Abstract: Aiming to solve the problems that existing infrared and visible image fusion models which ignored illumination factor in fusion process and used conventional fusion strategy leaded to the fusion results with the loss of detail information and inconspicuous salient information, a deep learning model for infrared and visible image fusion based on illumination weight allocation and attention was proposed. Firstly, an Illumination Weight Allocation Network (IWA-Net) was designed to estimate the illumination distribution and calculated illumination weights. Secondly, a CM-LI-norm fusion strategy was introduced to enhance dependency between pixels and achieved smooth processing of salient features. Finally, a decoding network composed of fully convolutional layers was employed to reconstruct fused images. The results of the fusion experiments on publicly available datasets show that the proposed model outperforms the compared models, with improvements observed in all six selected evaluation metrics. Specifically, the Spatial Frequency (SF) and Mutual Information (MI) metrics have increased by an average of 0.45 and 0.41, respectively, which effectively reduces edge blurring and enhances clarity and contrast of the fused image. The fusion results of this model exhibit superior performance in both subjective and objective aspects compared to other contrastive models.