support the free version with advertising. The paid version would have a small yearly fee, based on the number of cameras. Let's take a case of a bank. There are thousands of ATM machines. Each one normally is attached to the premises of a bank. If they have our software sitting there, it would be easy to monitor, and also not lose valuable time in case of an incident. We do intent to tie-up with leading players who undertake the job of installation. That would give us instant adoption and traction. Even the free version does a lot. If you take a look at the deck I shared yesterday, it'll give you an idea. I will share the software in a few days.

John. One more clarification. If a user is viewing the CCTV footage in REAL TIME via the internet, and our software is sitting on that machine, it will capture and aggregate images in real time. There is no need to export any file in this case. I should have mentioned this earlier.


I saw the deck slides and I've read the whole string.

To clarify (again): In order for the end user to tag, classify or notate images to be sent to 'the cloud', the end user would have to export the entire video database (each day), watch it, and while watching be clicking and notating and firing images to the cloud (for each camera)?

If that is the case, I think you are vastly underestimating the time requirements to accomplish this task (every day).

Your logic statement regarding the reasons people should pay attention to their systems more closely is sound. Unfortunately, the vast majority will simply not do it.

Lots of modern systems have added built-in health monitoring sensors that alert when something is wrong because of this learned phenomenon.

I like new ways to look at stuff.... but, from experience, I think your plan relies too heavily on the assumption that end users will do what you think they logically should. :(

I agree John. Majority won't do this. However, my last message should clarify that it is not necessary to download the file. Images can be captured in real time as well. In any case, I think that we still have a case for the affected party to get involved. I was also reading an article about how in several large stores, the people in charge look at cctv in real time, and capture images as and when needed.