Using entropy information measures for edge detection in digital images

Shannon information entropy measures were used as filters of different kernel sizes to detect edges in digital images. The concept is based on communication theory with splitting of edge detection kernel into source and destination parts. The arbitrary shape of the kernel parts and the fact that information filter output is a real number with reduced problem of edge´s continuity represents the major advantage of this approach. The results are compared with traditional edge detection algorithms like Sobel to illustrate performance and sensitivity of the information entropy filters. Besides the well known test image Lena, the real life examples are taken from medical X-Ray imaging of knee joints in order to illustrate the algorithm performance on real data.