A description of a twodimensional light intensity function f(x,y), where the letters x and yrepresents spatial co-ordinates and the ‘f’value at every point is directly relational to the gray level of the image atthat point is known to be an image 1. Digital image processing methods altersthe deficiency in an image into a modified improved version. From computer science,image processing is some process of signal processing which takes on an imageor frames of videos as the input and outputs either an image or set ofparameters that is connected to the image 1. Lately, image enhancement hasbecome one of the central topics in image processing. Image enhancement includesa collection of skills that are used to improve and develop an image visual appearance- a figure description in fig 1- or a conversion of the image to an improved formfor suitability and understanding of human and machine interpretation. Figure1: An enhanced imagedescription Several kinds of image andpictures are employed as the basis of information in recent communicationsystem and broad-based applications. Most images taken may have inherit some formof degradation such as blurred image. Similarly, an image converted fromcertain form to another such as scanning, transmitting, storing and a lot moremay also inherit some form of distortion.
Mostly the degradation may occur atthe output and therefore the need to enhance the output image for an improvedvisual appearance and appealing.The chief goal ofenhancement of an image is to process the image such that the outcome serve thepurpose for which it was intended for a specific target or set application. Inmost situations, the end effect of image enhancement can be recognizedvisually. Although enhancement of image is very challenging issue in manyresearch and application areas, this aspect of image processing is a necessityand needs expanded research to address certain technical challenges that occurin daily operation. Image enhancement techniques is needed to address or handle regular objects with regardsto image geometric transformations or recover certain characteristics by adjustingthe intensities or colors for the processing of medical images and other areasof application of image processing as biometric image processing, satelliteimage processing etc. There has not been any general theory of imageenhancement partly due to the fact that no universal standard for the qualityof an image has been proven.
Once an image is taken for visual interpretation,the eyewitness is the definitive judge of how well a specific approach works. Thusleading to the development of varying classes of techniques over the pastdecades 2 and 3.A full proof implementationof gray level image transformation is possible but a challenging task. In 2001,to automatically find a measure for central tendency of an image brightnesshistogram, 4 created an automatic contrast enhancement technique by shiftingand converting the histogram appropriately.
A contrast enhancement algorithm basedon curvelet was proposed by 5 which tapped the properties of curvelet forcontrast stretching for effective enhancement of image contrast. This approach isvery efficient for correcting images which may contain noise. A cumulativefunction was put forward by 6 to be used in together with histogramequalization to realize contrast enhancements. In this paper an implementationof gray level image enhancement techniques is undertaken. A software programdeveloped in Matlab is used to apply the techniques studied. The remaining ofthe paper is organized as follows: In section 2 Gray level image transformationtechniques is introduced followed by section 3 with a brief explanation of theapplication of Matlab and its implementation of the software. Section 4 detailsthe simulation experiment and results.
Finally the paper concludes with Section5.
