Image processing
EMVA Young Professional Award 2017 presented
This year's EMVA Young Professional Award goes to Boaz Arad. He is being honored for his work "Sparse Recovery of Hyperspectral Signal from Natural RGB Images".
The award-winning work describes the restoration of 31-channel hyperspectral information from 3-channel RGB images of natural scenes. Despite the extreme reduction in dimensionality of 31:3 when projecting hyperspectral information to RGB, the method presented allows the hyperspectral data to be recovered from the RGB data with an accuracy of 90 to 95 % for a large number of scenes.
Hyperspectral (HS) imaging systems can capture the complete spectral signature reflected from each point in a scene. This produces a much more detailed spectral image than is possible using RGB cameras. Until recently, scientific and industrial applications that rely on hyperspectral image information have relied almost exclusively on the familiar technology of hyperspectral scanning systems. However, these systems are expensive, bulky and often take up to a minute to capture a complete scene. Replacing these systems with inexpensive, compact and fast RGB cameras would open up interesting new possibilities in many areas of application.
Arad, 32, received his Bachelor of Science (B.Sc.) in Computer Science from Ben-Gurion University of the Negev (BGU) in 2012. On the accelerated path to a PhD at BGU's Interdisciplinary Computational Vision Laboratory (ICVL), he received his Master's degree (M.Sc.) in 2014 and hopes to complete his PhD in 2018. In addition to his studies, Boaz is currently the CTO of the startup company "HC-Vision".
The EMVA Young Professional Award is an annual prize that honors the exceptional and innovative work of students or young professionals in machine vision. The aim of the European Machine Vision Association (EMVA) is to further promote innovation in machine vision, to contribute to education in the machine vision discipline and to build a bridge between research and industry.










