Image Restoration: iMovie Stabilization feature

Despite the best of intentions, I have not yet been able to get around to testing Apple's new iMovie video stabilization feature for the Mac OS (Operating System). I will, however, whet you appetite with a Macworld review I've come across that gives examples.

Enjoy!

Image Recognition: Facial and license plate recognition news items

There are two on-line news articles related to image recognition that I would like to bring to your attention.

The first news article is about law enforcement in Tacoma, Washington, USA using a facial recognition package to match 16 years worth of prisoner mug shots with pictures taken by ATM (Automated Teller Machines) in a forgery and theft investigation to generate the lead needed to solve the case.

The second news article is from New Orleans, Louisiana, USA where local law enforcement used a license-plate recognition system to make 20 arrests and recover 23 stolen vehicles and license plates in a 25 day period.

The common thread here is, of course, automatic image recognition. These software algorithms have come a long way in the last decade. However, one should understand that the conditions are partially or completely controlled in both of these applications - i.e. distance, lighting, exposure, aspect (turned toward the camera), and (possibly) compression all fall within acceptable boundaries. In addition, with the license plate recognition problem, the character set and fonts were known in advance. The controlled conditions and a priori knowledge significantly increase the accuracy of the results tremendously and the chances of a successful investigation and prosecution.

(Hat-tip to JUSTNETNews, USA - I highly recommend this free service of the National Law Enforcement and Corrections Technology Center, or NLECTC, which is part of the National Institute of Justice, USA)

Image Forensics: An interview with Hany Farid

SiliconValley has an interview with image forensics researcher Hany Farid. My favorite excerpt was this:

Farid first conceived of the possibilities of digital forensics while at MIT, waiting in the checkout line at the library. Restless, he picked up a copy of the Federal Rules of Evidence, then started randomly reading. What he saw startled him: Film and digital images were equivalent evidence, in the eyes of the court.

"This just seemed like a bad idea,'' he said. "Not that I'm good at seeing the future — but I could tell this would be a problem."

Of course, susceptibility to tampering is not the only thing to be considered when evaluating the relative merits of digital imagery as evidence, but it is obvious that tampering with a digital copy is easier to do and (possibly) get away with than tampering with film or tape.

Enjoy!

CCTV: Work continues on automating CCTV surveillance

Wired News has an article about ongoing research at Ohio State University (USA) in automating surveillance using a network of CCTV cameras. Similar research and product development is occurring at many locations around the world, so I suggest viewing this article as a sample of where things are heading with CCTV technology.

Imaging: Improvements in through-wall radar imaging

StrategyPage has a short article about the two latest versions of through-wall radar imagers (namely the Xaver 400 and RadarVision). These devices allow soldiers and police to see objects, including people, on the other side of a wall made of standard building materials, such as concrete, wood, gypsum board (a.k.a. drywall), and the like, but not metal.

These devices work by emitting an RF (radio frequency) radar pulse, listening for the reflection of that pulse off of objects in the wall and on the other side, and using time-delay (i.e. how long it takes for the reflected pulse to come back) and direction (i.e. which direction did the reflection come from) information to construct a picture. This is similar to how bats and dolphins use acoustic echo location, just with an RF pulse instead of sound pulses.

This is a handy capability to have in hostage and other situations where a team has to enter a building where armed attackers or booby traps might be waiting for them. The engineering problems that had to be solved in order to come up with a deployable system were extremely challenging - my had is off to the teams that did this.

Enjoy!

PS. In case you are interested, I posted some time ago about a DARPA (Defense Advanced Research Projects Agency, US Department of Defense) effort that developed a through-wall motion detector which was compact and lightweight - so light-weight that you use it with a single hand, like a stud-finder. That post can be found here.

Image Recognition: Military requirements push technology forward

The Strategy Page has an article about how the proliferation of video on the battlefield (e.g. from surveillance cameras and even night vision goggles) is driving technology development. Initially, it was the UK (United Kingdom) that drove image recognition technology with its massive deployment of CCTV in public spaces (e.g. train stations, airports, city centers, and shopping malls). Now, it is the military.

The benefit from computer-assistance in analyzing video comes from the following:

1. Volume (i.e. the sheer number of cameras and, hence, images to be monitored and/or analyzed)
2. Concentration (i.e. the human visual system loses the ability to concentrate effectively on images after about 20 minutes of continuous viewing)
3. Memory (i.e. computers can track and, possibly, predict more things simultaneously than a human can because, generally speaking, computers are not nearly as attention and working-memory limited in the short-term as humans and far outstrip us in their ability to recall video sequences over the long term - they can just play back the video recording off of their hard disks)
   

The article claims that the abilities of computer systems to recognize patterns is rapidly improving and approaching that of humans - that is no small feat as humans are simply amazing in their ability to perform real-time pattern analysis.

One intriguing point made in the article is that the current conflicts are generating a lot of real-world recordings of "bad behavior" that can be used to train and test new pattern analysis and prediction algorithms - training data like this is like gold to us signal processing types!

Enjoy!

Image Forensics: Augmenting video restoration & enhancement with a forensic artist

The math, science, and technology used in video forensic filtering and forensic video analysis can work minor, and sometimes major, miracles. However, it is not "the be all and end all", as the saying goes. This article in Forensic Magazine uses an actual case to show where a professional forensic artist was able to use knowledge of human physiology and such to generate a full frontal facial drawing of a suspect from a single frame of a security camera video taken at from the standard elevated position. Let's hear it for the liberal art majors!

Image Recognition: Military uses pattern recognition to detect bomb planting activity

I wanted to call your attention to an article by the Strategy Page on how the US military in Iraq is using pattern recognition on imagery to detect IED (Improvised Explosive Device) emplacement activities.

A very simple implementation of this technique is to take a picture on Day 1, then another picture from the same location and in the same direction on Day 2. Next subtract the two images from each other and see what is left over. All the things that did not change between the two images will disappear and only those things that changed will be left! In this case, what will be see on the "difference image" might be some disturbed gravel, tire tracks off the side of the road (assuming it is a road-side bomb emplacement) or, if the military guys are really lucky, what will be left is a bunch of guys standing around with shovels and a bomb as they work to put it in place. Neat (and helpful)!

Image Forensics: New Best Practices Document for Forensic Video Analysis

The committer members over at SWGIT (Scientific Working Group on Imaging Technology) have been very busy lately. Following the recent release of two new sections for review, they have now added another - Best Practices for Forensic Video Analysis. It and other documents can be found be found by following this link.

Image Forensics: Good Practices

The Scientific Working Group on Imaging Technology (SWGIT) has released drafts of two new "good practices" documents for image forensics, titled "Best Practices for Forensic Photographic Comparison" and "Digital Imaging Technology Issues for the Courts".

These drafts, along with previously released good practices documents from SWGIT and other organizations, can be found on the IAI (International Association of Identification) website at this link.

(Hat Tip: MB)

Forensic Education: Courses for young and old who are interested in forensic audio and video

I've got two announcements regarding forensic education to pass along.

First off, the Department of Justice (USA) is funding Professor Rich Sanders, of the University of Colorado, Denver, to establish a new National Center for Audio/Video Forensics. The full university press release can be found here. It is scheduled to be up and running in the fall of 2009.

Second, the South Carolina Governor's School for Science and Mathematics (USA) runs its popular Summer Science Programs (SSP) every summer. SSP is a residential summer camp for rising 8th, 9th and 10th graders from across the state. The reason I mention it here is that I have been asked to teach two one-week runnings of a new course entitled A Mathematical Tour of Forensic Science. If you are interested, you can find out more on their website.

Update: I am informed by the director of the SSP that students from outside of South Carolina are also very welcome to participate.

Miscellaneous: More links

Working for a living interferes with blogging - Responsibility is such a cruel mistress.

As you can probably figure out from the intro, I have to give you my sincerest apologies, but work is leaving no time for thoughtful comments or analysis, so here are the links to the raw material that I hoped to turn into interesting posts:

Talk to the machine (Strategy Page)

Listening in with the council's lie detectors (Times Online; hat tip to A.F.)

Music Special: Five great auditory illusions (New Scientist)

Astronomy Technology Brings Nanoparticle Probes Into Sharper Focus (Science Daily)

Wanted: RAM's help in solving crimes (New Scientist) (a YouTube video explaining the attack can be found here)

Snakes Locate Prey Through Vibration Waves (Science Daily)

The 'non-lethal' flashlight (New Scientist Blogs)

Car Camera Recorder Pro: Video of Everything in Your Car's Path (TechnoRide)

Abuse of Auto-Tune (Cross Spectrum)

Justice Grants will Fund Research at CU-Denver (Business Journal)

FBI wants palm prints, eye scans, tattoos (CNN)

Enjoy! I hope to be back with more in-depth blogging soon.

Acoustics: Underwater acoustic modem

An Israeli start-up, Sea-Eye Underwater Ltd., has developed an underwater acoustic modem that is fast enough to transmit real-time video. The reported range of the modem is 100-200 meters (1 meter = 1 yard, approximately).

EETimes has the story.

Forensics: Name that codec in one note

Sometimes I feel that I lose track of the founding purpose of this blog - which is to educate people in a non-commercial way about audio and video forensics. In that spirit, I will let the non-specialists among you in on a current trick of the trade - namely, G-Spot. G-Spot is often the first thing a video forensics professional turns to when a piece of digital surveillance video evidence won't play on the laboratory's workstation.

G-Spot is a software application developed by Steve Greengerg to identify what codec is needed to play a video file. Given the plethora of low-cost digital video recorder (DVR) security systems on the market, many, if not most, with their own proprietary codecs, G-Spot can be a life saver. Here is a link to a review of the application (Note: there is some sexual innuendo, but I did not find it to be overly crude).

Now you know what drives many video examiners to distraction every day and what they turn to for the answer.

Speech Recognition: Application to university lectures

Technology Review (an MIT publication) has a comprehensive article on an up-to-date application of speech recognition, including both the recognition engine and the user interface layers - namely, MIT's Lecture Browser website, a search application for video lectures.  The article is very well written and is approachable by anyone with a bit of science background.

Forensics: Target gets recognition

The WCCO television station profiles the audio/video forensic guys at Target - yes, Target, the department store chain.  They truly are heroes in our profession, and largely unsung ones at that.  Target encourages and supports their in-house forensic unit in assisting law enforcement agencies in solving their audio/video forensic problems and they do it for free.  It is nice to see them getting some well deserved attention.

Physics: How are sound and light different?

Sound and light are very different, even though they are both waves.  For instance, one can travel in outer space and the other one cannot - at least outside of Hollywood, sound doesn't travel in a vacuum! 

To explain more fully, let's start off with how light and sound are the same.  Both are indeed waves and therefore have amplitudes, frequencies, wavelengths, and speeds.  Wavelength is the distance a wave travels before repeating, or, put another way, it is the distance between two successive troughs or peaks on a periodic (repeating) wave.  Frequency is the number of times a wave repeats in a second and is closely (and inversely) related to wavelength.  The speed of a wave is the frequency times the wavelength.

One way that they differ is in how they propagate. Sound requires a media to travel in because it is basically just a change in pressure.  For example, in air, it is a change in air pressure that carries the sound.  Sound can also be carried by other media, such as water (e.g. sonar).  Light, on the other hand, is an electromagnetic wave, which means that it is a combination of electric and magnetic fields that travel together.  Therefore, it does not need air, water, or any other medium to carry it - in essence, it carries its medium with it.

Another way in which they differ is in their speeds.  Sound travels about a foot (approximately a third of a meter) in a millisecond.  Light, on the other had, travels 186,000 miles (300,000 kilometers) per second.  Talk about the tortoise and the hare!

So, despite many similarities, sound and light differ in ways that are significant and have consequences for many different life forms - but that is a topic for another posting.

Military: Concealment and Deception or a "Magic Trick"?

The British Ministry of Defence is making news with its "invisible tank" demonstration.  Not to trivialize what their engineers have accomplished, but there are obvious parallels between what they've done and magic tricks - controlled conditions, including the viewing angle of the observers, and unseen technology, including cameras and projectors.  

That is exactly what "concealment and deception", as it is called, is all about, isn't it?  Exploit your control of the local environment and the limitations of your "opponent" to deceive them.  In this case, it involved a tank and not David Copperfield and the Statue of Liberty  (YouTube).

Computer: MacBook update

In case you are interested in the experiences of a recent convert to the "Cult of Apple", my impressions of the Apple MacBook are still very positive. At this point, I have had far fewer issues with this laptop than with any previous laptop that was 'designed and built for Windows'.  I phrased it that way because, as you might recall, I am running Mac OS-X and Windows XP Pro on this machine.  I've had no significant problems with either one.

As far as tested software packages, I've extended the list of applications I've installed and tested on the Mac OS to include Matlab (a signal processing software environment useful for filtering and plotting data, including audio and video), Octave (a freeware alternative to Matlab), Remote Desktop (for connecting to a networked Windows desktop in the office), QuickSilver (a desktop search application to beat all desktop search applications), and Gimpshop (a freeware alternative to Adobe Photoshop).  I haven't fully tested all of them, but I can say that they all install and run correctly.  Octave did give a bit of trouble during installation, but once I got it to launch the first time, it seems to be behaving properly now.

Image Recognition: License plate scanners

The technology manufacturing cycle is at work with license plate (tag) scanners - decreasing prices lead to more sales and deployments, which lead to further cost decreases and so on.  We've all seen it with personal computers, cell/mobile phones, DVD players, flat screen displays, and a multitude of other devices.  Now this market force is at work with license plate scanners , which are now being deployed on police cars.  The scanners are automating what was before a completely manual process - namely running stolen tag numbers and such - and doing it much faster.