一种基于竞争层神经网络的新型图像轮廓线分组算法

Novel image contour grouping algorithm based on competitive layer neural network

提出了一种基于竞争层神经网络(CLNN)的新型图像轮廓线分组算法。该算法借助于CLNN对生物大脑视觉皮层侧向连接和WTA连接模式的功能性模拟,采用与生物神经元类似的视觉信号处理机制对图像的相应区域进行分组感知。实验证明,所提出的算法具有分组能力强、抗噪声能力突出的优点。

An image contour grouping algorithm based on ComPetitive Layer Neural Network （CLNN） was proposed. The presented algorithm, which is inspired from the lateral interactions and WTA interaction within the biological visual cortex, performs the grouping perception on the image by using the visual signal processing mechanisms similar to the biological neurons. The results of experiments show that the proposed algorithm exhibits stronger grouping capability and promising antinoise capability.