Gradient-threshold Edge Detection based on Perceptually Adaptive Threshold Selection

Despite the prevalence of conventional methods for gradient-threshold edge detection, the underlying global threshold is not adaptive to the image content with respect to the human perception. To address the inherent challenges, local thresholds are adaptively selected in this paper taking into account the activity masking characteristic of the human visual system. The selected local thresholds are then utilized for edge labeling in the gradient image. Extensive experimental results have demonstrated the effectiveness of the proposed method for edge detection in perceptual quality.