Algorithm 2: Histogram of Oriented Gradient (HOG)

Input: Upper face image, boundary coordinate, size of blob Output: HOG feature of each boundary coordinate image $\leftarrow$ Upper face image x_landmark $\leftarrow$ x coordinate of boundary y_landmark $\leftarrow$ y coordinate of boundary distance $\leftarrow$ size_of_blob/2 for i=0 to length of boundary coordinate do   if x_landmark[i] > distance then     x_start_blob $\leftarrow$ x_landmark[i] − distance   else     x_start_blob $\leftarrow$0   end   if y_landmark[i] > distance then     y_start_blob $\leftarrow$ y_landmark[i] − distance   else     y_start_blob $\leftarrow$ 0   end   x_end_blob $\leftarrow$ x_start_blob + size_of_blob   y_end_blob $\leftarrow$ y_start_blob + size_of_blob   blob $\leftarrow$ image[x_start_blob, x_end_blob][y_start, y_end_blob]   hog_feature[i] $\leftarrow$ HOG(blob) end for