Word length reduction for the integral image

Abstract

The integral image is an image containing accumulated sums of pixel values taken from an input image. It is an important concept for multi-scale image processing algorithms, for it provides a very economic way to compute the sum of pixel values in any rectangular input image region.Unfortunately, the integral image requires a large binary word length to represent the accumulated sums. This is an issue for platforms having limited memory, power, and bandwidth like in mobile devices. Our paper deals with two methods for word length reduction, involving computation through the overflow and rounding with error diffusion. We show by experiment that, based on a word length reduced integral image, the Viola and Jones face detector for a VGA resolution can work on a 16-bit CPU (i.s.o. 27 bits, which becomes 32 bits on byte-oriented CPUs), enabling face detection on a wider range of platforms.

A New Integral Image Structure for Memory Size Reduction

“…However, the method of this paper, and the method of [4] can be used together. The paper [4] specifies the data size of the integral image, using the modulo operation according to the maximum size of the box filter.…”

“…Comparison with [4] is difficult, because the data size depends on the box filter size. However, the method of this paper, and the method of [4] can be used together.…”

“…As described in [4], the size of the integral image memory requires 27-bit word length memory for 640 × 480 input image size.…”

“…Research, such as [4], reduces integral image memory using modulo operation and rounding. However, this approach has drawbacks, including rounding errors and an additional constraint of fixed size of the box filter.…”

A New Method of Storing Integral Image for Memory Efficiency Using Modified Block Structure

“…As described in [4], the size of the integral image memory requires 27-bit word length memory for 8bit 640 × 480 input image, and its size becomes even larger for HD or full HD images.…”