Our customers often ask us for specific functions and filters to be included in our software. Paying attention to user requests and managing our development priorities according to them is probably one of the things Amped Software is best known for.
However, not all requests, even if technically feasible, are a match for the purpose of our solutions. Take for example Amped FIVE: some of the most common use cases are enhancing license plates or faces.
Quite a few customers asked for some functions to perform super resolution of faces from a single image. While this may be technically very interesting, most of the implementations have a fatal flaw that prevents them from being used for forensic applications: they introduce data external to the case.
Yesterday a GitHub project called “srez” caught my attention.
Image super-resolution through deep learning. This project uses deep learning to upscale 16×16 images by a 4x factor. The resulting 64×64 images display sharp features that are plausible based on the dataset that was used to train the neural net.
Here’s an random, non cherry-picked, example of what this network can do. From left to right, the first column is the 16×16 input image, the second one is what you would get from a standard bicubic interpolation, the third is the output generated by the neural net, and on the right is the ground truth.
The annual Digital Experience 2016 conference, hosted by our partner DataExpert, will be held this year on Tuesday, September 20 and Wednesday, September 21, at the Veenendaal Hotel, Netherlands.
This conference is organized for digital investigation professionals, fraud analysts, security officers, and IT-mangers, but also for Security Engineers, SOC/CERT-managers, Compliance Officers, Data Protection Officers and anyone interested in digital fraud investigations, Cyber Security, Incident response and/or Data Intelligence & Analysis.
This year, Amped Software is sending our international trainer, David Spreadborough (Spreadys), who will be conducting a workshop titled “Video – Fact or Fiction”, scheduled on Wednesday from 11:45 – 12:45. In this workshop, Spreadys will speak about the importance of correctly acquiring, analyzing and enhancing video and images within a legal environment.
We will also have a booth set up during this event. So stop by to meet the Amped team, to get a personal demo of our products, or to get answers to any of your questions.
We look forward to seeing you soon!
Amped Software heads to Africa to break ground in the emerging economies of Eastern Africa.
SecProTec East Africa is being held at the KICC (Kenyatta International Convention Center) from the 15th to the 17th of September.
If you would like a personal demonstration of our solutions, or if you would like to explore partnership opportunities with us, we welcome you to visit our booth B31 where we will be exhibiting with our South African Partner Risk Diversion.
SecProTec East Africa powered by Intersec is the most important trade fair for the security, safety and protection industry in the entire East African region.
This week I have been delivering a new and exciting course in The Netherlands. At the request of some of our expert users, we designed a new course that focuses on the latest updates to Amped FIVE.
Hosted by our partner, DataExpert, this three-day course has been designed for people who had their original Amped FIVE training over a year ago but want to learn about the new features that were added to the software since then.
Due to our unique and very fast development path, getting a good understanding of the updates, new filters, and increased functionality of the software is vital in today’s fast-paced world.
It seems like only yesterday that I was flying out to Calgary to present image processing training at the 2007 LEVA Training Conference. I was so nervous to present in front of my peers. These many years later, I’m still at it … albeit with a new employer and an entirely new set of tools.
Recently, LEVA announced the schedule for their 2016 Training Conference. Once again, I’m happy to say that I’ll be there as a presenter. I’m also happy to say that I’m bringing a bunch of friends.
Direct Preview in Amped FIVE from the DVR HDD Application, DVR Examiner, now means a seamless workflow between the two products.
Native data recovery from DVR HDD’s can be a tricky business. DVR Examiner, by DME Forensics, can take a lot of this pain away.
When viewing a directly attached (write blocked) HDD, or a forensic disk image, if the file system is supported, all the clips will be viewable.
Right clicking a clip brings up various options. Under Preview With.. you will now see Amped FIVE as a choice.
One of Amped FIVE‘s many input features is having the ability to preview and capture a video signal, directly within the interface.
Any Directshow compatible video device can be used. When selected, the Input Filter settings dialogue box will open, giving you the ability to choose the device and configure its settings.
Although no longer near the top of acquisition options, screen capturing is one of those tasks that when it is needed, it pays to have prepared your capture environment beforehand.
Screen capturing is fraught with difficulties. The main issue now, from a forensic image analysis point of view, is that many players change the encoded data for their display purposes. This can severely affect the video’s ability to be analyzed correctly for restoration or enhancement. As a result, relying on that captured video may be unwise in certain circumstances.
I have written about this effect, and its most common cause in CCTV, many times but I think it’s necessary to take a closer look.
To start with, this is a great little video showing the effect in action….
The two pieces of mesh have the same grid pattern. As the top piece is moved, a pattern appears. This is the Moire.
I see this daily in CCTV Images that have been obtained by taking either a single image, or a video, directly from a television or monitor screen.
OK, sometimes this is required. The immediate publication of some form of imagery is often an ideal way to obtain recognition intelligence at an early stage. Especially if the native, evidential acquisition of the visual material would delay this process.
The evidence can be recovered correctly later, and further images publicized at a later time.
Let’s now look at this image loaded into Amped FIVE… Continue reading
Doing the same thing to a number of files is as simple as Copy & Paste!
It’s a simple function, but one that can dramatically speed up the processing of files in Amped FIVE. It also ensures that any configurable settings are ‘copied’ exactly. No need to write down or screenshot settings – just copy the filter!
In this example, I have three .dav files that I need to process. After dragging the first one into FIVE, the Direct Play dialogue box appears. As these are proprietary streams, they need preparing first as they have no header or index.