基于眼球微动机理,提出了一种结合静态、动态机理的贴近非经典感受野的模型。该模型以感受野轴两侧的两个半椭圆环为非经典感受野抑制区,并在其中设立子区域。通过抑制方向角度模拟微动,结合抑制权值与图像亮度特征,得到最后能量值。研究结果表明,利用该模型提取目标轮廓时,可较好地抑制背景纹理,保留较多轮廓,相比于传统模型有较好的效果。

Based on the mechanism of eyeball fretting, a model combining with the static and dynamic mechanisms is proposed, which is close to the non-classical receptive field model. In this model, two semi-elliptical rings on both sides of the receptive field axis are regarded as the non-classical receptive field suppression areas and the sub-regions are established as well. The fretting is simulated by suppressing the direction angle, and the final energy value is obtained by combining the suppression weight and the luminance features of the image. The experimental results show that using the proposed model to extract the target contours can restrain the background texture well and retain more contours, which is better than that of the traditional model.