A Perceptual-Based Robust Measure of Image Focus

The focus measure is commonly used in the passive autofocus imaging system, which gives a sign of the image focus level. However, most methods are sensitive to image noise. The construction of an anti-noise focus measure is the primary concern of this letter. For this purpose, the core of the algorithm is to quantify the image sharpness information under noise conditions. Inspired by the mechanism of the low-level visual system, we proposed a perceptual-based robust focus measure based on the local difference of Gaussian (DoG). The monotonicity, effective range and SDA index are used to assess the performance of the proposed method. The experimental results under different types of random noise show that the robustness of the proposed method outperforms other competing methods.