Simple and effective method to quantify the optical performance of camera phones

Abstract

A simple method based on binary line pair patterns to evaluate the image quality of camera phones is proposed in this paper, where four different spatial resolution line pairs are located at four different quadrants respectively. The value of Modulation Transfer Function (MTF) of the camera phones can be tested at four different spatial resolutions for the entire field of view (FOV) simultaneously. In addition, other key parameters of optical imaging quality criteria, such as image distortion, relative illumination can also be characterized in the same time. Moreover, the possible color moiré due to color aliasing from RGB channels can be detected and evaluated using the same test chart. Wavelet transforms on the gray level of test images based on the Haar wavelet and the Meyer wavelet are implemented in this paper. The test image is decomposed into the space Vj, (j=-3,-2,-1). The j=-3 corresponds a lower resolution space. Using the multi-resolution wavelet analysis, the 100lp/mm region can be partially recovered at a low resolution space, for example at the space V-2. The results are shown. Some noisy ripples at the original image are also magnified in a lower resolution space.