Abstract
The integral image, an intermediate image representation, has found extensive use in multi-scale local feature detection algorithms, such as Speeded-Up Robust Features (SURF), allowing fast computation of rectangular features at constant speed, independent of filter size. For resource-constrained real-time embedded vision systems, computation and storage of integral image presents several design challenges due to strict timing and hardware limitations. Although calculation of the integral image only consists of simple addition operations, the total number of operations is large owing to the generally large size of image data. Recursive equations allow substantial decrease in the number of operations but require calculation in a serial fashion. This paper presents two new hardware algorithms that are based on the decomposition of these recursive equations, allowing calculation of up to four integral image values in a row-parallel way without significantly increasing the number of operations. An efficient design strategy is also proposed for a parallel integral image computation unit to reduce the size of the required internal memory (nearly 35% for common HD video). Addressing the storage problem of integral image in embedded vision systems, the paper presents two algorithms which allow substantial decrease (at least 44.44%) in the memory requirements. Finally, the paper provides a case study that highlights the utility of the proposed architectures in embedded vision systems.
Highlights
Proposed as the summed-area table for texture-mapping in computer graphics in the mid-1980s [1], the integral image is comparatively new in the image processing domain
Employment of the integral image eliminates computationally expensive multiplications for box filter calculation, reducing it to three addition operations [2]. This allows all box filters to be computed at a constant speed, irrespective of their size; this is a major advantage for computer vision algorithms, especially feature detection techniques which utilize multi-scale analysis
This paper presents two generic methods for reducing the storage requirements of the integral image significantly which can benefit both custom hardware design and software implementation on programmable processor architectures for resource constrained embedded vision systems
Summary
Proposed as the summed-area table for texture-mapping in computer graphics in the mid-1980s [1], the integral image is comparatively new in the image processing domain. The idea of using an integral image was introduced as an intermediate image representation by the Viola-Jones face detector [2] Since it has been useful for fast implementation of image pyramids in multi-scale computer vision algorithms such as Speeded-Up Robust Features (SURF) and Fast. Employment of the integral image eliminates computationally expensive multiplications for box filter calculation, reducing it to three addition operations [2] This allows all box filters to be computed at a constant speed, irrespective of their size; this is a major advantage for computer vision algorithms, especially feature detection techniques which utilize multi-scale analysis. Such algorithms generally require calculation of variable-size box filters to implement different scales of an image pyramid. SURF requires computation of 9 × 9 box filters for implementation of the smallest and 195 × 195 for the largest scale of its image pyramid [3]; without an integral image, these larger filters would take almost 500 times longer than the smallest one to compute
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
