본문 바로가기

AI73

5.3 Generalized Hough Transform In this class, we willl learn about generalized Hough Transform. ● Hough Circles Let's find circles by Hough Transform. Equation of circle is If radius is known, we should consider 2D Hough space and Accumulator array would be A(a, b). A point in 2D space becomes a circle in Hough space. More points become more circles. If radius is not known, we should consider 3D Hough space. In this case, the.. 2020. 8. 7.
5.2 Hough Transform In this class, we will learn about Hough Transform. How do we find image boundaries (lines)? ● Hough Transform - Hough Transform is a generic framework for detecting a shape/object. - Edges don't have to be connected. - Lines can be occluded. - The key idea is 'Democratic' Detector. → Each image edge votes for the possible line model. ● Image and parameter space We can see the image as parameter.. 2020. 8. 5.
5.1 Lines Parameterization In this class, we will learn about lines parameterization. ● Slope intercept form We can express line in slope intercept form. Slope intercept form uses slope and y-intercept to express line. ● Double intercept form In another way, line can be expressed in double intercept form. Double intercept form uses x-intercept and y-intercept to express the eqation of line. ● Normal Form Finally, normal f.. 2020. 8. 5.
5.0 Defining Boundaries Today, we will learn about how to define boundaries. What are the object boundaries? This picture would be Human annotated boundaries. By using Edge Detection technique, we can get this kind of edge. By using multi-scale edge detection, we can get this picture. Edge strength does not necessarily correspond to our perception of boundaries. These are pictures we can feel more gap between human ann.. 2020. 8. 5.