Pre-processing of the x-ray to increase the sensitivity of visual analysis
Keywords:low-contrast images, radiograph, histogram equalization, fuzzy methods, visual analysis, membership function, intensification operator
In the field of medical imaging, it is fundamental to improve medical images of different physical nature to increase the likelihood of diagnosis based on them. X-rays are one of the oldest techniques used to analyze dense tissue abnormalities. Insufficient quality of X-rays is due to both the physical characteristics of the equipment used and the process of their for-mation. There are two main approaches to digital image processing - spatial methods, which are based on direct manipulation of the pixels of the original image and frequency conversion methods. These image processing methods consider pixel values as exact constants, while there are objective reasons for the presence of digital uncertainties, which are due to loss of information when displaying objects from three-dimensional (3-D) space, to 2-D projections, uncertainty of the gray level, statistical randomness, etc. To account for these factors, new methods are currently being developed that are based on the ideas of ambiguity. This approach is a kind of nonlinear transformation that allows you to take into account factors that are ambiguous. Fuzzy methods are based on mapping gray brightness levels to a fuzzy plane using membership transformations. The image is represented as a mass of fuzzy sets relative to the analyzed property with the value of the membership function that varies in the range [0-1]. The aim of this article is to assess the impact on the quality of the bright characteristics of the X-ray image of the results of using a combination of spatial methods of histogram equalization, fuzzy intensification and improvement in the frequency domain. The proposed algorithm provides a redistribution of the brightness of the histogram in the middle range of gray levels, which corresponds to the best visual (according to Weber - Fechner's law) per-ception, allows to increase the contrast and resolution of the image. There is a significant effect on the result of improving the image of the parameters of fuzzy intensifications. Experimental results are given on the example of real images.
An analysis of x-ray image enhancement methods for vertebral bone segmenta -ion IEEE 10th International Colloquium on Signal Processing & its Applications (CSPA2014) (7 - 9 Mac. 2014, Kuala Lumpur, Malaysia). 2014.
M. S. V Sokashe, “Computer assisted method for cervical vertebrae segmentation from x-ray images,” Int. J. Adv. Res. Comput. Commun. Eng., vol. 2, no. 11, 2013, pp. 4387–4389.
Pratt W.K. Digital Image Processing / Pratt W.K. – New York; – Chichester; – Weinheim; – Brisbane: John Wiley and Sons Inc., 2001. – 723 р.
Honsales R. Tsyfrova obrobka zobrazhenʹ/Honsales R., Vuds R.; [Per. z anhl. za red. P.A.Chochia]. - M.: Tekhnosfera, 2006. -1070 s.
Сhi Z. Fuzzy algorithms: With Applications to Image Processing and Pattern Recognition / Сhi Z., Yan H., Pham T. – Singapore; – New Jersey; – London; – Hong Kong: Word Scientific, 1998. – 225 p.
Horst Haußecker. Handbook of Computer Vision and Applications. -V. 2. Signal Processing and Pattern Recognition / Horst Haußecker, Hamid R. Tizhoosh.- Academic Press. −1999. −722р.
Ahmetshina L.G., A.A. Egorov. Nechetkaya klasterizaciya polutonovyh izobr-?zhenij na osnove preobrazovaniya iskhodnyh dannyh //Iskusstvennyj intellekt.− Kiev, − 2018. № 3.− S. 36-43.
Ahmetshina L.G., S.K. Mitrofanov. Polіpshennya і segmentacіya slabokontrastnih zobrazhen na osnovі metodu nechіtkoї іntensifіkacіi //materіali ІV Vseukr. nauk.-prakt. konf. 27–29 listopada, 2019. Dnipro, Ukraїna. Perspektivni napryamki suchasnoї elektronіki, іnformacіjnih ta komp’ yuternih sistem (MEISS’2019), -
This work is licensed under a Creative Commons Attribution 4.0 International License.