Recently in various fields, numerous researches are going on for the assessment of material damage on the basis of crack initiation and propagation. Various methods are available in NDT for this purpose, among which analysis using released acoustic emission (AE) waves due to crack propagation is very effective due to its dynamic monitoring features. Various approaches are proposed for long time to make it an ideal method for accurate monitoring of crack behaviors in materials. In fragmentation theory there are some proportionality among the relations of AE event, AE energy, area and volume of cracks etc., which are calculated from the released AE waves from any dynamic crack. It has been found that the necessity of calculating the fractal dimension is important in verifying these relationships. This parameter is emphasized for determining the geometry of the irregularity in crack surface and crack volume. In this paper a novel approach based on image processing is proposed to find out the fractal dimension for analyzing the crack propagation characteristics. Finally, the proportionality relationships of AE parameters with crack propagation behavior in ferrite cast iron under fatigue loading are demonstrated experimentally.