Dynamic textures and covariance stationary series analysis using strategic motion coherence

Dynamic texture describes images sequence that continuously demonstrates movement of pixels intensity change patterns in time, for example, smoke, fire, waterfall, sea-waves, foliage, traffic on highway and so on. Motion coherence analysis on dynamic textures is usually observed through their motion vector fields. We implement strategic motion coherence analysis to evaluate the coherent motion on dynamic texture. In this article, motion coherence index is proposed as a metric to evaluate the coherent motion. It represents corresponding angles quantity of vectors on the motion vector field. The coherent motion in a video frame will be calculated and accumulated from every four adjacent vectors. The average of the accumulation represents the motion coherence index of dynamic texture in the video frame. The coherence index is utilized as an attribute to distinguish the difference of dynamic textures. We propose a new technique to distinguish dynamic texture by investigating the covariance stationary series of motion coherence index. Our investigation shows that, there are significant differences among the series of motion coherence index of dynamic textures. For instance, dynamic fire texture and dynamic smoke texture can be distinguished by the covariance stationary and covariance non-stationary of motion coherence index series respectively.