Automatic analysis of human facial expression is one of the challenging problems in machine vision systems. The most expressive way humans display emotion is through facial expression. In this paper, we extend texture based facial expression recognition, with a method of 2D image processing implemented for extraction of features and a new neural network based decision trees. The algorithm applies a set of preprocessing and divides the image into two main parts (eyes and lips apart), and implements Discrete Cosine Transform (DCT) on each part to reduce image data size in different parts of the face. Different decision tree models have been tried in order to find the best recognition rate. Experimental results show that, such a combination of decision tree with neural network to identify different facial expressions improves the recognition rate significantly.