期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2019
卷号：97
期号：21
页码：3132-3143
出版社：Journal of Theoretical and Applied
摘要：Images with low contrast may have intensity properties that make it difficult to use the traditional algorithms and approaches to enhance its contrast. Image�s histogram may contain high frequency values in specific area in the image and low frequency in the remaining area. That leads to inconsistency in the histogram and results in image with unacceptable contrast. Two algorithms are proposed to solve this problem, namely MFGLG and ACEIF/MFGLG approaches. The first approach is Modified Fast Gray Level Grouping (MFGLG). It is a modification of Fast Gray Level Grouping (FGLG). MFGLG uses two sets of gray level bins and uses them as a startup bins assigned to the histogram. The proposed approach results in more efficient contrast enhancement compared to FGLG. The algorithm has no user intervention. Moreover, the proposed algorithm can be applied to vast range of contrast perturbed images. It also can be applied to images with the highest frequency in the histogram concentrated in any image location. The second approach is Automatic Contrast Enhancement Image Fusion based on Modified Fast Gray Level Grouping (ACEIF/MFGLG). With ACEIF/MFGLG, the output of MFGLG is fused with the original image to get more detailed image. The original image is used to provide more accurate contrast and intensity. The proposed two approaches are applied on low-contrast images and provides high quality images. The algorithm can be applied without human intervention with acceptable speed compared to the other methods in literature.
