Abstract—We have proposed a very simple yet highly reliable face recognition algorithm using Adjacent Pixel Intensity Difference Quantization (APIDQ) histogram previously. In this paper, we present a modified quantization method to improve recognition performance. After the intensity variation vectors for all the pixels in an image are calculated, each vector is quantized directly in (dIx, dIy) plane instead of r-θ plane. By counting the number of elements in each quantized area in the (dIx, dIy) plane, a histogram can be created. This histogram, obtained by APIDQ for facial images, is utilized as a very effective personal feature. Furthermore, by utilizing rough location information of facial parts, the facial area is divided into 5 individual parts, and then APIDQ is applied on each facial component. Recognition results are firstly obtained from different parts separately and then combined by weighted averaging. Experimental results show that the top of recognition rate of the whole research groups is achieved by using FB task of the publicly available face database of FERET.

Index Terms—Face recognition, adjacent pixel intensity difference quantization (APIDQ), histogram.

Feifei Lee, Qiu Chen, and Tadahiro Ohmi are with New Industry Creation Hatchery Center, Tohoku University, Sendai, 980-8579 Japan (e-mail:fei@fff.niche.tohoku.ac.jp).
Koji Kotani is with Department of Electronics, Graduate School of Engineering, Tohoku University, Sendai, 980-8579 Japan.

Cite: Feifei Lee, Koji Kotani, Qiu Chen, and Tadahiro Ohmi, "Face Recognition Algorithm Based on Adjacent Pixel Intensity Difference Quantization in Rectangular Coordinate Plane," International Journal of Computer and Electrical Engineering vol. 4, no. 4, pp. 554-557, 2012.