Category Archives: New features

Identify Social Media Files with Amped Authenticate

Amped Authenticate Update 10641 introduced the new Social Media Identification filter. It can be found in the File Analysis filter group.

The filters in the File Analysis group are generally looking at the file’s container to return relevant information about the file. The Social Media Identification filter examines the file for traces of information that may indicate the file’s social media source. The key word here is “may.”

The workflow that I will explain here is typical in the US and Canada. Take from it what you need in order to apply it to your country’s legal system.

Let’s begin.

Continue reading

Amped Authenticate Update 10641: Social Media Identification, Griffeye Integration & Many New Filter Options

Hi everyone!

David Spreadborough here, the International Trainer at Amped Software. It’s great to be back writing a blog post. The past few months have been very busy at Amped and our image authentication software, Amped Authenticate, has become the ‘go-to’ tool for many requiring an image’s history or to identify signs of manipulation. Helped maybe by the huge amounts of press detailing fake stories and images.

In this crazy world of multimedia forensics, we cannot stand still. The tech wizards at Amped HQ have been hard at work integrating new filters and tools to assist you further. So, let’s dive in and take a look!

Social Media Identification Filter

Under the File Analysis category, you will find this new filter.

Its purpose is to detect traces in the file formats left on images by social media platforms. As most of you probably already know, it is very easy to save other people’s images from sharing sites. With a simple right-click, you can save the displayed image to your computer.

An image from someone’s Facebook timeline

This filter now enables you to identify if the images you are examining originate from a Social Media Platform.

Continue reading

Amped DVRConv for transcription?

In our most recent update of Amped DVRConv, we added the ability to separate the audio and video streams in your DME files – to save the audio as a separate file. For some, this functionality went unnoticed. For others, it was a huge deal.

Two very specific use cases required this functionality. You asked. We delivered.

Case #1 – Child Exploitation/Human Trafficking

Agencies responsible for investigating cases of child exploitation/human trafficking were spending a lot of time redacting video files (blurring faces and other sensitive information) in order to send files off for audio transcription. The distribution of files in child exploitation cases (files that can be considered child pornography) for transcription is now made a lot easier with DVRConv. All of the evidentiary videos can be loaded into the tool and processed without having to view the footage. DVRConv helps to dramatically speed up the process of getting files to transcription whilst protecting identities and shielding staff from the harmful psychological and legal effects of viewing/distributing such material.

Case #2 – Police Generated Video

Agencies that have deployed body worn/vehicle-based cameras or have interview room recorders often have to send the resulting video files to outside companies for transcription. Like the case above, they are faced with having to redact the visual information prior to releasing the files to their contractor. Even if the agency has chosen a CJIS compliant transcription contractor, they may have agency policies that require the redaction of the visual information prior to release. DVRConv eliminates the need to perform a visual redaction ahead of such a release of files. Having this ability is already saving agencies a tremendous amount of time/money.

Users of DVRConv do not require specialized training. The tool can be used by anyone. It’s drag-drop easy. Plus, the settings can be configured so that the resulting audio file meets the requirements of your transcription vendor.

If you’d like to know more about Amped DVRConv, or any of our other Amped Software products and training options, contact us today.

Amped DVRConv Update 10098: more formats, more speed, more options

Today we released an update to Amped DVRConv, the easiest way to convert videos from proprietary DVR formats.

We have been working on this update for some time and a few users have received beta updates in order to support formats that were urgently required. During this period we have re-engineered a good part of the architecture to improve stability, speed and format compatibility.

Continue reading

The Amped FIVE Assistant Video Tutorial

We recently announced the release of the latest version of Amped FIVE (10039) where we introduced a new operational mode through a panel called the “Assistant”.

The Assistant provides a set of predefined workflows which can be used to automate common operations or guide new users, but it’s not obtrusive. You can use it or not, and you can always add filters or do anything, as usual, it’s just an additional option.

We’ve created a video tutorial so you can see it in action. See below or watch on YouTube now!

We’ll be adding more videos to our YouTube channel soon, so follow us to get more videos like this.

The Sparse Selector

With over 100 filters and tools in Amped FIVE, it’s easy to lose track of which filter does what. A lot of folks pass right by the Sparse Selector, not knowing what it does or how to use it. The simple explanation of the Sparse Selector’s function is that it is a list of frames that are defined by the user. Another way of explaining its use: the Sparse Selector tool outputs multiple frames taken from random user selected positions of an input video.

How would that be helpful, you ask? Oh, it’s plenty helpful. Let me just say, it’s one of my favorite tools in FIVE. Here’s why.

#1. – Setting up a Frame Average

You want to resolve a license plate. You’ve identified 6 frames of interest where the location within the frame has original information that you’re going to frame average to attempt to accomplish your goal. Unfortunately, the frames are not sequentially located within the file. How do you select (easily / fast) only frames 125, 176, 222, 278 314, and 355? The Sparse Selector, that’s how.

Continue reading

Amped FIVE Update 10039: Introducing the Assistant, and much more

I am not exaggerating when I say that this is one of the most important Amped FIVE updates in years!

In this version, we have released a new operational mode in Amped FIVE through a panel called the “Assistant”. The concept is extremely simple but opens a world of possibilities. I had this idea during a meeting in the US about one year ago. I hacked a quick and dirty prototype in a couple of hours and showed it to a few users at the LEVA conference to gather feedback. From then on, we worked to improve it and prepare some scripts.

In the last few years, Amped FIVE has grown like crazy, with more than 100 filters for every kind of issue. It has been adopted by experienced analysts and beginners, and used on cases of local, national and international level.

This raised a few interesting challenges:

  • With so many filters, how are you supposed to know the best tool to use in every case?
  • How do you enforce your agency SOPs for specific needs and workflows?
  • How do you easily help beginners and your new colleagues that are new to the job with the wealth of options available?
  • How do you automate repetitive and boring tasks?
  • How do you avoid human error with repeatable and documented practices?

We had a few ideas for a funny and helpful character.

Or maybe a more awesome classical wizard with its annoying modal interface.

Not really. The solution is much simpler.

Meet the Amped FIVE Assistant.

Continue reading

Experimental validation of Amped Authenticate’s Camera Identification filter

We tested the latest implementation (Build 8782) of PRNU-based Camera Identification and Tampering Localization on a “base dataset” of 10.069 images, coming from 29 devices (listed in the table below). We split the dataset in two:
– Reference set: 1450 images (50 per device) were used for CRP estimation
– Test set: 8619 images were used for testing. On average, each device was tested against approximately 150 matching images and approximately 150 non-matching images.

It is important to understand that, in most cases, we could not control the image creation process. This means that images may have been captured using digital zoom or at resolutions different than the default one, which makes PRNU analysis ineffective. Making use of EXIF metadata, we could filter out these images from the Reference set. However, we chose not to filter out such images from the Test set: we prefer showing results that are closer to real-world cases, rather than tricking the dataset to obtain 100% performance.

Using the above base dataset, we carried out several experiments:
– Experiment 1) testing the system on images “as they are”
– Experiment 2) camera identification in presence or rotation, resize and JPEG re-compression
– Experiment 3) camera identification in presence of cropping, rotation and JPEG re-compression
– Experiment 4) discriminating devices of the same model
– Experiment 5) investigating the impact of the number of images used for CRP computation.

Continue reading

PRNU-based Camera Identification in Amped Authenticate

Source device identification is a key task in digital image investigation. The goal is to link a digital image to the specific device that captured it, just like they do with bullets fired by a specific gun (indeed, image source device identification is also known as “image ballistics”).

The analysis of Photo Response Non-Uniformity (PRNU) noise is considered the prominent approach to accomplish this task. PRNU is a specific kind of noise introduced by the CMOS/CCD sensor of the camera and is considered to be unique to each sensor. Being a multiplicative noise, it cannot be effectively eliminated through internal processing, so it remains hidden in pixels, even after JPEG compression.

In order to test if an image comes from a given camera, first, we need to estimate the Camera Reference Pattern (CRP), characterizing the device. This is done by extracting the PRNU noise from many images captured by the camera and “averaging” it (let’s not dive too deep into the details). The reason for using several images is to get a more reliable estimate of the CRP, since separating PRNU noise from image content is not a trivial task, and we want to retain PRNU noise only.

After the CRP is computed and stored, we can extract the PRNU noise from a test image and “compare” it to the CRP: if the resulting value is over a given threshold, we say the image is compatible with the camera.

Camera identification through PRNU analysis has been part of Amped Authenticate for quite some time. However, many of our users told us that the filter was hard to configure, and results were not easy to interpret. So, since the end of last year, a new implementation of the algorithm was added (Authenticate Build 8782). The new features included:

Advanced image pre-processing during training
In order to lower false alarms probability, we implemented new filtering algorithms to remove artifacts that are not discriminative, something that is common with most digital cameras (e.g., artifacts due to Color Filter Array demosaicking interpolation).

Continue reading

Advanced File Information – A Closer Look!

In the recent Amped FIVE Update (Rev. 9010), there were some big additions to the advanced “File Info”.

The updated tool has already received high praise from regular users who can now do all their frame, stream, hex and format analysis from within the same application – Amped FIVE!

It’s time therefore to take a much closer look at this new and powerful functionality.

Continue reading