文摘
As nowadays haze and other inclement weather becomes frequency, weak contrast vehicle target recognition has become a problem needed to solve urgently. At present, the existing vehicle recognition algorithms for weak contrast target vehicle are hard to achieve ideal effect. Therefore, in this paper, the Green Associative Mechanism Model based on human vision perceptual theory is proposed. This model has excellent associative function which is improved by introducing the neural interaction mechanism. Then, associative WHAT pathway used for object recognition in human visual cortex is abstracted properly. And on this basis a hierarchical Associative Mechanism Model is established, which is used for weak contrast vehicle recognition. The experiments show that the recognition rates of the model reach 88.6%, which is far above these of the existing recognition algorithms.