Hardware acceleration of image recognition through a visual cortex model

• Published in: Optics and laser technology 2008-09, Vol.40 (6), p.795-802
• Main Authors: Rice, Kenneth L., Taha, Tarek M., Vutsinas, Christopher N.
• Format: Article
• Language: English
• Subjects: Computers in experimental physics; Exact sciences and technology; Fundamental areas of phenomenology (including applications); Hardware; Image recognition; Imaging and optical processing; Instruments, apparatus, components and techniques common to several branches of physics and astronomy; Neural networks, fuzzy logic, artificial intelligence; Optical computers, logic elements, interconnects, switches; neural networks; Optical elements, devices, and systems; Optics; Physics
• ISSN: 0030-3992; 1879-2545
• DOI: 10.1016/j.optlastec.2007.12.007

Recent findings in neuroscience have led to the development of several new models describing the processes in the neocortex. These models excel at cognitive applications such as image analysis and movement control. This paper presents a hardware architecture to speed up image content recognition through a recently proposed model of the visual cortex. The system is based on a set of parallel computation nodes implemented in an FPGA. The design was optimized for hardware by reducing the data storage requirements, and removing the need for multiplies and divides. The reconfigurable logic hardware implementation running at 121 MHz provided a speedup of 148 times over a 2 GHz AMD Opteron processor. The results indicate the feasibility of specialized hardware to accelerate larger biological scale implementations of the model.