Nicholas Sia Pik Kong

Brightness Preserving Dynamic Histogram Equalization for Image Contrast Enhancement

Histogram equalization (HE) is one of the common methods used for improving contrast in digital images. However, this technique is not very well suited to be implemented in consumer electronics, such as television, because the method tends to introduce unnecessary visual deterioration such as the saturation effect. One of the solutions to overcome this weakness is by preserving the mean brightness of the input image inside the output image. This paper proposes a new method, known as brightness preserving dynamic histogram equalization (BPDHE), which is an extension to HE that can produce the output image with the mean intensity almost equal to the mean intensity of the input, thus fulfill the requirement of maintaining the mean brightness of the image. First, the method smoothes the input histogram with one dimensional Gaussian filter, and then partitions the smoothed histogram based on its local maximums. Next, each partition will be assigned to a new dynamic range. After that, the histogram equalization process is applied independently to these partitions, based on this new dynamic range. For sure, the changes in dynamic range, and also histogram equalization process will alter the mean brightness of the image. Therefore, the last step in this method is to normalize the output image to the input mean brightness. Our results from 80 test images shows that this method outperforms other present mean brightness preserving histogram equalization methods. In most cases, BPDHE successfully enhance the image without severe side effects, and at the same time, maintain the mean input brightness1.

Bi-histogram equalization with a plateau limit for digital image enhancement

Many histogram equalization based methods have been introduced for the use in consumer electronics in recent years. Yet, many of these methods are relatively complicated to be implemented, and mostly require a high computational time. Furthermore, some of the methods require several predefined parameters from the user, which make the optimal results cannot be obtained automatically. Therefore, this paper presents bi-histogram equalization with a plateau level (BHEPL) as one of the options for the system that requires a short processing time image enhancement. First, BHEPL divides the input histogram into two independent sub-histograms. This is done in order to maintain the mean brightness. Then, these sub-histograms are clipped based on the calculated plateau value. By doing this, excessive enhancement can be avoided. Experimental results show that this method only requires 34.20 ms, in average, to process images of size 3648 × 2736 pixels (i.e. 10 Mega pixels images). The proposed method also gives better enhancement results as compared with some multi-sections mean brightness preserving histogram equalization methods.

Color image enhancement using brightness preserving dynamic histogram equalization

Histogram equalization (HE), although one of the most popular techniques used for digital image enhancement, is not very suitable to be implemented directly in consumer electronics, such as television, because this method tends to produce an output with saturation effect. To overcome this weakness, it is suggested that the mean intensity of the input image be maintained in the output image. Previously, we proposed a method known as brightness preserving dynamic histogram equalization (BPDHE) which can fulfill this requirement for grayscale images. In this paper, we present several possibilities to extend this method for color images.

Image sharpening using sub-regions histogram equalization

Histogram equalization (HE) based methods are commonly used in consumer electronics. Histogram equalization improves the contrast of an image by changing the intensity level of the pixels based on the intensity distribution of the input image. This paper presents subregions histogram equalization (SRHE). First, the method partitions the image based on the smoothed intensity values, which are obtained by convolving the input image with a Gaussian filter. By doing this, the transformation function used by HE is not based on the intensity of the pixels only, but the intensity values of the neighboring pixels are also taken into the consideration. Besides, this paper also presents a more robust histogram equalization transformation function. Experimental results show that the proposed method is not only can enhance the contrast, but this method also successfully sharpens the image.

Enhancement of Microscopic Images Using Modified Self-Adaptive Plateau Histogram Equalization

In this paper, we modified a method known as Self-Adaptive Plateau Histogram Equalization (SAPHE) to enhance microscopic images acquired using optical microscope. First, our method decides the plateau threshold value, automatically; based on the histogram itself. Then, using this plateau threshold, the bins of the histogram are modified. Finally, histogram equalization based on this modified histogram is carried out. Experimental results show that this modified SAPHE method is able to enhance the microscopic images without over amplifying the noise level, as compared with global histogram equalization.

Enhancement of Color Microscopic Images Using Toboggan Method

The images taken from an optical electronic microscope often contain data corrupted by noise, and the subject under study is normally blur due to improper focusing during image acquisition. Thus, in this paper, we propose the use of toboggan contrast enhancement method to solve this problem. First, the noise level in the image is reduced by using Gaussian filter. Gaussian filtering results more blur edges. Thus, toboggan contrast enhancement is applied to restore the edges. Experimental result, from an image of algae taken by an optical microscope, shows that toboggan contrast enhancement successfully sharpen the image.

Improving the visual quality of abdominal magnetic resonance images using histogram equalization

Histogram equalization is one of the well known enhancement techniques commonly used for medical images. Although easy to be implemented, global histogram equalization sometime produces unnecessary visual deterioration, such as saturation in intensity values. One of the available improved versions of histogram equalization is brightness preserving dynamic histogram equalization (BPDHE). This method has been proposed to be used in consumer electronics. In this paper, we investigate whether this method is also suitable to be employed in medical application or not, especially for abdominal magnetic resonance images. Our preliminary results show that in general, BPDHE can significantly improve the quality of medical data.

The effect of shape and weight towards the performance of Simple Adaptive Median filter in reducing impulse noise level from digital images

Recently, Simple Adaptive Median (SAM) filter has been introduced for the purpose of reducing the impulse noise level in digital images. SAM filter, which uses square filter as its basis, has an ability to change the size of the filter, spatially, based on the approximated local noise level. This paper investigates the effect of shape and weight into the performance of SAM filter in terms of quality improvement, and also execution time. In this work, SAM filter has been compared with its three new derivatives. These three new methods are Circular SAM (CSAM), Weighted SAM (WSAM), and Weighted CSAM (WCSAM). For the purpose of evaluation, fifty test images of size 1600×1200 pixels have been used as the test images. The results show that the performance of these methods is almost identical. All methods successfully restore the images corrupted up to 95% of impulse noise. Circular filter and weighting process slightly improve the performance of SAM. However, circular filter dramatically increases the processing time due to its complicated implementation.

Investigation on several methods to correct the intensity inhomogeneity in magnetic resonance images

Magnetic resonance image (MRI) has opened up new avenues of diagnosis and treatment that were not previously available. However, there are a number of artifacts which can arise in the MRI process and make subsequent analysis more challenging. The most undesired visual effect is the intensity in-homogeneity in MRI. Therefore, study on correcting this in-homogeneity problem is essential to extract information from MRI correctly. In this project, we present three methods to correct the in-homogeneity in MRI which are histogram matching, histogram equalization, and homomorphic unsharp masking method. Histogram matching and histogram equalization are used to distribute the intensity of the input MRI uniformly. Another method which is homomorphic unsharp masking method functions as a sort of band notch filter, where a certain spatial frequency range in the image is selected and removed. It removes low frequency components from an image and it does not alter the tissue boundaries. This method is the simplest method and it can be done by using different kind of filter such as mean and median. From the result provided by the methods mention above, some methods do remove the in-homogeneity of MRI successfully while some methods do not. As a result, different methods do correct the in- homogeneity of the MRI in different ways.

Image Enhancement with Noise Suppression Ability Using Histogram Manipulations

A simple contrast enhancement method based on histogram manipulation is presented. The method is based on histogram equalization and histogram stretching. The proposed method successfully enhances the microscopic image without over amplifying the noise level.