Toward under-specified queries enhancement using retrieval and classification platforms

Radiography images are used usually for diseases detection and fracture that can be visible on lateral view. Poor contrast of x-ray images do not provide momentous information concerning pathologies that are of interest to the radiologist. Magnification of contrast and sharpness of x-ray images will afford plenty and satisfactory visual information to radiologist and clinician and thus, allow better segmentation and indexing subsequent modules in the computer aided diagnosis (CAD) system for an autonomous disease diagnosis. Therefore, in this paper it intends to describe a new strategy to cater for under-specified queries enhancement using retrieval and classification platforms. In the retrieval platform (RPF), gamma correction (GC) function was employed on the under-specified query (USQ) image to generate dispersion versus location (DL) descriptor that measures the relationship between the local contrast and the local brightness, measured respectively with the help of estimators of location and dispersion. Subsequently, it employs appropriate near optimal search between the DL features of the USQ image and the corresponding similarity measurement in the archive database. In the classification platform (CPF), an approach was examined to predict the gain value of GC function using statistical pixel-level (SPL) features extracted from the radiography images along with the ANNs model classifier. The quality of the retrieved image is determined by referring to the USQ image. In addition, the problem of gain value estimation is transformed to a classification problem solved using an ANN model with three different measurement modes. Results indicated that the proposed approach can significantly improved image quality as confirmed by the DL descriptor which shown a more balance condition.