New tricks in automated image analysis

Using software to analyze images is not new. In fact, it's almost a mature technology. But there are still issues to solve, especially when you need to analyze images in real time. Advanced Imaging Magazine has looked at three particular industry segments which are using these technologies: geosciences, life sciences and military applications. This post is focused on two specific applications, cancer detection and automatic image matching for unmanned aerial vehicles (UAVs).

Using software to analyze images is not new. In fact, it's almost a mature technology. But there are still issues to solve, especially when you need to analyze images in real time. In this article, Advanced Imaging Magazine looks at three particular industry segments which are using these technologies: geosciences, life sciences and military applications. This post is focused on two specific applications, cancer detection and automatic image matching for unmanned aerial vehicles (UAVs).

The article starts by looking at remote sensing, "where the goal is to call out details from imagery generated by airborne or orbital platforms using a range of sensors." And it shows how the HALCON software developed by MVTec Software GmbH, from Munich, Germany, is used for remote sensing and aerial image interpretation.

Then, the article looks at software designed to analyze pathology specimens. For example, "Lifecan developed image analysis software for a major program aimed at detecting cancer cells from normal cells, by examining the nuclei to identify internal features that statistically distinguish cancer from normal cells."

Please note that the article mentions Lifecan or Lifescan Biosciences as the provider for this software. In fact, the real name of this company is LifeSpan Biosciences, Inc., based in Seattle, Washington.

Below are two images showing how the LifeSpan Biosciences automated image analysis can be used for cancer detection (Credit: LifeSpan Biosciences). On the left, you can see how a cancer is detected based on hematoxylin morphology. And on the right, after cancer a cancer has been localized, "a marker is quantified in nuclear and cytoplasmic areas." You'll find other examples of these imaging technologies on this page.

Automated image analysis for cancer detection

Of course, this technology is not designed to replace human pathologists.

"The target is more realistically to look in the specimen and get a "yes/no" answer as to whether or not it is normal, and what is the probability of it being cancer. You can either use this technology to review specimens before the human looks at them to draw his or her attention to certain features, or you can use it to look at specimens after the human is done to make sure that nothing was missed. Both are valid approaches," [said Joseph Brown, Lifespan president and CEO.]

Now, let's turn to military applications of automated image analysis to remote sensing, automated targeting or automated navigation.

[For these last two applications,] the challenge often involves dynamically changing backgrounds, moving platforms capturing the imagery, a lack of perspective control and the need for real-time analysis under harsh battlefield conditions. Since the output often results in life-or-death decisions, reliability has to be high.

These are the challenges faced by Octec Ltd., based in Bracknell, UK.

"You can do image analysis in industrial machine vision types of applications fairly quickly, though rarely full frame rate because there isn't actually that much of the necessity to do that. The added challenge for us is to do image analysis in real-time and then get it into a piece of hardware that is sensible and manageable in a military environment," [said Roger Joel, sales and marketing manager for Octec.]

Below are two examples of the Automatic Image Matching (AIM) technology used by Optec for real time video image processing or image enhancement. "It would allow, for example, matching of an infrared image taken from the payload of a Unmanned Aerial Vehicle (UAV) to a visible band satellite image. This could assist the UAV in terms of navigation or target detection." On the left, you can see "a thermal image extracted from an airborne platform. Given the position and orientation of the imager, it is possible to approximately match the viewpoint of the imager by warping a geo-referenced photo of the area, as shown by the warped aerial photo on the right." (Credit: Optec)

Automatic image matching for UAVs

Octec also developed the software which was used for the Autonomous Airborne Refueling Demonstration (AARD) System to perform the first ever autonomous probe and drogue airborne refueling engagement on August 30, 2006 at Edwards Air Force Base, California, as you can read in this press release from Radstone Embedded Computing, Towcester, UK.

For your viewing pleasure, here is a link to a short video of this first autonomous refueling engagement (Windows Media format, 1 minute)

Sources: Keith Reid, Advanced Imaging Magazine, December 4, 2006; and various websites

You'll find related stories by following the links below.