Kernel Operators

Wesley E. Snyder; Hairong Qi

doi:10.1017/9781316882641.009

5 - Kernel Operators

from Part II - Preprocessing

Published online by Cambridge University Press: 25 October 2017

Wesley E. Snyder and

Hairong Qi

Show author details

Wesley E. Snyder: Affiliation:
North Carolina State University
Hairong Qi: Affiliation:
University of Tennessee

Book contents

Get access

Summary

You need boundaries … Even in our material creations, boundaries mark the most beautiful of places, between the ocean and the shore, between the mountains and the plains, where the canyon meets the river

– Wm. Paul Young

Introduction

In this chapter, linear operations on images are investigated. The derivative, probably the most common linear operator, is considered first. That discussion is extended into edge detection, the result of applying derivative kernels on images. A variety of methods for accomplishing this objective are considered.

• (Section 5.2)We first define what makes a linear operator and explain the kernel operator, or sum of products, which is the type of linear operator that will appear most frequently in this book.

• (Section 5.3) We show in this section how to use vector representation of images to convert kernel operators to the matrix multiplication form in order to facilitate some analytical study.

• (Section 5.4) We focus on the discussion of derivatives, since it is probably the most common kernel operator used in Computer Vision applications. We discuss how to find the derivative kernel operators through the definition of derivatives, by function fitting, using vector representation of images, and taking derivatives of special blurring kernels.

• (Section 5.5) Applying derivative operators on images would result in edge images. We describe in this section the different types of edges and a popular edge detector, the Canny edge detector.

• (Section 5.6) The concept of scale space is explained and how the scale space is associated with the detection of edges at different scales.

• (Section 5.7) In Chapter 1, we briefly described the biological vision. Here, we revisit this issue and ask the question how human beings perform lower-level vision, i.e., edge detection. We introduce the Gabor filter, which is believed to well characterize the functionality of receptive fields in the visual cortex.

• (Section 5.8) We wrap up this chapter by performing an experimental study on the various types of derivative kernels in terms of their ability in detecting edges (both angle and magnitude) of different orientations.

Type: Chapter
Information: Fundamentals of Computer Vision , pp. 51 - 91

DOI: https://doi.org/10.1017/9781316882641.009 [Opens in a new window]

Publisher: Cambridge University Press

Print publication year: 2017

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

[5.1] M., Alfaraj, Y., Wang, and Y., Luo. Enhanced isotropic gradient operator. Geophysical Prospecting, 62, 2014.Google Scholar

[5.2] V., Anh, J., Shi, and H., Tsai. Scaling theorems for zero crossings of bandlimited signals. IEEE Trans. Pattern Anal. and Machine Intel., 18 (3), 1996.Google Scholar

[5.3] J., Babaud, A., Witkin, M., Baudin, and R., Duda. Uniqueness of the Gaussian kernel for scalespace filtering. IEEE Trans. Pattern Anal. and Machine Intel., 8 (1), 1986.Google Scholar

[5.4] G., Bilbro and W., Snyder. Optimization of functions with many minima. IEEE Transactions on SMC, 21 (4), July/August 1991.Google Scholar

[5.5] P., Burt and E., Adelson. The Laplacian pyramid as a compact image code. CVGIP, 16, 1981.Google Scholar

[5.6] J., Canny. A computational approach to edge detection. IEEE Trans. Pattern Anal. and Machine Intel., 8 (6), 1986.Google Scholar

[5.7] J., Crowley. A Representation for Visual Information, Ph.D. Thesis. CMU, 1981.

[5.8] J., Daugman. Two-dimensional spectral analysis of cortical receptive fields. Vision Research, 20, 1980.Google Scholar

[5.9] J., Daugman. Uncertainly relation for resolution in space, spatial frequency, and orientation optimized by two-dimensional visual cortical filters. J. Optical Soc. America, 2 (7), 1985.Google Scholar

[5.10] W., Frei and C., Chen. Fast boundary detection: A generalization and a new algorithm. IEEE Transactions on Computers, 25 (2), 1977.Google Scholar

[5.11] R., Haralick and L., Shapiro. Computer and Robot Vision, volume 1. Addison-Wesley, 1992.

[5.12] M., Van Horn,W., Snyder, and D., Herrington. A radial filtering scheme applied to intracoronary ultrasound images. Computers in Cardiology, Sept. 1993.

[5.13] D., Hubel and T., Wiesel. Receptive fields, binocular interaction, and functional architecture in the cat's visual cortex. Journal of Physiology (London), 160, 1962.Google Scholar

[5.14] D., Hubel and T., Wiesel. Functional architecture of macaque monkey visual cortex. Proceedings of the Royal Society, B, 198, 1978.Google Scholar

[5.15] P., Jackway and M., Deriche. Scale-space properties of the multiscale morphological dilationerosion. IEEE Trans. Pattern Anal. and Machine Intel., 18 (1), 1996.Google Scholar

[5.16] J., Jones and L., Palmer. An evaluation of the two-dimensional gabor filter model of simple receptive fields in the cat striate cortex. J. Neurophysiology, 58, 1987.Google Scholar

[5.17] E., Joseph and T., Pavlidis. Bar code waveform recognition using peak locations. IEEE Trans. Pattern Anal. and Machine Intel., 16 (6), 1994.Google Scholar

[5.18] M., Kelly. Machine Intelligence 6. Univ. Edinburgh Press, 1971.

[5.19] M., Kisworo, S., Venkatesh, and G., West. Modeling edges at subpixel accuracy using the local energy approach. IEEE Trans. Pattern Anal. and Machine Intel., 16 (4), 1994.Google Scholar

[5.20] T., Lee. Image representation using 2-d Gabor wavelets. IEEE Trans. Pattern Anal. and Machine Intel., 18 (10), 1996.Google Scholar

[5.21] Y., Leung, J., Zhang, and Z., Xu. Clustering by scale-space filtering. IEEE Trans. Pattern Anal. and Machine Intel., 22 (12), 2000.Google Scholar

[5.22] T., Lindeberg. Scale-space for discrete signals. IEEE Trans. Pattern Anal. and Machine Intel., 12 (3), 1990.Google Scholar

[5.23] T., Lindeberg. Scale-space theory, a basic tool for analysing structures at different scales. Journal of Applied Statistics, 21 (2), 1994.Google Scholar

[5.24] D., Marr and E., Hildreth. Theory of edge detection. Proc. Royal Society of London, B, 207, 1980.Google Scholar

[5.25] D., Marr and T., Poggio. A computational theory of human stereo vision. In Proc. Royal Society of London, 1979.

[5.26] E., Pauwels, L., Van Gool, P., Fiddelaers, and T., Moons. An extended class of scale-invariant and recursive scale space filters. IEEE Trans. Pattern Anal. and Machine Intel., 17 (7), 1995.Google Scholar

[5.27] P., Perona and J., Malik. Scale-space and edge detection using anisotropic diffusion. IEEE Transactions Pattern Analysis and Machine Intelligence, 12, 1990.Google Scholar

[5.28] H., Tagare and R., deFigueiredo. Reply to on the localization performance measure and optimal edge detection. IEEE Trans. Pattern Anal. and Machine Intel., 16 (1), 1994.Google Scholar

[5.29] P., Verbeek and L., van Vliet. On the location error of curved edges in low-pass filtered 2-d and 3-d images. IEEE Trans. Pattern Anal. and Machine Intel., 16 (7), 1994.Google Scholar

[5.30] I., Weiss. High-order differentiation filters that work. IEEE Trans. Pattern Anal. and Machine Intel., 16 (7), 1994.Google Scholar

[5.31] R., Young. The Gaussian derivative model for spatial vision: I. retinal mechanisms. Spatial Vision, 2, 1987.Google Scholar

Book contents

5 - Kernel Operators

Summary

Access options

References

Save book to Kindle

Save book to Dropbox

Save book to Google Drive