Adaptive gamma correction for automatic contrast enhancement of Chest-X-ray images affected by various lung diseases

• Published in: Multimedia tools and applications 2024-01, Vol.83 (29), p.73457-73475
• Main Authors: Yadav, Vivek Kumar, Singhai, Jyoti
• Format: Article
• Language: English
• Subjects: 1232: Human-centric Multimedia Analysis; Adaptive algorithms; Chest; Computer Communication Networks; Computer Science; Coronaviruses; Data Structures and Information Theory; Image acquisition; Image contrast; Image enhancement; Image quality; Lung diseases; Medical imaging; Multimedia Information Systems; Parameters; Performance evaluation; Special Purpose and Application-Based Systems; Statistical methods; Viral diseases; X ray imagery
• ISSN: 1573-7721; 1380-7501; 1573-7721
• DOI: 10.1007/s11042-023-18083-x

Lung and respiratory ailments are among the leading causes of illness and fatalities. Coronavirus disease (COVID-19), caused by the SARS-CoV-2 virus, has convinced the world that early and affordable detection improves treatment. X-ray imaging systems are inexpensive and widely available. Chest X-ray (CXR) images are inadequate due to the acquiring environment and technician skill. Hence, CXR image contrast enhancement is necessary for a correct diagnosis. Various lung diseases create variable spatial variation in CXR image contrast and brightness; hence, a single contrast enhancement procedure cannot improve it. In the proposed method CXR images are first classified into four categories depending upon their quality defined by their statistical parameters, before applying adaptive gamma correction for contrast enhancement. The performance of the proposed method is compared with existing methods on four datasets for five different types of lung diseases. The performance of the proposed algorithm is evaluated using parameters, such as Root Mean Square Contrast (RMSC) to determine the relation of contrast enhancement between the original and enhanced image, Contrast Improvement Index (CII) to measure the achieved contrast enhancement and Tenengrad which calculates the variation of intensity in the direction of maximum gradient descent . The qualitative and quantitative performance of the proposed method is found better than the existing methods for CXR images for all five lung diseases, which shows the stable performance of the proposed method and improvement in the processed images.