为了模拟视觉通路的特征抽象与自学习能力,在视神经信息分层处理机制的基础上提出一种特征感知模型.在改进卷积神经网络框架的基础上,首先依据视网膜中神经元的方向选择性、空间局部性以及神经元间的侧抑制性,在初级视觉特征处理中构建一种视网膜拓扑映射;然后在中级视觉特征处理中引入生物神经的稀疏表达法,构建神经突触激活函数,解决了神经计算中常见的过拟合问题;最后提出模拟腹部通路信息传递的具有计算感知不变性的分层视觉特征感知计算模型.应用不同规模数据集进行测试的结果表明,该模型对大规模的目标识别问题具有较好的识别效果,目标识别的平均准确率可达85%.%To reproduce the ability of feature abstraction and self-learning of visual pathways, a feature percep-tion model is proposed based on the hierarchical processing mechanism of optic nerve. On the basis of the im-proved convolution neural network, the model firstly established a retinal topological mapping in the primary visual feature processing according to the directional selectivity, the spatial locality and the lateral inhibition of neurons in retina; and then, provided a neural synaptic activation function to solve the problem of over-fitting in neural computation, by introducing the sparse representation of the biological nerve in the secondary visual fea-ture processing; finally, the hierarchical visual perception computing model with computing perceptual invariance was proposed by reproducing information transmission of the ventral stream. The test results applying different scale data sets show that the proposed model has sound recognition effect on large scale target recognition prob-lems, and the average recognition accuracy can reach 85%.
展开▼
