文中提出了一种基于遗传算法并考虑空域距离关系来设计细胞神经网络模板参数并实现森林红外图像动物边缘检测的算法。仿真结果表明，该算法与其他的森林红外图像边缘检测算法相比，能很好地抑制噪声对边缘检测的干扰，保留边缘信息，同时能较好地检测出红外图像动物的边缘，检测精度高，具有良好的应用前景。

This paper proposes an algorithm based on genetic algorithm and considering the spatial distance factor to design the template parameters of cellular neural networks. Then this algorithm is used to realize the animal edge detection in forest infrared image. The simulation results show that: comparing with other edge detection algorithm in forest infrared image, this algorithm can effectively suppress noise on edge detection, and preserve the edge information. It can detect the infrared image edge of the animal much better. The detection precision is high, which has a good application prospect.