A novel biologically inspired ELM-based network for image recognition

摘要: In this paper, a novel biologically inspired network for image recognition has been introduced. The Hierarchical model and X (HMAX) the extreme learning machine (ELM) are combined, to construct five-layer feed-forward network: S1-C1-S2-C2-H. previous four layers, originating from HMAX, provide robust feature representation of specific object, classification stage in H layer is implemented with ELM. HMAX simulates hierarchical processing mechanism primate visual cortex, calculate complex features representation. As biological algorithm generalized SLFNs, ELM learns much faster good generalization performance, performs well applications. Four groups experiments performed on three datasets, results compared state-of-the-art techniques. Experimental show that our proposed performance fast speed.