Sample Question 1 You are given four binary objects that belong 0 0 0 0 0 0 0 0 0 0 to 2 classes (A & B) as shown in figure 14, 0 0 0 0 0 0 0 0 1 1 0 0 1 01 1 0 15, 16 and 17. Your objective is to develop a 1 1 1 0 computer program to automatically classify 0 1 1 1 0 1 1 1 1 1 those binary objects. Propose a feature and 0 1 1 1 0 0 1 1 1 0 briefly explain on how the proposed feature 0 0 1 0 0 0 0 0 0 0 can be extracted and used to classify those 0 0 0 0 0 0 0 0 0 0 objects automatically. Calculate the proposed Figure 14:1 Object A1 feature for object 1 to 4 and verify if it can help to achieve your objective based on the calculated results. [15 marks] Figure 15:1Object A2 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 1 1 1 0 0 1 0 1 0 0 1 0 0 0 0 0 0 1 0 0 1 1 1 0 0 1 0 1 0 0 1 0 0 0 Answers: 0 1 1 1 0 0 1 1 1 0 Feature to be used: Thinness’ ratio 0 0 0 0 0 0 0 0 0 0 A T 4 2 [2] Figure 16: Object B1 Figure 17: Object B2 P How to extract the feature: Area, A is defined as the total number of pixels that forms the object. [1] Perimeter P is defined as the total pixels that constitute the edge of the object. It can be done by scanning for the following pattern, and set the center pixel to 0 when it is found. Next calculate the area of the object with edge pixel only and the result is the perimeter. [3] 0 1 0 1 1 1 0 1 0 How to use the feature: Since object of class B is thinner compare to those of class A, the thinness ratio of object from class B would be lower compared to those from class A [3] Calculate the feature for each object: 11 Object 1(Class A): A1 = 11, P1 = 8 => T1 4 2 2.16 8 11 Object 2(Class A): A2 = 11, P2 = 8 => T1 4 2 2.16 8 11 Object 3(Class B): B1 = 11, P3 = 11 => T1 4 2 1.14 11 11 Object 4(Class B): B2 = 11, P4 = 11 => T1 4 2 1.14 11 [2] Since the thinness ratio for object from different class are different and is the same for object of similar class, it can be used to achieve the objective to classify the objects automatically. [4] Sample Question 2 Propose a binary object feature (other than projection) that can differentiate object A (figure 6) and object B (figure 7) shown below. Briefly explain how the proposed feature can be extracted and how it can be used to differentiate the two objects automatically. [6 marks] 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 1 1 0 0 1 0 1 0 0 1 1 1 0 0 0 1 0 1 0 0 1 1 0 0 0 1 0 1 0 0 1 1 0 0 0 1 1 1 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 Figure 6: Object A 1 Figure 7: Object B Since object A has hole and B has no hole [1], Euler number can be used to differentiate the two objects[1]. Object A should have smaller Euler number compared to object B. [1] Euler number = number of objects – number of hole s [1] Euler number can be approximate by scanning the number of concavities and convexities[1] (see patterns shown below) in the image. 0 0 0 1 0 1 1 1 Convexities Concavities [1] Euler number number of convexities – number of concavities [1] Object A: Euler number = 1 – 1 = 0 (smaller) Object B: Euler number = 1 – 0 = 1 [1]