Segmenting focused objects based on the Amplitude Decomposition Model

In this paper, we present an inherent model of the low depth-of-field (DOF) images, referred as the Amplitude Decomposition Model, which turns out to be useful for the detection and segmentation of focused objects in the low DOF images. By analyzing the low DOF image in frequency domain, the Amplitude Decomposition Model is firstly investigated, i.e., the amplitude spectrum of the low DOF image can be decomposed into the amplitude of its totally defocused version and the high-frequency difference amplitude of its focused regions. Based on this model, we propose a method for detecting focused objects. Using the detection result, we then utilize a thresholding method to segment the focused objects and employ the graph cut technique to refine the focused object boundary. Experimental results show that the proposed method can extract focused objects effectively and is comparable to the state-of-the-art methods.

► The Amplitude Decomposition Model for low depth-of-field images is presented.
► A simple method for detecting focused objects from low DOF images is proposed.
► A scheme for segmenting focused objects is presented.