Abstract-Stereo image rectification is a pre-processing step of disparity estimation intended to remove image distortions and to enable stereo matching along an epipolar line. A real-time disparity estimation system needs to perform real-time rectification which requires solving the models of lens distortions, image translations and rotations. Look-up-table based rectification algorithms allow image rectification without demanding high complexity operations. However, they require an external memory to store large size look-up-tables. In this work, we present an intermediate solution that compresses the rectification information to fit the look-up-table into the onchip memory of a Virtex-5 FPGA. The low-complexity decompression process requires a negligible amount of hardware resources for its real-time implementation. The proposed image rectification hardware consumes 0.28% of the DFF and 0.32% of the LUT resources of the Virtex-5 XCUVP-110T FPGA, it can process 347 frames per second for a 1024×768 pixels image resolution, and it does not need the availability of an external memory.

International audienceReal-time disparity estimation requires real-time rectification which involves solving the models of lens distortions, image translations and rotations. Low complexity look-up-table based rectification algorithms usually require an external memory to store large look-up-tables. In this chapter, we present an implementation of the look-up-table based approach which compresses the rectification information to fit the look-up-table into the on-chip memory of a Virtex-5 FPGA. First, a very low complexity compressed look-up-table based rectification algorithm (CLUTR) and its real-time hardware are presented. The implemented CLUTR hardware rectifies stereo images with moderate lens distortion and camera misalignment. Moreover, an enhanced version of the compressed look-up-table based rectification algorithm (E-CLUTR) and its novel real-time hardware are presented. E-CLUTR solves more extreme camera alignment and distortion issues than CLUTR while maintaining the low complexity architecture