TAILIEUCHUNG - Urban Transport and Hybrid Vehiclesedited Part 4

Tham khảo tài liệu 'urban transport and hybrid vehiclesedited part 4', kỹ thuật - công nghệ, cơ khí - chế tạo máy phục vụ nhu cầu học tập, nghiên cứu và làm việc hiệu quả | Computer Vision Techniques for Background Modelling in Urban Traffic Monitoring 53 The following index sets have been considered as a valuable quantification of relative performance of each algorithm S Ịsf Sj 1 Sy E . 25 E0 50 E0 75 . The first set includes fitness coefficients with ideal value equal to 1 while the second set includes fitness errors with ideal value equal to 0. The values of each one of these coefficients will be averaged for all the chosen frames. In addition the typical deviation of each one is calculated. It can be noticed that the Sec coefficient has been dropped from the analysis as in agreement with the above comments it exhibited a very poor sensitivity yielding very high scores for every algorithm. Table 5 details the average ụ and typical deviation Ơ for the chosen fitness indexes and the traffic light video sequence. According to the authors in Manzanera Richefeu 2007 parameter N 2 was recommended for the SD SDSP and SDM algorithms. However in our experiments N 4 performed slightly better in some videos. Therefore both values have been considered for each one of the four sigma-delta alternatives. Results of the proposed sigma-delta algorithm with confidence measurement are clearly on top according to the fitness coefficients both for N 2 and N 4. On the other hand according to the fitness error coefficients the proposed algorithm is significantly better than any of the other algorithms featuring also a moderate typical deviation. Algorithm Sf Sj 1 Sy E Ơ E Ơ E Ơ E Ơ E Ơ E Ơ SD N 2 0 44 0 156 0 29 0 125 0 66 0 072 0 36 0 143 0 30 0 115 0 22 0 078 SD N 4 0 59 0 172 0 44 0 161 0 87 0 062 0 45 0 152 0 36 0 124 0 26 0 087 SDSP N 2 0 63 0 180 0 48 0 177 0 92 0 027 0 42 0 168 0 34 0 137 0 24 0 097 SDSP N 4 0 56 0 185 0 41 0 167 0 94 0 018 0 50 0 151 0 41 0 123 0 29 0 087 SDM N 2 0 42 0 160 0 28 0 126 0 66 0 075 0 40 0 141 0 33 0 114 0 24 0 078 SDM N 4 0 55 0 169 0 40 0 153 0 87 0 064 0 49 0 142 0 40 0 116 0 28 0 082 SDC

TỪ KHÓA LIÊN QUAN
Đã phát hiện trình chặn quảng cáo AdBlock
Trang web này phụ thuộc vào doanh thu từ số lần hiển thị quảng cáo để tồn tại. Vui lòng tắt trình chặn quảng cáo của bạn hoặc tạm dừng tính năng chặn quảng cáo cho trang web này.