Forensic analysis investigates images
Ray Liu, Ashwin Swaminathan, and Min Wu of the Electrical and Computer Engineering Department at the University of Maryland (College Park, MD, USA; www.umd.edu) have developed a method for forensic analysis of digital camera images based on the observation that color interpolation leaves distinct intrinsic traces on images. These intrinsic fingerprints can be identified and can verify the authenticity of digital data. Using an imaging model and applying component analysis techniques, the researchers can determine which interpolation algorithm is being used, estimate the parameter settings, and thus determine the camera that took the picture. Classification accuracy rate is about 90%. For example, photos taken with different cameras can be combined to form a new tampered image. The method, when applied to tamper detection, can determine that the tampered image consists of two portions from different cameras.
null
Vision delivers toothbrush variety
High mechanical quality is mandatory in an electric toothbrush, but the market—especially the US market—also requires variety in color and design. A major manufacturer of electric toothbrushes has chosen to meet the demand for rapidly changing color schemes for molded plastic toothbrush shells by installing a machine-vision system developed by Wolf Systeme (Neulingen-Göbrichen, Germany; www.wolfsysteme.de)
The major challenge was inspecting for color changes to existing products. Given the small batches produced of each color and the fact that each shell consists of two or three injection-molded pieces, it is difficult to ensure that the colored parts are properly matched. While the geometrical forms of the plastic shells remain the same, the color combinations continually change. And, to enhance appearance, glittering metal may be added to the components, which increases the challenge of surface inspection.
To meet the complex demands for sorting out mismatches and defects, Wolf Systeme modified its flexible MVplus machine-vision system. “We were increasingly faced with the problem that we always had to adopt common software products for every new customer application and every new product,” says Markus Wolf, CEO of the company. He says MVplus now uses HALCON software from MVTec Software (Munich, Germany; www.mvtec.com) because it is object-oriented, flexible software with a standard library that runs on parallel processors.
Nanoshells improve early cancer detection
To improve early cancer detection, researchers are developing in vivo biomedical imaging techniques with the resolution to distinguish between healthy and malignant tissue. Yet these techniques could still be improved with contrast enhancement agents. Gold nanostructures are promising for optical imaging because they can absorb and scatter light at specific wavelengths. This phenomenon, called localized surface plasmon resonance (LSPR), can improve contrast by enhancing or damping the optical signals characteristic of certain types of tissue.
To be useful in biomedical imaging, the LSPR must be tuned for maximum absorption in the near-infrared (near-IR) range, where light attenuation by blood and soft tissue is minimal. As a result, Younan Xia of the Department of Biomedical Engineering of Washington University (St. Louis, MO, USA; ww.wustl.edu) prepared gold nanoshells (composite particles with a metallic shell and dielectric core) and nanorods tuned into the near-IR. Because they are difficult to predictably prepare, however, Xia is using hollow, porous structures called nanocages.
Vision inspects surgical implants
At its facility in Switzerland, DePuy Spine, (Raynham, MA, USA; www.depuyspine.com) produces small implants such as screws for spinal surgery. Previously, inspection had been done manually, but with regulations growing stricter, it became clear that only vision technology could provide the necessary quality of inspection. Using a desktop robot, Sony (Park Ridge, NJ, USA; www.sony.com/videocameras) camera, and Cognex (Natick, MA, USA; www.cognex.com) vision library, system-integrator Compar (Pfäffikon, Switzerland; www.compar.ch) developed a flexible part automation system that proved more suitable and economical than fully automated in-line inspection. Each implant is placed in the cell of a blister pack. Each batch is identified by a code, and certain implant nests may be empty. The camera is mounted on the robot’s axis, while the blister packs are moved under the camera in the x direction. Johann Faneca, production manager at DePuy Spine, says, “We are very pleased with the vision installation, which we estimate has saved more than $95,000.”
null