Machine Intelligent System Algorithm to Recognize Different Shaped Color Targets


Geometrical shape plays a vital role in computer vision applications. In this study, a new suggested method used to detect geometric shapes (square and circle) of different colors (red, green, and blue), and different arrangements within the same input image. The introduced algorithm classified 256 tested input images at the same time and feedback the recognized index number to each individual image. These tested images have different details for the same color target by means occurrences like different in orientation, position, and partially appeared shape. A new idea of indexing number used to describe shape and color of input target images for all possibility of occurrences. There were 42 possible arrangements of image cases because of the color and shape targets that used in this study. The introduced algorithm detects shape, and color with an accuracy of 100%.