Local Dominant Directional Symmetrical Coding Patterns for Facial Expression Recognition

Figure 2

The coding sketch of LDDSCP encoder. Two groups of symmetrical direction masks are convoluted with the original expression image, and the corresponding convolution values of each group are presented in the middle of sketch. Then, we encode the dominant direction information by comparing each convolution value with the average strength of its belonging group and obtain the LDDSCP-1 code and LDDSCP-2 code subsequently (shown in the right).