Method for automatic assessing of the gradient filter threshold for fast processing of dynamic scenes objects

In this paper, we present a new method based on the use of information provided by the gradient methods for determining the geometric parameters of objects with high accuracy. The algorithm is based on the use of data obtained after image processing by gradient filters. Also, it reacts to the slightest change in the contours of objects of the dynamic image scenes. Repeated experiments using more than 5000 real images were processed to improve the theory. Given a high refresh rate of modern systems, a position of the Center Of Gravity (COG) in the dynamic images is changing gradually even for rapid motion. Using this feature, COG for each frame of a training sample is defined under various threshold values by means of an algorithm. A number of elements (frames) in the training sample are selected depending on the type of the dynamic object, a task set and on the initial conditions. The suggested method is recommended for further use by the expert system, in parallel with its own operation, with a goal to maintain a threshold value on the optimal level in case of dynamic perturbing factors. After the research, we found that the prediction accuracy increased that essentially improved results. A number of experiments demonstrated increasing the accuracy of determination of the center of blurred objects. Also, we have eliminated the human factor. All of the calculations are done automatically. These data are very useful and important for all areas of science where high accuracy of the results is necessary.