基于SSA-CatBoost的颜色校正模型仿真
Simulation of Color Correction Model Based on SSA-CatBoost
沈楠翔 1顾文娟 1李志文 1王洪成1
作者信息
- 1. 昆明理工大学机电工程学院,云南 昆明 650504
- 折叠
摘要
针对图像采集设备在采集过程中出现的颜色失真以及传统方法存在校正效果不佳的问题,提出了一种结合SSA算法的CatBoost颜色校正模型.首先,建立三个CatBoost模型,并使用SSA算法分别搜索三个模型的最佳参数;然后将所得到的最优参数赋值给三个CatBoost模型,并分别对三通道的测量值与真实值进行拟合;最后,与传统多项式回归方法进行对比,校验此模型校正效果.模型仿真结果表明,上述模型可将色差值ΔE控制在 3 以内,证实了SSA-CatBoost模型具有更好的颜色校正.
Abstract
In order to solve the problem of color distortion in the process of image acquisition equipment and the traditional method has poor correction effect,a CatBoost color correction model combined with SSA algorithm is pro-posed.Firstly,three CatBoost models were established,and the SSA algorithm was used to search the best parameters of the three models.Then the obtained optimal parameters were assigned to three CatBoost models,and the measured values of the three channels were fitted with the real values respectively.Finally,compared with the traditional polyno-mial regression method,the correction effect of this model was verified.The model simulation results show that shows that the model makes the color difference value(Δ E)is controlled within 3,which proves that the SSA-CatBoost model has better color correction.
关键词
颜色失真/麻雀搜索算法/对称决策树/颜色校正Key words
Color distortion/Sparrow search algorithm/Oblivious trees/Color correction引用本文复制引用
出版年
2024
NETL
NSTL
万方数据