LCD character defect detection algorithm based on multi-feature moment fusion

Table 1. Hu moment under different poses
View table
View in Article
Table 1. Hu moment under different poses
Hu矩姿态
A B C
I₁ 0.690 0 0.697 3 1.000 4
I₂ -0.097 3 -0.041 1 1.181 2
I₃ 0.484 3 0.452 4 -0.884 9
I₄ 0.183 1 0.236 9 -1.817 1
I₅ 0.472 4 0.518 4 -3.750 1
I₆ 0.133 3 0.216 1 -0.976 6
I₇ 0.150 3 0.282 7 -2.569 5

Table 2. Zerniek moment under different poses
View table
View in Article
Table 2. Zerniek moment under different poses
Zernike矩姿态
A B C
$Z_{6.0}$ 1.872 4 1.883 2 0.990 3
$Z_{7.1}$ 0.033 0.037 0.111 6
$Z_{8.0}$ 2.317 1 2.378 3 1.672
$Z_{9.1}$ 0.057 2 0.059 4 0.040 7
$Z_{10.0}$ 2.878 5 2.903 3 1.92
$Z_{11.1}$ 0.046 4 0.048 2 0.059 9

Table 3. Color moments under brightness defects
View table
View in Article
Table 3. Color moments under brightness defects
颜色矩 $F_{c o l o r}$ 正常亮度过高亮度
$μ_{H}$ 0.367 4 0.226 6
$σ_{H}$ 0.684 9 0.572 1
$S_{H}$ 0.349 8 0.519 9
$μ_{S}$ 0.057 9 0.050 8
$σ_{S}$ 0.254 6 0.351 7
$S_{S}$ 0.329 8 0.459 9
$μ_{V}$ 0.052 6 0.085 7
$σ_{V}$ 0.103 4 0.377 6
$S_{V}$ 0.346 3 0.180 2

Table 4. Gray moments under brightness defects
View table
View in Article
Table 4. Gray moments under brightness defects
灰度矩G 正常亮度过高亮度
$m$ 70.938 2 116.086 3
$σ^{2}$ 6 757 10 196
$e$ 6.683 5 6.306 7
$ξ$ 78.331 3 60.614 6

Table 5. Comparison of fusion features and single features
View table
View in Article
Table 5. Comparison of fusion features and single features
误判率/% 漏判率/% 效率/s
Hu矩 12 6 0.5
Zernike矩 8 2 0.6
颜色矩 14 0 0.7
灰度矩 9 0 0.5
本文算法 1 0 0.6

Tools

Get Citation

Copy Citation Text

Xin CHEN, Cheng-gang FANG. LCD character defect detection algorithm based on multi-feature moment fusion[J]. Chinese Journal of Liquid Crystals and Displays, 2022, 37(10): 1326

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites