Object recognition inspired by biological visual information processing mechanism is one of the research subjects in current computer vision research field, the main idea is to simulate the hierarchical process of visual information in brain visual cortex and build mathematic model to achieve target recognition. However, traditional hierarchical calculation model is usually built based on front feed information transfer, and the passive hard wired way is used between layer and layer. The multi level decomposition of visual information is emphasized, but less involved in the active perception and learning process of visual nervous system. So the convolutional neural networks with sparsely connection thought, self learning mechanism and good network topology structure are chosen as the framework. Based on classical convolutional neural network model, with hierarchical and biomimetic idea, a new enhanced level convolution neural network (CNN) model IH-CNN based on visual nerve is proposed. Experimental results show that target recognition issue in large scale images can be better solved through IH-CNN model and the target recognition accuracy rate is 84%.