Shadow elimination based on multiple feature differences and glvq
Yaomin Hu, Weiming Liu · 2021
Aimed at problem of that shadow elimination methods based on model or single feature are all having inherent drawbacks. A shadow removal algorithm based on multiple feature differences between pixels and the reference background pixels is proposed. The multi feature differences of the current pixel and the corresponding pixel of the background are inputted into GLVQ to classify the current pixel.The experiments proved the proposed method has better performance than that of some other methods.