Real-time image recognition using HLAC features at 1000 fps

Real-time image recognition at 1000 fps is realized by implementing a parallel processing circuit module to calculate higher-order local auto-correlation (HLAC) features on a high-speed vision platform. The circuit module is compactly designed in order to decrease the number of multiplications required in the HLAC calculation. The circuit module is integrated on a user-specific FPGA of the high-speed vision platform. The high-speed vision platform, on which the HLAC circuit module is hardware-implemented, can extract 25 HLAC features at 1000 fps from 1024 × 1024 pixel images, which include 0th, 1st, and 2nd order HLAC features. In the experimental results, projected images switching at frame rates as high as 250 fps are recognized by using HLAC features extracted at 1000 fps on the high-speed vision platform.