I

Enhancement

1 Fundamental Enhancement Techniques Raman B. Paranjape 3

2 Adaptive Image Filtering Carl-Fredrik Westin, Hans Knutsson, and Ron Kikinis 19

3 Enhancement by Multiscale Nonlinear Operators Andrew Laine and Walter Huda 33

4 Medical Image Enhancement with Hybrid Filters Wei Qian 57

Rangaraj M. Rangayyan

University of Calgary

Medical images are often deteriorated by noise due to various sources of interference and other phenomena that affect the measurement processes in imaging and data acquisition systems. The nature of the physiological system under investigation and the procedures used in imaging also diminish the contrast and the visibility of details. For example, planar projection nuclear medicine images obtained using a gamma camera as well as single-photon emission computed tomography (SPECT) are severely degraded by Poisson noise that is inherent in the photon emission and counting processes. Although mammograms (X-ray images of the breast) are not much affected by noise, they have limited contrast because of the nature and superimposition of the soft tissues of the breast, which is compressed during the imaging procedure. The small differences that may exist between normal and abnormal tissues are confounded by noise and artifacts, often making direct analysis of the acquired images difficult.

In all of the cases just mentioned, some improvement in the appearance and visual quality of the images, even if only subjective, may assist in their interpretation by a medical specialist.

Image enhancement techniques are mathematical techniques that are aimed at realizing improvement in the quality of a given image. The result is another image that demonstrates certain features in a manner that is better in some sense as compared to their appearance in the original image. One may also derive or compute multiple processed versions of the original image, each presenting a selected feature in an enhanced appearance. Simple image enhancement techniques are developed and applied in an ad hoc manner. Advanced techniques that are optimized with reference to certain specific requirements and objective criteria are also available.

Although most enhancement techniques are applied with the aim of generating improved images for use by a human observer, some techniques are used to derive images that are meant for use by a subsequent algorithm for computer processing. Examples of the former category are techniques to remove noise, enhance contrast, and sharpen the details in a given image. The latter category includes many techniques in the former, but has an expanded range of possibilities, including edge detection and object segmentation.

If used inappropriately, enhancement techniques themselves may increase noise while improving contrast, they may eliminate small details and edge sharpness while removing noise, and they may produce artifacts in general. Users need to be cautious to avoid these pitfalls in the pursuit of the best possible enhanced image.

The first chapter, by Paranjape, provides an introduction to basic techniques, including histogram manipulation, mean and median filtering, edge enhancement, and image averaging and subtraction, as well as the Butterworth filter. Applications illustrate contrast enhancement, noise suppression, edge enhancement, and mappings for image display systems. Dental radiographic images and CT images of the brain are used to present the effects of the various operations. Most of the methods described in this chapter belong to the ad hoc category and provide good results when the enhancement need is not very demanding. The histogram equalization technique is theoretically well founded with the criterion of maximal entropy, aiming for a uniform histogram or gray-level probability density function. However, this technique may have limited success on many medical images because they typically have details of a wide range of size and small gray-level differences between different tissue types. The equalization procedure based on the global probability with a quantized output gray scale may obliterate small details and differences. One solution is the locally adaptive histogram equalization technique described in this chapter. The limitations of the fundamental techniques motivated the development of adaptive and spatially variable processing techniques.

The second chapter by Westin et al. presents the design of the adaptive Wiener filter. The Wiener filter is an optimal filter derived with respect to a certain objective criterion. Westin et al. describe how the Wiener filter maybe designed to adapt to local and spatially variable details in images. The filter is cast as a combination of low-pass and high-pass filters, with factors that control their relative weights. Application of the techniques to CT and MR images is illustrated.

The third chapter by Laine et al. focuses on nonlinear contrast enhancement techniques for radiographic images, in particular mammographic images. A common problem in contrast or edge enhancement is the accompanying but undesired noise amplification. A wavelet-based framework is described by Laine et al. to perform combined contrast enhancement and denoising, that is, suppression of the noise present in the input image and/or control of noise amplification in the enhancement process. The basic unsharp masking and subtracting Laplacian techniques are included as special cases of a more general system for contrast enhancement.

The fourth and final chapter of the section, by Qian, describes a hybrid filter incorporating an adaptive multistage nonlinear filter and a multiresolution/multiorientation wavelet transform. The methods address image enhancement with noise suppression, as well as decomposition and selective reconstruction of wavelet-based subimages. Application of the methods to enhance microcalcification clusters and masses in mammograms is illustrated.

Together, the chapters in this section present an array of techniques for image enhancement: from linear to nonlinear, from fixed to adaptive, and from pixel-based to multiscale methods. Each method serves a specific need and has its own realm of applications. Given the diverse nature of medical images and their associated problems, it would be difficult to prescribe a single method that can serve a range of problems. An investigator is well advised to study the images and their enhancement needs, and to explore a range of techniques, each of which may individually satisfy a subset of the requirements. A collection of processed images may be called for in order to meet all the requirements.

0 0

Post a comment