Page 40 - Read Online
P. 40

Tang et al. Intell Robot 2022;2(2):130­44  I http://dx.doi.org/10.20517/ir.2022.07  Page 138

                                              Table 1. Parameters of the proposed method

                                      Parameters  Values    Remarks
                                                    5     A given threshold in Equation (3)
                                                    1     A given threshold in Equation (5)
                                        0          20     The initial detection radius
                                                   16     The initial update rate
                                        0
                                                   0.2    A given threshold in Equations (6) and (7)
                                        0
                                                    1     A given threshold in Equation (9)
                                        1
                                                   0.2    A given threshold in Equation (11)
                                        2
                                                   0.7    A given threshold in Equation (11)
                                        3
                                                   43     A given threshold in Equation (12)
                                        1


                     Clip1






                     Clip2


                                (a)         (b)          (c)         (d)          (e)          (f)
               Figure 4. The moving object detection experiments on the video Walk (Clip1) and Bungalows (Clip2): (a) the first frame; (b) the frame for
               detection; (c) the ground-truth; (d) the result of GMM; (e) the result of G-ViBe; and (f) the result of I-ViBe.


                            Table 2. The valuation of the three methods for moving object detection in Walk and Bungalows
                            The valuation    The video clip of Walk    The video clip of Bungalows
                               indices  GMM  [32]  G-ViBe  [35]  I-ViBe  GMM  [32]  G-ViBe  [35]  I-ViBe
                            SP           0.9995  0.9804     0.9985   0.9310    0.9401   0.9854
                            RE           0.7758  0.9483     0.9612   0.8570    0.7999   0.9636
                            FPR          0.0004  0.0195     0.0014  0.0689     0.0598   0.0145
                            FNR          0.2241  0.0516     0.0387   0.1429    0.2000   0.0363
                            PWC          0.0060  0.0203     0.0021  0.0826     0.0879   0.0186
                            PRE          0.9779  0.5647     0.9272   0.7395    0.7702   0.9390
                            F            0.8652  0.7079     0.9493   0.7939    0.7847   0.9511



               effectively than the other two methods (see Figure 4e,f).


               4.2. The experiment for multiple objects detection
               To test the performance of the proposed approach in multiple moving objects detection, two experiments
               were conducted on the dataset Highway (Clip1) and People (Clip2). The results are shown in Figure 5, and the
               evaluations for the three methods in this experiment are shown in Table 3.



               The results of the experiment on Highway (Clip1) show that there are lots of errors based on the GMM method
               and the general ViBe method, because there are some leaves shaking in the background having similar color
               attribute with the vehicles. However, the proposed approach can deal with this problem efficiently, which is
   35   36   37   38   39   40   41   42   43   44   45