Spectral interference in Red Green Blue-Near InfRared (RGB-NIR) images captured by single sensor results in colour distortion and detail information ambiguity of the reconstructed standard Red Green Blue (RBG) and Near InfRared (NIR) images. To resolve this problem, a demosaicing method based on deep learning was proposed. In this method, the grandient dppearance and dispersion problems were solved by introducing long jump connection and dense connection, the network was easier to be trained, and the fitting ability of the network was improved. Firstly, the low-level features such as pixel correlation and channel correlation of the mosaic image were extracted by the shallow feature extraction layer. Secondly, the obtained shallow feature graph was input into successive and multiple residual dense blocks to extract the high-level semantic features aiming at the demosaicing. Thirdly, to make full use of the low-level features and high-level features, the features extracted by multiple residual dense blocks were combined. Finally, the RGB-NIR image was reconstructed by the global long jump connection. Experiments were performed on the deep learning framework Tensorflow using three public data sets, the Common Image and Visual Representation Group (IVRG) dataset, the Outdoor Multi-Spectral Images with Vegetation (OMSIV) dataset, and the Forest dataset. The experimental results show that the proposed method is superior to the RGB-NIR image demosaicing methods based on multi-level adaptive residual interpolation, convolutional neural network and deep residual U-shaped network.

The existing boundary prior based saliency algorithm model has the problem of improper selection of reasonable saliency prior region, which leads to the inaccurate foreground region and influence the final result. Aiming at this problem, a salient target detection algorithm based on contrast optimized manifold ranking was proposed. The image boundary information was utilized to find the background prior. An algorithm for measuring the priori quality was designed by using three indexes, namely, saliency expection, local contrast and global contrast. A priori quality design with weighted addition replaced simple multiplication fusion to make the saliency prior more accurate. When the salient regions were extracted from the a priori, the strategy of selecting the threshold was changed, the foreground region was selected more rationally, and the saliency map was obtained by using the manifold ranking, so that the saliency detection result was more accurate. The experimental results show that the proposed algorithm outperforms the similar algorithms, reduces the noise, which is more suitable for human visual perception, and ahead of the depth learning method in processing time.