Construction algorithm for convolution mask in scale space inspired by mechanism of retina

In view of the complexity of traditional mask constructing methods, the paper analyzes a spatial summation model between photoreceptor and ganglion cells in a retina. Secondly, a novel approach is proposed based on a mechanism of human vision model. Thirdly, a family of scale masks can be constructed only using different sampling rate of standard gauss curve. The paper also demonstrates some practical application of the filter in temporal and frequent domain. Compared with traditional constructing methods of Gauss scale space mask, this approach has the simple and normative characteristics.