基于卷积神经网络与视觉显著性的图像融合分析

Image fusion analysis based on convolutional neural network and visual saliency

红外、可见光图像在传统融合当中主要采取手动的形式进行图像的提取,但是无论是方法还是特征都较为单一,并不能够满足当前发展需求。通过深度学习实现图像特征的自动化提取,使传统提取方法得到了丰富与创新,在此过程中要充分利用卷积神经网络,使视觉显著性的图像融合更为全面,在条件随机场的作用下,精确分割图像,以更为显著性的方式对目标进行提取。

In traditional fusion,infrared and visible images are mainly extracted manually,but both methods and features are relatively simple,which cannot meet the current development needs.The automatic lifting of image features is realized through deep learning,which enriches and innovates the traditional extraction methods.In this process,it is necessary to make full use of convolutional neural network to achieve more comprehensive image fusion of visual significance.Under the effect of conditional random field,the image is accurately segmented,and the target is extracted in a more significant way.