Chapter 4: Image Enhancement Introduction and Gray-Scale Modification

Introduction Image enhancement techniques are used to emphasize and sharpen image features for display and analysis. In general, image enhancement is used to generate a visually desirable image. It can be used as a preprocess or a postprocess. Highly application dependent. A technique that works for one application may not work for another.

Introduction There are three types of image enhancement techniques: Point operations: each pixel is modified according to a particular equation, independent of the other pixels. Mask operations: each pixel is modified according to the values of the pixel’s neighbors. Global operations: all the pixel values in the image or subimage are taken into consideration.

Overview of Gray-Scale Modification Gray-scale modification methods belong in the category of point operations. They function by changing the pixel’s gray-level value by using a mapping equation. The mapping function maps the original gray-level values to other, specified values. The primary operations applied to the gray scale of an image are to compress or stretch it.

Overview of Gray-Scale Modification Gray-level compression is done to gray-level ranges that are of little interest. Gray-level stretching is done to gray-level ranges where we desire more information. The gray-level compression and stretching can be illustrate using the graph of modified gray-level vs. original gray-level.

Overview of Gray-Scale Modification 255 128 75 28 Original gray-level values Modified gray-level values Stretch slope>1 Gray-level stretching

Overview of Gray-Scale Modification Original Image Image after gray-level stretching

Overview of Gray-Scale Modification If the mapping line has a slope between 0 and 1, this is called gray-level compression. If the slope is greater than one, then it is called gray-level stretching. In the previous example, the range of gray-level values from 28 to 75 is stretched, while other gray-level values are left alone.

Overview of Gray-Scale Modification Stretching a particular gray-level range can expose a previously hidden visual info. In some cases, we may want to stretch a specific range of gray levels, while clipping the values at the low and high ends. The effect of doing this is that the contrast of the image is enhanced.

Overview of Gray-Scale Modification 255 128 50 Original gray-level values Modified gray-level values Gray-level stretching with clipping at ends 200

Overview of Gray-Scale Modification Original Image Image after gray-level stretching

Overview of Gray-Scale Modification Another type of mapping equation is called the intensity-level slicing. Used for feature extraction. Here we select specific gray-level values of interest and map them to a specified, typically higher, value. Using this method, we can “bring out” the feature of interest in the image.

Overview of Gray-Scale Modification

Overview of Gray-Scale Modification

Overview of Gray-Scale Modification

Histogram Modification An alternate perspective to gray-level modification that performs a similar function is referred to as histogram modification. The gray-level histogram of an image is the distribution of the gray levels in an image. The characteristics of an image can be determined from its histogram (refer to “Histogram Features” in Chapter 2).

Histogram Modification

Histogram Modification The histogram can be modified by a mapping function which will stretch, shrink or slide the histogram. This will change the contrast or brightness of the image. The graphical representation of histogram stretch, shrink and slide can be seen in the following diagrams.

Histogram Modification

Histogram Modification

Histogram Modification

Histogram Modification The mapping equation for histogram stretch can be found as follows: I(r,c)MAX is the largest gray level value in the image I(r,c) I(r,c)MIN is the smallest gray level value in I(r,c) MAX and MIN correspond to the maximum and minimum gray-level values of the new range.

Histogram Modification This equation will take an image and stretch the histogram across the entire gray-level range. This will increase the contrast of a low-contrast image. If a stretch is desired over a smaller range, different MAX and MIN values can be specified.

Histogram Modification Low-contrast image Histogram of low-contrast image

Histogram Modification Image after histogram stretching Histogram of image after stretching

Histogram Modification If most of the pixel values in an image fall within a small range, but a few outliners force the histogram to span the entire range, a pure histogram stretch will not improve the image. In this case, it is useful to allow a small percentage of the pixel values to be clipped at the low and high end of the range.

Histogram Modification Original Image Histogram of the original image

Histogram Modification Image after histogram stretching without clipping Histogram of the image

Histogram Modification Image after histogram stretching with clipping 3% low and high value Histogram of the image

Histogram Modification The opposite of histogram stretch is a histogram shrink, which will decrease image contrast by compressing the gray levels. The histogram shrinking equation is generally the same as the one for stretching. But for histogram shrinking, MAX and MIN should be set to the maximum and minimum of the new, compressed range.

Histogram Modification Original image Histogram of original image

Histogram Modification Image after histogram shrink to the range [75, 175] Histogram of the image

Histogram Modification In general, histogram shrink reduces contrast and may not seem to be useful as image enhancement tool. However, there is an image-sharpening technique algorithm that uses the histogram shrink process as a part of the enhancement technique.

Histogram Modification The histogram slide technique can be used to make an image either darker or lighter. Darker: slide histogram towards low end. Lighter: slide histogram towards high end. Histogram slide is done by adding or subtracting a fixed number from all the gray-level values.

Histogram Modification Any values slid past the minimum or maximum values will be clipped to the respective minimum and maximum. A positive OFFSET will increase the overall brightness. A negative OFFSET will create a darker image.

Histogram Modification Original image Histogram of original image

Histogram Modification Image after positive-value histogram sliding Histogram of image after sliding

Histogram Modification Histogram equalization is a popular technique for improving the appearance of a poor image. Its function is similar to that of histogram stretch but often provides more visually pleasing results across a wider range of images.

Histogram Modification The histogram equalization process consists of four steps: Find the running sum of the histogram values. Normalize the values from step 1 by dividing by the total number of pixels. Multiply the values from step 2 by the maximum gray level value and round. Map the gray-level values to the result from step 3 using one-to-one correspondence.

Histogram Modification Example: You are given a 3 bits/pixel image with the following histogram: Next, perform the four steps histogram equalization process as mentioned before. The result can be seen in the tables in the next slide. Gray-level 1 2 3 4 5 6 7 No of Pixel 10 8 9 14

Histogram Modification The first three steps: Gray-level 1 2 3 4 5 6 7 No of Pixel 10 8 9 14 Run Sum 18 27 29 43 44 49 51 Normalized 10/51 18/51 27/51 29/51 43/51 44/51 49/51 51/51 Multiply by 7 The fourth step: Old 1 2 3 4 5 6 7 New

Histogram Modification Original image Histogram of original image

Histogram Modification Image after histogram equalization Histogram after equalization

Histogram Specification Sometimes, it is useful to be able to define a histogram and modify the histogram of the original image to match the histogram that we define. Such as process is called histogram specification. This process can be implemented in 4 steps:

Histogram Specification Find the mapping table to histogram-equalize the image (this is basically the result of histogram equalization). Specify the desired histogram. Find the mapping table to histogram-equalize the values of the desired histogram (this is done by applying histogram equalization to the specified histogram in step 2). Map the original values to the values from step 3.

Histogram Specification Step 1: Use histogram equalization result from last example Original Gray-level Value - O 1 2 3 4 5 6 7 Histogram Equalized Values - H Step 2: Specify the desired histogram Gray-Level Value 1 2 3 4 5 6 7 Number of Pixels in Desired Histogram 10 15 20

Histogram Specification Step 3: Find the histogram equalization mapping table for the desired histogram Gray-Level Value 1 2 3 4 5 6 7 Histogram Equalized Values - S Step 4: Map the original values to the values from step 3 O 1 2 3 4 5 6 7 H S M

Adaptive Contrast Enhancement Adaptive Contrast Enhancement (ACE) filter is used with an image with uneven contrast. In this case, we want to adjust the contrast differently in different regions of the image. Regions with low contrast should be given more contrast compared to other regions. This is different from image modification techniques, which are based only on global parameters.

Adaptive Contrast Enhancement ACE works by using both the local and global image statistics to determine the amount of contrast adjustment required. The image is processed using the sliding window concept. The local image statistics are found by considering only the current window. The global statistics are found by considering the entire image.

Adaptive Contrast Enhancement The ACE equation is as follows: mI(r,c) = is the mean for the entire image I(r,c) σl = local standard deviation (in the window) ml = local mean (average in window) k1, k2 = constants, vary between 0 and 1

Adaptive Contrast Enhancement This filter subtracts the local mean from the original data and weights the result by the local gain factor k1[mI(r,c)/σl(r,c)]. This has the effect of intensifying local variations. Can be controlled by the constant k1. Areas of low contrast (low values of σl(r,c)) are boosted.

Adaptive Contrast Enhancement The mean is then added back to the result, weighted by k2 to restore the local average brightness. In practice, it is often helpful to shrink the histogram of image before applying this filter. It is also helpful to limit the range of the local gain factor, i.e. set a minimum and maximum for the local gain factor.

Adaptive Contrast Enhancement Original Image Histogram equalized version of original image

Adaptive Contrast Enhancement Image after being applied with ACE filter. k1 = 0.9, k2 = 0.5 Local gain max = 25 Histogram equalized version of ACE filtered image