International Journal of Computer Applications (0975 – 8887) Volume 98 – No.20, July 2014 44 0 200 400 600 800 1000 1200 1400 1600 Intensity 0 50 100 150 200 250 0 200 400 600 800 1000 1200 1400 1600 Intensity 0 50 100 150 200 250 0 200 400 600 800 1000 1200 1400 0 50 100 150 200 250 Analytical Adjustment of Image Contrast Tajbia Karim Department of Electrical and Electronic Engineering American International University-Bangladesh Banani, Dhaka, Bangladesh Tasnuva Tasneem Department of Electrical and Electronic Engineering American International University-Bangladesh Banani, Dhaka, Bangladesh ABSTRACT Detailed analysis based on histogram stretching for image contrast adjustment has been performed. However unprocessed image includes unwanted noise which needs to be reduced for performing the stretching accurately. The level of noise reduction greatly effects the contrast adjustment and varies from images to images depending on the intensity of noise itself. This paper discusses about image contrast adjustment technique along with the algorithm. Results are analyzed corresponding to images of different quality based on histogram stretching considering different level for the noise reduction. General Terms Histogram stretching, Visual quality, Image contrast Keywords Intensity histogram, Underwater image, Medical image, Contrast adjustment 1. INTRODUCTION Contrast adjustment is an important part of image processing [1]. Contrast of an image can be determined by the ratio between the brightest and the darkest pixel intensities [2]. Unprocessed images having very low contrast are not suitable for human eyes to read. By improving the image contrast, images can be made more suitable for human vision [3]-[6]. This sort of image processing can play a vital role in digital photography, medical images and LCD display images. Adjusting contrast means simply making light colors lighter and dark colors darker simultaneously. It can be done by setting all color components below a specified lower bound to zero, and all color components above a specified upper bound to the maximum intensity (that is, 255)[7]. An efficient method is used here that is called histogram stretching [8]. A histogram can be defined as the probability distribution of the pixel values in an image. It represents the frequency of occurrence of all the gray levels in an image [9]. It is broken into three histograms of the three component channels for RGB images. Two operations affecting the pixel values are used in the stretching method. They are- i) Adding a value to all the pixels adds that amount to the histogram, visually this shifts the histogram ii) Multiplying all the pixel values by a certain amount of scales where the histogram data appears, visually this stretches the histogram. The rest of the paper is ordered as- Section 2 describes about the detailed methodology that has been used to perform the analysis. Section 3 is the result and analysis part. A comparison is shown by varying the amount of pixel removal for an underwater image that has been captured in an adverse environment in section 3.1. It also shows that effect of excessive amount of pixel removal can result an unnatural image. Section 3.2 includes the histogram stretching effects on images of different categories including a medical image. Experimental results are shown in section 3.3. Discussion and Conclusion are included in Section 4 and 5 respectively. 2. METHODOLOGY To extract maximum information from an image it is necessary to distinguish between different objects and the background. Contrast adjustment is a conventional method to meet up the purpose which can be performed efficiently by appropriate histogram stretching. Unfortunately, raw images, specially captured at adverse environment, suffer from noises that can affect the further stretching which is needed for the purpose of contrast adjustment. Therefore, noise reduction should be carried out before the stretching technique. The detailed method can be summarized as follows- Step-I: Plotting Intensity histogram of an image Step-II: Selecting portions to remove/shift (over red arrow) Pixel counts Intensity Pixel counts Pixel counts Intensity Intensity Step-III: Stretching the remaining pixels