Pneumoconiosis X-ray Image Enhancement

This paper focuses on the study of Fuzzy Enhancement and Multi-Channel Adaptive Homomorphic Enhancement, and the applications in pneumoconiosis X-ray films enhancements. Pneumoconiosis is one of the occupational diseases with widely influence and severe hazard. And there is no effective cure so far. Reference occupational diseases history and X-ray film examination are chiefly adopted to discern and diagnose pneumoconiosis. However, X-ray film features an excessively broad range, over -exhaustive details and poor contrast. Also, the pathologic changes of early pneumoconiosis symptoms are not obvious. These factors are disadvantages to early diagnoses of pneumoconiosis. With the help of digital image enhancements, the contrast of pneumoconiosis X-ray film can be enhanced, and such early symptoms will be emphasized with better clarity in the feedback. Thereby, it is helpful to reduce the misdiagnosis rate.As a powerful tool for processing uncertain problems, Fuzzy sets theory now has been used in image processing and pattern identification successfully. Improved fuzzy algorithm simplifies the complex transforms and inverse transforms in Pal algorithm and it adopts novel enhancement operators. The threshold value of membership grade can be adjusted dynamically according to image types. Thus it overcomes threshold as a fixed number and reduces iterative times. Adaptive Fuzzy Contrast Enhancement Based on Homogeneity Detection (HD-AFCE) uses image fuzziness and visual property that sensitiveness to noise of human eyes is higher in area of homogeneity than in area of detail to enhance local contrast of image and suppress amplification of noise effectively. First the image is mapped to fuzzy region. Then we detect the image homogeneity and self-adaptively change contrast magnification coefficients. Multi-channel Adaptive Homomorphic Enhancement (MC-AHME) uses the following properties: nonlinearity, multi-channel filtering and sensitivity to high frequency information. This method can enhance image details of various sizes effectively. Firstly the image is transformed to logarithm region, then is filtered by three sizes filter groups. Afterwards, we utilize LF and HF coefficients to enhance the filtering results. At last we get the average of three channels and map the image to the grey region. The results indicate that with enhancement processing on silicosis' X-ray films image contrast improves obviously and the whole contour is distinct. Pneumoconiosis shadows are prominent. Thus it can help the doctors to interpret and can be propitious to pneumoconiosis forepart diagnoses.