.png){kind=icon}
返回
Illumination correction via optimized random vector functional link using improved Harris hawks optimization
.png){kind=icon}
PRE
AI
DOI：10.1007/s11042-022-11986-1.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
To enhance the accuracy of illumination estimation, this study proposes illumination correction using a modified random vector functional link (RVFL) algorithm based on the multi-verse optimizer (MVO)-improved Harris hawks optimization (HHO). The MVO is first utilized to initialize a set of optimized populations for the HHO algorithm, enhancing the real-time performance and improving the accuracy of the HHO algorithm. Further, the MVO-HHO is used to determine the optimal parameters of the RVFL, i.e., the input weights and biases, increasing the prediction accuracy and stability of the RVFL. After the predicted illumination information is obtained, the image can be restored through diagonal transformation. Through comparative experiments, the average chromaticity error of illumination estimation with the proposed MVO-HHO-RVFL algorithm is 0.025959, which is 26.29%, 32.55%, and 25.27% lower than those of the improved RVFL based on the HHO, improved extreme learning machine based on the HHO, and improved backpropagation based on the Levenberg-Marquardt algorithms, respectively. The obtained results demonstrate that the proposed algorithm effectively improves the accuracy of illumination estimation and that there are significant differences between it and the other algorithms.
Keyword:
Illumination correction
Multi-verse optimizer
Harris hawks optimization
Random vector functional link
Diagonal transformation
期刊
IF:
3
论文数:
2.0W
被引数:
3.2W
机构
引用论文
分享收藏
Latent tuberculosis infection is associated with increased unstimulated levels of interferon-gamma in Lima, Peru.png)
PLOS ONE
OAAI
PLOS ONEIF0
OAAI分享收藏
查看更多内容