A Comparison of Thresholding Based Image Enhancement Techniques?

Image Enhancement is a vast area of Image Processing with its applications in different areas. Image Enhancement is used to transform digital images to enrich the visual information inside it. It is an initial operation for almost all vision and image processing assignment in several areas such as computer vision, biomedical image Processing, forensic image analysis, remote sensing and fault detection. In image enhancement certain transformations are applied upon an input image to obtain a visually more acceptable, more comprehensive and less noisy output image. In this paper four histogram equalization based Image Enhancement Techniques, CHE (Conventional Histogram Equalization), BBHE (Brightness Preserving Bi-Histogram Equalization), DSIHE (Dual Sub-Image Histogram Equalization) and MMBEBHE (Minimum Mean Brightness Error Bi-HE), are compared. All these techniques are based on partitioning of histogram of image and then equalizing each part separately. These techniques are assessed qualitatively and after examining output image visually, we see if it retains an appearance which is perfectly natural.